A Handbook of Environmental Management PDF

A HANDBOOK OF ENVIRONMENTAL
MANAGEMENT
A Handbook of
Environmental Management
Edited by
Jon C. Lovett
Professor of Sustainable Development in a North South
Perspective, University of Twente, The Netherlands and
former Director, Centre for Ecology, Law and Policy,
University of York, UK
David G. Ockwell
Lecturer in Geography and Fellow, Sussex Energy Group and
Tyndall Centre for Climate Change Research, University of
Sussex, UK
Edward Elgar
Cheltenham, UK • Northampton, MA, USA
© Jon C. Lovett and David G. Ockwell 2010
All rights reserved. No part of this publication may be reproduced, stored

in a retrieval system or transmitted in any form or by any means, electronic,
mechanical or photocopying, recording, or otherwise without the prior
permission of the publisher.
Published by
Edward Elgar Publishing Limited
The Lypiatts
15 Lansdown Road
Cheltenham
Glos GL50 2JA
UK
Edward Elgar Publishing, Inc.

William Pratt House
9 Dewey Court
Northampton
Massachusetts 01060
USA
A catalogue record for this book

is available from the British Library
Library of Congress Control Number: 2009940631
ISBN 978 1 84064 207 0 (cased)
Printed and bound by MPG Books Group, UK

03
Contents
List of contributors vii
1 Introduction 1
Jon C. Lovett and David G. Ockwell
2 Global biodiversity conservation priorities: an expanded review 8
Thomas M. Brooks, Russell A. Mittermeier, Gustavo A.B. da
Fonseca, Justin Gerlach, Mike Hoffmann, John F. Lamoreux,
Cristina G. Mittermeier, John D. Pilgrim and Ana S.L. Rodrigues
3 Integrated conservation and development projects: a positive
role for forest conservation in tropical Africa? 30
Neil Burgess, David Thomas, Shakim Mhagama,
Thomas Lehmberg, Jenny Springer and Jonathan Barnard
4 Biodiversity conservation in managed landscapes 75
Tom M. van Rensburg and Greig A. Mill
5 How do institutions affect the management of environmental
resources? 119
Bhim Adhikari
6 Analysing dominant policy perspectives – the role of discourse
analysis 168
David G. Ockwell and Yvonne Rydin
7 Theoretical perspectives on international environmental
regime effectiveness: a case study of the Mediterranean Action
Plan 198
Sofia Frantzi
8 The price of fish and the value of seagrass beds:
socioeconomic aspects of the seagrass fishery on Quirimba
Island, Mozambique 241
Fiona R. Gell
9 The link between ecological and social paradigms and the
sustainability of environmental management: a case study
of semi-arid Tanzania 282
Claire H. Quinn and David G. Ockwell
10 Exploring game theory as a tool for mapping strategic
interactions in common pool resource scenarios 309
Vanessa Pérez-Cirera
v
vi A handbook of environmental management
11 Economic valuation of different forms of land-use in semi-

arid Tanzania 331
Deborah Kirby
12 Economic growth and the environment 359
Dalia El-Demellawy
13 Biodiesel as the potential alternative vehicle fuel: European
policy and global environmental concern 408
Mahesh Poudyal and Jon C. Lovett
Index 431
Contributors
Bhim Adhikari, Institute for Water, Environment and Health, United

Nations University, Canada
Jonathan Barnard, BirdLife International, UK
Thomas M. Brooks, NatureServe, USA; World Agroforestry Centre
(ICRAF), University of the Philippines; and School of Geography and
Environmental Studies, University of Tasmania, Australia
Neil Burgess, Institute of Biology, University of Copenhagen, Denmark
Dalia El-Demellawy, Environment Department, University of York, UK
Gustavo A.B. da Fonseca, Global Environment Facility, USA and
Departamento de Zoologia, Instituto de Ciencias Biologicas, Universidade
Federal de Minas Gerais, Brazil
Sofia Frantzi, Institute for Environmental Studies (IVM), Vrije Universiteit
Amsterdam, The Netherlands
Fiona R. Gell, Wildlife and Conservation Division, Department of
Agriculture, Fisheries and Forestry, Isle of Man Government, British
Isles
Justin Gerlach, Nature Protection Trust of Seychelles, UK
Mike Hoffmann, IUCN Species Survival Commission, c/o United Nations
Environment Programme – World Conservation Monitoring Centre, UK
Deborah Kirby, Kirby Consulting, Nairobi, Kenya
John F. Lamoreux, Randolph, New Hampshire, USA
Thomas Lehmberg, Danish Ornithological Society, Denmark
Jon C. Lovett, University of Twente, The Netherlands
Shakim Mhagama, Wildlife Conservation Society of Tanzania
Greig A. Mill, De Montfort University, UK
Cristina G. Mittermeier, International League of Conservation
Photographers, USA
Russell A. Mittermeier, Conservation International, USA
vii
viii A handbook of environmental management
David G. Ockwell, Department of Geography, University of Sussex, UK

Vanessa Pérez-Cirera, WWF Mexico
John D. Pilgrim, BirdLife International in Indochina, Vietnam and Global
Biodiversity Network, Vietnam
Mahesh Poudyal, University of York, UK
Claire H. Quinn, School of Earth and Environment, University of Leeds,
UK
Ana S.L. Rodrigues, Department of Zoology, University of Cambridge,
UK and Centre d’Ecologie Fonctionnelle et Evolutive, CNRS–UMR
5175, France
Yvonne Rydin, Bartlett School of Planning, University College London,
UK
Jenny Springer, WWF US, USA
David Thomas, BirdLife International, UK
Tom M. van Rensburg, Department of Economics, National University of
Ireland, Galway, Republic of Ireland
1. Introduction
Jon C. Lovett and David G. Ockwell
Environmental management has risen from being the task of technical

natural resource specialists to being the concern of everyone on the planet.
This has led to a rapid expansion in the range of jobs dealing with environ-
mental issues. Not only are ecologists, conservationists, hunters, farmers
and fishers involved; we now also have professionals in social science fields
such as environmental economics, law and politics. Previously a topic that
was dominated by the application and interpretation of technical meas-
ures such as species diversity and population growth rates, environmental
management is now being debated in terms such as property rights and
market trading. Sometimes the technical and social aspects make uneasy
bedfellows: for example, ecologically minded conservationists can find
themselves at loggerheads with human rights lawyers seeking equitable
access to protected areas for indigenous peoples. In this book we aim to
provide overviews and specific examples of case studies and techniques
that are used in environmental management from the local level to
international environmental regimes.
The recognition of a division between technical and social fields of
study is not new. In 1959 the scientist, administrator and novelist C.P.
Snow gave a lecture in Cambridge entitled ‘The Two Cultures and the
Scientific Revolution’. This focused on the idea that the ‘intellectual life of
the whole of western society is increasingly split into two groups’, literary
intellectuals and scientists (Snow, 1998). The ‘Two Cultures’ theme was
taken up again nine years later in another famous paper, ‘The Tragedy of
the Commons’, written by the biologist Garrett Hardin (Hardin, 1968). In
this paper, Hardin addressed the classic academic divide between social
and natural sciences. He described the gulf thus:
An implicit and almost universal assumption of discussions published in pro-

fessional and semi-popular scientific journals is that the problem under discus-
sion has a technical solution. A technical solution may be defined as one that
requires a change only in the techniques of the natural sciences, demanding
little or nothing in the way of change in human values or ideas of morality.
The class of ‘no technical solution problems’ has members . . . They think that
farming the seas or developing new strains of wheat will solve the problem –
technologically. I try to show here that the solution they seek cannot be found.
(Hardin, 1968, p. 1243)
1
2 A handbook of environmental management
The point of Hardin’s paper is that technical solutions alone cannot

solve environmental management problems. The environment does not
exist in isolation from human society or the economic systems that operate
within society. The environment both defines, and is shaped by, the
activities of human beings. In the past, however, there has been a tendency
amongst environmental managers to try to implement technical natural-
science-based solutions to environmental problems without any attempt
to understand the socioeconomic dynamics that underlie the context
within which such technical solutions are applied. This has often resulted
in misaligned management objectives and ultimately management failure.
Leach and Mearns’ (1996) study of the fuel wood shortage in Africa
in the 1980s provides a good example of this (Ockwell and Rydin, 2006).
The fuel wood shortage was perceived by most environmental managers
as the result of a wood supply gap. In other words, demand for fuel wood
exceeded supply and the technical solution was simple – plant more trees.
Unfortunately, this tree planting approach failed to address the problem
of the fuel wood shortage for the many African people whose livelihoods
were affected. It emerged later that the problem was not a simple issue of
a lack of supply, but a far more complex problem related to the nature
of ownership and use of trees as a source of fuel in Africa. Following the
broad-scale failure of the tree planting policy to address the fuel wood
shortage, social scientists working together with natural scientists later
demonstrated that the basic assumptions that define the idea of a supply
gap ignore more subtle issues such as the fact that most fuel wood comes
from clearing wood for agriculture or from lopping branches valued for
fruit and shade. From the perspective of people affected by the fuel wood
shortage there was not one simple problem of a lack of supply, but many
more complex socioeconomic problems associated with command over
trees and their products to meet a wide range of basic needs. This high-
lights the need to attend to a range of socioeconomic issues in environ-
mental management, such as the nature of property rights regimes, local
cultural practices and the subjective, often conflicting, understanding of
different resource users (Ockwell, 2008).
The constructive outcomes of natural and social scientists working
together to solve environmental management problems have led to an
increasing awareness of the need for interdisciplinary approaches to envi-
ronmental management. This requires managers to combine insights from
both the natural and social sciences in order to ensure sustainable out-
comes. In an attempt to help current and future managers to understand
how they might complement their natural science approaches with insights
from the social sciences, this Handbook of Environmental Management
contains a range of case studies that demonstrate the complementary
Introduction 3
application of different social science techniques in combination with

ecological management thinking.
Tom Brooks et al. (Chapter 2) highlight the importance of an awareness
of how conservation funding is spent. Allocation of money essentially rep-
resents a key way in which human beings interact with the environment.
The approach taken to prioritizing which areas benefit from the $6bn
spent annually on conservation has obvious consequences for global bio-
diversity. In their chapter, a shorter version of which was originally pub-
lished in the journal Science, Brooks et al. present a comprehensive review
of the concepts, methods, results, impacts and challenges of approaches to
nine templates of global biodiversity priorities that have been proposed by
biodiversity conservation organizations over the last decade. Their review
is rooted within the theoretical irreplaceability/vulnerability framework of
systematic conservation planning. This chapter makes an important con-
tribution to improving understanding of these prioritization approaches,
which in turn makes it possible to orient more efficient allocations of
geographically flexible conservation funding.
Neil Burgess et al. (Chapter 3) discuss people versus environment and
people and environment policies in the context of wildlife conservation
as a divide between those promoting ‘fortress conservation’ and those
promoting ‘people-focused conservation’. In their chapter they argue that
for environmental managers involved with implementing conservation
projects on the ground in the developing world, these polarized views
often represent impractical extremes. Furthermore, for people living in
rural areas of developing countries, the divide between ‘development’
and ‘conservation’ is also often quite artificial. Burgess et al. highlight a
third approach to environmental management that falls between the two
extremes. Projects that take this middle ground approach are known as
Integrated Conservation and Development Projects (ICDPs). The authors
present a detailed analysis of the successes and failures of ICDPs over the
years and develop some practical ecological, social and economic criteria
by which ICDPs might be assessed. They then provide a practical example
of how to apply such criteria by using them to analyse the successes and
failures of two ICDPs with which they have had personal involvement.
By combining ecological criteria with social and economic criteria, the
authors’ analysis enables them to make a series of practical management
recommendations for making the ICDP model more effective in achieving
conservation at the same time as sustaining and improving the lives of the
people that live in these areas.
Management efforts to maintain biodiversity do not need to be focused
solely on protected areas. Economically productive landscapes contain
many of the world’s species, and many protected areas now regarded
as wilderness were formerly managed for livestock, hunting or exten-

sive agriculture. Tom van Rensburg and Greig Mill (Chapter 4) take a
functional approach to the ecology of biodiversity conservation with an
emphasis on disturbance. With rising human population leading to ever-
increasing demand for food and other natural products, policies that offer
incentives for combining biodiversity conservation with other productive
management objectives will become ever more important in the future.
The move away from protection to production has resulted in new laws
that shift the focus of management from central government control of
natural resources such as forests, towards community involvement in
management with corresponding changes in access and utilization. Bhim
Adhikari (Chapter 5) demonstrates the role that social institutions play
in the management of common pool resources (CPRs – natural resources
that are communally owned and managed). Drawing on the new insti-
tutional economics literature, Adhikari shows how an understanding of
the nature of social institutions is vital if environmental managers are to
be successful in intervening in the management of CPRs. This includes a
need to understand the property rights determining the nature of resource
ownership as well as any unwritten social ‘contracts’ that permit members
of the community to access and use the resource. When CPRs begin to
be degraded, it is often as a result of external pressures that erode the
social institutions that have traditionally governed resource manage-
ment regimes. Management interventions that fail to understand these
traditional institutions and the way in which they have been disrupted are
unlikely to be successful in restoring natural resource use to a sustainable
pattern.
David Ockwell and Yvonne Rydin (Chapter 6) explore the idea of
policy discourses in theoretical and methodological terms. They provide
a practical example of how environmental managers might formally
approach the analysis of the hidden assumptions, values and beliefs
that often underpin dominant framings of environmental problems (for
example, the fuel wood supply gap mentioned above) and expose them
to more critical scrutiny. These dominant framings often prevent more
sustainable, alternative policy solutions from gaining policy influence.
Exposing them to critical scrutiny is one way in which to demonstrate the
policy relevance of alternative knowledge. Ockwell and Rydin focus on
the now well-established field of ‘discourse analysis’. They introduce some
of the core theoretical principles behind different approaches to discourse
analysis before demonstrating the methodological and practical implica-
tions of these different approaches via their application to a case study of
fire management in Cape York, northern Australia. Their chapter pro-
vides a practical example of ‘how to do discourse analysis’. At the same
Introduction 5
time it clearly highlights the insights that environmental managers might

derive by using discourse analysis to better understand the hidden assump-
tions that lie behind different management options.
Differing discourses are not only found in approaches to the manage-
ment of specific resources or areas of land, but also in much higher-level
policy. Sofia Frantzi (Chapter 7) reviews different perspectives on interna-
tional environmental regime effectiveness using the Mediterranean Action
Plan (MAP) as an example. Originally conceived by an ‘epistemic commu-
nity’ of scientists as a means of combating pollution in the Mediterranean
Sea, the MAP can also be regarded as a tool that enables Mediterranean
countries to come together for negotiation, with the technical manage-
ment goal of pollution reduction being subsidiary to the objective of more
general political cooperation. This insight is fundamental to understand-
ing why scientists often become frustrated with policy-makers when the
science findings they are trying to promote take a back seat to consid-
erations of trade and security, which are the main drivers of national
interests.
Staying in the marine environment, but with a local rather than inter-
national focus, Fiona Gell (Chapter 8) provides a detailed demonstration
of how understanding the economics behind natural resource use can lead
to a better understanding of how to ensure that resource use is sustain-
able. Gell looks at the economics of a seagrass fishery in the Quirimba
Archipelago, northern Mozambique. Through an in-depth analysis of the
socioeconomic dynamics of the people who rely on the fishery for their
livelihoods, Gell makes an informed set of management recommendations
for the long-term sustainable management of the fishery.
Claire Quinn and David Ockwell (Chapter 9) highlight an issue
common to many of the chapters in this handbook – that environmental
management for sustainable development needs to protect the environ-
ment at the same time as protecting and developing the livelihoods of
those people who depend on it. This is particularly important for some
of the world’s poorest people whose livelihoods are often most dependent
on natural resources. Using the case study of semi-arid Tanzania, Quinn
and Ockwell highlight the reciprocal relationship that exists between the
environment and society. They then demonstrate how the paradigms that
define environmental managers’ and policy-makers’ conceptions of eco-
logical and social problems are integral to defining the policy discourses
that shape the choice of management solutions. The authors provide a
clear example of how the traditional emphasis on the ecologically centred
‘equilibrium theory’ led to a view of pastoralism as responsible for envi-
ronmental degradation in semi-arid regions of sub-Saharan Africa. In
contrast, the alternative, more recently emerging ecological paradigm
based on ‘non-equilibrium theory’ lends itself to a new perspective that

sees pastoralists and other indigenous populations as being an integral
part of the environment. Most importantly, Quinn and Ockwell show
that these different ecological and social paradigms have fundamental
implications for the policy discourses that are adopted and that define
appropriate management strategies. Equilibrium theory was fundamental
to the policy discourse of people versus environment that has traditionally
defined colonial-influenced approaches to environmental management
in Africa. Non-equilibrium theory, on the other hand, supports a policy
discourse of people and environment, which sees the appropriate manage-
ment response to be inclusive of indigenous people and their knowledge.
Quinn and Ockwell’s central argument is that if the ecological paradigms
that underpin policy discourses fail to recognize the reciprocal link
between the environment and society, the resulting management solutions
can only protect the environment at the expense of the livelihoods of poor
people, thus failing to achieve sustainable development.
The two following chapters by Vanessa Pérez-Cirera (Chapter 10) and
Deborah Kirby (Chapter 11) provide differing methodological perspec-
tives on the semi-arid environment discussed by Quinn and Ockwell.
Pérez-Cirera explores the application of game theory whereas Kirby illus-
trates the use of production function economics. Together, these chapters
provide detailed examples of the potential for different approaches to
environmental decision-making.
The scale of environmental management changes to the macroeconomic
considerations of economic growth and the environment in Chapter 12
by Dalia El-Demellawy. Whilst environmental riches are associated with
either a complete absence of economic activity in pristine wilderness, or
low-level hunter-gatherer economies, it is perhaps wealthier countries that
can afford to have both the technological benefits of development and sus-
tainable environmental policies. Intermediate economies are characterized
by natural resource exploitation and pollution. This observation is formal-
ized in the ‘environmental Kuznet’s curve’, which suggests that there is an
inverted ‘U’-shaped curve of environmental degradation associated with
development. If this is the case, then the macroeconomic environmental
management solution is to enhance economic development to the point
where the whole planet is enjoying environmental sustainability. However,
as explained by El-Demellawy, reality is a bit more complicated.
The final chapter by Mahesh Poudyal and Jon Lovett (Chapter 13) deals
with a controversial issue that has created an environmental management
conundrum. Scientists are agreed that the release of greenhouse gases by
modern economic activity, in particular the burning of fossil fuels, has
resulted in global warming. The environmental effects of global warming
Introduction 7
are predicted to be enormous, changing the whole ecology of the planet. A

technical solution is to replace fossil fuel with renewable resources such as
biofuel derived from agricultural crops. Policy-makers in Europe can see
a wide range of benefits from this move: meeting commitments under the
Kyoto Protocol, improved security of fuel supplies, enhanced European
integration through agricultural subsidies to new European partners and a
strong market for biofuels from developing countries, which can help meet
the Millennium Development Goals. Set against these potential benefits
are environmentally negative changes in land use such as the destruction
of tropical rainforest for biofuel crops and the introduction of large-scale
mono-cultures. The future will reveal if ‘second generation’ biofuels
from wood products are the answer, or if biofuels offer a false dawn for
maintaining our fuel dependence in light of global warming.
Each of the chapters in this handbook provides practical examples
of the ways in which insights from the social sciences can complement
knowledge from the natural sciences to make environmental management
more effective. Sustainable development presents the dual challenge of
maintaining environmental quality whilst improving the livelihoods of
the people who rely on natural resources. In the past, implementation
of environmental management has been hampered by a tendency to rely
on technical solutions without understanding the socioeconomic context
within which these technical solutions were applied. The complementary
application of different social science techniques in combination with
ecology-based management thinking, as demonstrated in this handbook,
provides practical solutions to overcoming this problem. Such an inter-
disciplinary approach to environmental management, working across the
social and natural sciences, is integral to developing effective management
solutions and achieving sustainable development.
References
Hardin, G. (1968), ‘The tragedy of the commons’, Science, 162(3859), 1243–8.
Leach, M. and R. Mearns (1996), The Lie of the Land. Challenging Received Wisdom on the
African Environment, Oxford, UK: James Currey.
Ockwell, D.G. (2008), ‘“Opening up” policy to reflexive appraisal: a role for Q Methodology?
A case study of fire management in Cape York, Australia’, Policy Sciences, 41(4), 263–92.
Ockwell, D. and Y. Rydin (2006), ‘Conflicting discourses of knowledge: understanding the
policy adoption of pro-burning knowledge claims in Cape York Peninsula, Australia’,
Environmental Politics, 15(3), 379–98.
Snow, C.P. (1998), The Two Cultures, Cambridge, UK: Cambridge University Press.
2. Global biodiversity conservation
priorities: an expanded review1
Thomas M. Brooks, Russell A. Mittermeier,
Gustavo A.B. da Fonseca, Justin Gerlach,
Mike Hoffmann, John F. Lamoreux,
Cristina G. Mittermeier, John D. Pilgrim and
Ana S.L. Rodrigues2
Human actions are causing a biodiversity crisis, with species extinction

rates now up to 1000 times higher than the background rate (Pimm et al.,
1995). Moreover, the processes driving extinction are eroding the envi-
ronmental services on which humanity depends (Millennium Ecosystem
Assessment, 2005). People care most about what is close to them, so most
responses to this crisis will be local or national (Hunter and Hutchinson,
1994). Thus, approximately 90 per cent of $6bn annual conservation
funding originates in, and is spent within, economically rich countries
(James et al., 1999). However, this still leaves globally flexible funding of
hundreds of millions of dollars annually from multilateral agencies (for
example, Global Environment Facility), bilateral aid and private sources
including environmentally focused corporations, foundations and indi-
viduals (Balmford and Whitten, 2003). These resources are frequently the
only ones available where conservation is most needed, because biodiver-
sity is unevenly distributed and the most biodiverse places are often the
most threatened and poorest economically (Balmford and Long, 1994;
Balmford et al., 2003; Baillie et al., 2004). Accordingly, geographically
flexible resources exert disproportionate influence on conservation world-
wide, and have a key role in the recently agreed intergovernmental 2010
target to reduce significantly the rate of biodiversity loss (Balmford et al.,
2005).
Since the pioneering work of Myers (1988) on how to best allocate flex-
ible conservation resources, no less than nine major institutional templates
of global biodiversity conservation prioritization have been published,
each with involvement from non-governmental organizations (Figure 2.1).
These strategies have attracted considerable attention (Figure 2.2), result-
ing in much progress as well as controversy. The diversity of approaches
has raised criticisms of duplication of efforts and lack of clarity (Mace et
8
Global biodiversity conservation priorities 9
Crisis ecoregions (Hoekstra et al., 2005)

305 ecoregions with ≥20% habitat conversion and within
High vulnerability
which the percentage conversion is ≥2 times the percentage

protected area coverage
Biodiversity hotspots (Myers et al., 2000)

34 biogeographically similar aggregations of ecoregions
holding ≥0.5% of the world’s plants as endemics, and with
≥70% of primary habitat already lost [Myers 1990, 1991;
Mittermeier et al., 1998, 1999, 2004]
Endemic bird areas (Stattersfield et al., 1998)
218 regions holding ≥2 bird species with global ranges of
<50 000 km2, and with more of these endemic than are
shared with adjacent regions [ICBP 1992; Crosby 1994;
Long et al., 1996] EBA
Centres of plant diversity (WWF and IUCN 1994–97)

234 mainland sites holding >1000 plant species, of which
High irreplaceability
≥10% are endemic either to the site or the region; or islands

containing ≥50 endemic species or ≥10% of flora endemic CPD
Megadiversity countries (Mittermeier et al., 1997)

Countries holding ≥1% of the world’s plants as endemics
[Mittermeier, 1988]
MC
Global 200 ecoregions (Olson and Dinerstein, 1998)
142 terrestrial ecoregions within biomes characterized
by high species richness, endemism, taxonomic uniqueness,
unusual phenomena, or global rarity of major habitat
G200
type [Olson and Dinerstein, 2002]
High-biodiversity wilderness areas (Mittermeier et al., 2003b)
Five biogeographically similar aggregations of
ecoregions with ≥0.5% of the world’s plants as endemics,
and with ≥70% of primary habitat remaining and HBWA
≥5 people per km2 [Mittermeier et al., 1998, 2002]
Low vulnerability
Frontier forests (Bryant et al., 1997)

Forested regions large enough to support viable
populations of all native species, dominated by native
tree species, and with structure and composition driven
by natural events FF
Last of the wild (Sanderson et al., 2002)

10% wildest 1-km2 grid cells in each biome, with wildness
measured with an aggregate index of human density, land
transformation, access and infrastructure LW
Note: For each proposal, we note primary references, definitions paraphrased from the
primary references, maps from the primary references except for Global 200 ecoregions
(Olson and Dinerstein, 2002) and biodiversity hotspots (Mittermeier et al., 2004) for which
more recent maps are available, and secondary references in square brackets. The Global
200 ecoregions also include 53 freshwater and 43 marine ecoregions, not mapped.
Figure 2.1 Global biodiversity conservation prioritizations

800
CE BH EBA
700
CPD MC G200
600
Number of citations
HBWA FF LW
500
400
300
200
100
0
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Year
Note: Codes as in Figure 2.1. Growth measured by number of citations (in parentheses)
of the primary (most cited) reference for each global conservation prioritization recorded
in the Web of Science (accessed 23 December 2009, searching comprehensively using ‘Cited
Reference Search’).
Figure 2.2 Growth of attention to global biodiversity conservation

prioritization
al., 2000). Although attempts have been made to summarize conservation

planning strategies by scale (Redford et al., 2003), no one has considered
them within the framework of conservation planning generally (Margules
and Pressey, 2000). We review the published concepts and methods behind
global biodiversity conservation prioritization, assess the remaining chal-
lenges and highlight how this synthesis can already inform allocation of
globally flexible resources.
Global prioritization in context

A framework of ‘irreplaceability’ relative to ‘vulnerability’ is central to
conservation planning theory (Margules and Pressey, 2000). Conservation
of all components within such a framework – ‘representation’ of
biodiversity – is not antagonistic to prioritization, as some have argued
(Schmidt, 1996). Rather, the latter is a subset of the former; representa-
tion identifies everything that biodiversity conservation aims to preserve,
whereas prioritization identifies what it aims to preserve first (Ginsberg,
1999). Importantly, in the conservation context, prioritization is quite
distinct from, and should not be confused with, triage. Prioritization
A proactive reactive B proactive reactive
EBA, CPD
Irreplaceability
Irreplaceability
BH MC, G200
HBWA
FF CE
LW
Vulnerability Vulnerability
Note: Codes as in Figure 2.1. (A) Purely reactive (prioritizing high vulnerability) and
purely proactive (prioritizing low vulnerability) approaches. (B) Approaches that do not
incorporate vulnerability as a criterion (all prioritize high irreplaceability).
Source: Revised version of figure originally published in Brooks et al. (2006).
Figure 2.3 Global biodiversity conservation priority templates placed

within the conceptual framework of irreplaceability and
vulnerability
provides a means of scheduling responses within such an overall frame-

work (Mittermeier et al., 2003a). Triage, by contrast, has been interpreted
as writing threatened biodiversity off the conservation agenda as beyond
hope (Pimm, 2000) – discounting the high vulnerability components of the
framework.
Conceptually all nine templates of global biodiversity conservation
priority fit within the framework of conservation planning theory (Figure
2.3). Importantly, though, they map onto different portions of the frame-
work – while most of the templates prioritize high irreplaceability, some
prioritize high and others low vulnerability. These differences are key to
understanding how, and why, the nine prioritizations differ, yielding pri-
ority maps that cover from less than one-tenth to more than one-third of
Earth’s land surface.
Measures of irreplaceability
Six of the nine templates of global conservation priority incorporate irre-
placeability – measures of spatial conservation options (Margules and
Pressey 2000; Pressey and Taffs, 2001). The most common measure of irre-
placeability is plant (WWF and IUCN, 1994–97; Mittermeier et al., 1997,
2003b; Myers et al., 2000) or bird (Stattersfield et al., 1998) endemism, often
supported by terrestrial vertebrate endemism overall (Mittermeier et al.,
1997, 2003b; Myers et al., 2000). The logic for this is that the more endem-
ics a region holds, the more biodiversity is lost if that region is lost (even
if anywhere holding even one endemic is irreplaceable in a strict sense). In
addition to numbers of endemic species, other aspects of irreplaceability
have been proposed including taxonomic uniqueness, unusual phenomena
and global rarity of major habitat types (Olson and Dinerstein, 1998), but
these remain difficult to quantify. Despite the fact that species richness
within a given area is sometimes assumed to be important in prioritization
(Prendergast et al., 1993), none of the approaches rely on species richness
alone. This is because species richness is driven by common, widespread
species, thus strategies focused on species richness tend to miss exactly
those biodiversity features most in need of conservation (Orme et al., 2005;
Possingham and Wilson, 2005; Lamoreux et al., 2006). Three approaches
do not incorporate irreplaceability (Bryant et al., 1997; Sanderson et al.,
2002; Hoekstra et al., 2005).
The choice of measures of irreplaceability is to some degree subjective,
in that data limitations currently preclude the measurement of biodiver-
sity wholesale. Further, these same data constraints have meant that, with
the exception of endemic bird areas (Stattersfield et al., 1998), the meas-
ures of irreplaceability used in global conservation prioritization have
necessarily been derived from specialist opinion. Subsequent tests of plant
endemism estimates (Krupnick and Kress, 2003) have proven this expert
opinion to be quite accurate. However, reliance on specialist opinion
means that results cannot be replicated, raising questions concerning the
transparency of the approaches (Humphries, 2000; Mace et al., 2000). It
also prevents formal measurement of irreplaceability, which requires the
identities of individual biodiversity features, such as species names, rather
than just estimates of their magnitude expressed as a number (Balmford,
et al., 2000; Humphries, 2000; Mace et al., 2000; Brummitt and Lughadha,
2003).
Measures of vulnerability
Five of the templates of global conservation priority incorporate
vulnerability – measures of temporal conservation options (Margules and
Pressey, 2000; Pressey and Taffs, 2001). A recent classification of vulner-
ability (Wilson et al., 2005) recognizes four types of measures based on:
environmental and spatial variables; land tenure; threatened species; and
expert opinion. Of these, environmental and spatial variables have been
used most frequently in global conservation prioritization, measured as
proportionate habitat loss (Myers et al., 2000; Sanderson et al., 2002;
Mittermeier et al., 2003b; Hoekstra et al., 2005). Species–area relation-

ships provide justification that habitat loss translates into biodiversity loss
(Pimm et al., 1995; Brooks et al., 2002). However, use of habitat loss as a
measure of vulnerability has several problems, in that it is difficult to assess
using remote sensing for xeric and aquatic systems, does not incorporate
threats such as invasive species and hunting pressure and is retrospective
rather than predictive (Wilson et al., 2005). The frontier forests approach
(Bryant et al., 1997) uses absolute forest cover as a measure, although this
has been criticized as not reflective of vulnerability (Innes and Er, 2002).
Beyond habitat loss, land tenure, measured as protected area coverage,
has also been incorporated into two approaches (Olson and Dinerstein,
1998; Hoekstra et al., 2005). Other possible surrogates not classified by
Wilson et al. (2005) include human population growth and density, which
are widely thought to be relevant (Sisk et al., 1994; Cincotta et al., 2000;
O’Connor et al., 2003; Shi et al., 2005; Veech, 2003; Balmford et al.,
2001), and were integral to two of the systems (Sanderson et al., 2002;
Mittermeier et al., 2003b). None of the global conservation prioritiza-
tion templates used threatened species or expert opinion as measures of
vulnerability. Household dynamics (Liu et al., 2003) and political and
institutional capacity and governance (O’Connor et al., 2003; Smith et
al., 2003) affect biodiversity indirectly, but have not been incorporated to
date. This is true for climate change as well, which is worrying given that
its impact is likely to be severe (Parmesan and Yohe, 2003; Thomas et al.,
2004; McClean et al., 2006). Finally, while costs of conservation generally
increase with threat, no proposals for global biodiversity conservation
priority have yet incorporated costs directly, despite the availability of
techniques to do this at regional scales (Moran et al., 1997; Wilson et al.,
2006). Two of the templates of global conservation prioritization do not
incorporate vulnerability (WWF and IUCN, 1994–97; Mittermeier et al.,
1997) and the remaining two only incorporate it peripherally (Olson and
Dinerstein, 1998; Stattersfield et al., 1998).
Spatial units
The spatial units most commonly used in systematic conservation planning
are equal-area grids. However, data limitations have precluded their use
in the development of actual templates of global biodiversity conservation
priority to date. Instead, all proposals, with the exception of megadiver-
sity countries (Mittermeier et al., 1997), are based on biogeographic units.
Typically, these units are defined a priori by specialist perception of the
distribution of biodiversity. For example, ‘ecoregions’, one of the most
commonly used such classifications, are ‘relatively large units of land
containing a distinct assemblage of natural communities and species’
(Olson et al., 2001, p. 933). Only in the endemic bird areas approach are
biogeographic units defined a posteriori by the distributions of the species
concerned (Stattersfield et al., 1998). Relative to equal-area grids, biogeo-
graphic units bring advantages of ecological relevance, while megadiver-
sity countries (Mittermeier et al., 1997) bring political relevance.
However, reliance on biogeographic spatial units raises several compli-
cations. Various competing bioregional classifications are in use (Jepson
and Whittaker, 2002), with the choice of system having considerable
repercussions for resulting conservation priorities (Pressey and Logan,
1994). Further, when unequally sized units are employed, priority may be
biased towards large areas as a consequence of species–area relationships.
Assessment of global conservation priorities should, therefore, factor out
area, either by taking residuals about a best-fit line to a plot of species
against area (Balmford and Long, 1995; Brooks et al., 2002; Werner and
Buszko, 2005; Lamoreux et al., 2006) or by rescaling numbers of endemics
using a power function directly (Veech, 2000; Brummitt and Lughadha,
2003; Hobohm, 2003; Ovadia, 2003). Nevertheless, use of a priori bio-
regional units for global conservation prioritization will be essential until
data of sufficient resolution become available to enable the use of grids.
Spatial patterns
In Figure 2.4, we map the overlay of the global biodiversity conservation
priority systems into geographic space from the conceptual framework
of Figure 2.3. Figure 2.4A illustrates the large degree of overlap between
templates that prioritize highly vulnerable regions of high irreplace-
ability: tropical islands and mountains (including montane Mesoamerica,
the Andes, the Brazilian Atlantic forest, Madagascar, montane Africa,
the Western Ghats of India, Malaysia, Indonesia, the Philippines and
Hawaii), Mediterranean-type systems (including California, central Chile,
coastal South Africa, south-west Australia and the Mediterranean itself),
and a few temperate forests (the Caucasus, the central Asian mountains,
the Himalaya and south-west China). Highly vulnerable regions of lower
irreplaceability (generally, the rest of the northern temperate regions)
are prioritized by fewer approaches. Figure 2.4B shows a large amount
of overlap between templates for regions of low vulnerability but high
irreplaceability, in particular the three major tropical rainforests of
Amazonia, the Congo and New Guinea. Regions of simultaneously lower
vulnerability and irreplaceability, such as the boreal forests of Canada and
Russia, and the deserts of western USA and central Asia are prioritized
less often.
Two general observations are apparent. First, most land (79 per
cent) is highlighted by at least one of the prioritization systems. Second,
Note: (A) Reactive approaches, corresponding to the right-hand side of Figure 2.3A,
that prioritize regions of high threat and those that do not incorporate vulnerability as a
criterion (Figure 2.3B); the latter are only mapped where they overlap with the former. (B)
Proactive approaches, corresponding to the left-hand side of Figure 2.3A, that prioritize
regions of low threat and those that do not incorporate vulnerability as a criterion (Figure
2.3B); again, the latter are only mapped where they overlap with the former. Shading
denotes number of global biodiversity conservation prioritization templates, in both cases.
Figure 2.4 Mapping the overlay of approaches prioritizing reactive and

proactive conservation
despite this, a noticeable pattern emerges from the overlay of different

approaches. There is significant overlap among templates that prioritize
irreplaceable regions (WWF and IUCN, 1994–97; Mittermeier et al., 1997,
2003b; Olson and Dinerstein, 1998; Stattersfield et al., 1998; Myers et al.,
2000), among those that prioritize highly vulnerable regions (Myers et al.,
2000; Hoekstra et al., 2005), and among those that prioritize regions of
low vulnerability (Bryant et al., 1997; Sanderson et al., 2002; Mittermeier
et al., 2003b), but not between approaches across each of these three
general classes (Table 2.1). This provides useful cross-verification of
priority regions (Fonseca et al., 2000).
These patterns of overlap reflect two approaches to how vulnerability is
incorporated into conservation in the broadest sense: reactive (prioritizing
areas of high threat and high irreplaceability) and proactive (prioritizing
areas of low threat but high irreplaceability). The former are considered
the most urgent priorities in conservation planning theory (Margules and
Pressey, 2000; Pressey and Taffs, 2001) because unless immediate con-
servation action is taken within them, unique biodiversity will soon be
lost. The latter are often de facto priorities, because the opportunities for
conservation in these are considerable (Norris and Harper, 2003; Cardillo
et al., 2006). Biodiversity conservation clearly needs both approaches, but
the implementation of each may correspond to different methods. On the
one hand, large-scale conservation initiatives may be possible in wilder-
ness areas, such as the establishment of enormous protected areas, like
the 3 800 000 ha Tumucumaque National Park, created in the Brazilian
state of Amapá in 2003. On the other hand, finely tuned conservation will
be essential in regions of simultaneously high irreplaceability and threat,
where losing even tiny patches of remnant habitat, like the sites identi-
fied by the Alliance for Zero Extinction (Ricketts et al., 2005), would be
tragic.
Impact of global prioritization

The appropriate measure of impact is the success of prioritization in
achieving its main goal – influencing globally flexible donors to invest in
regions where these funds can contribute most to conservation. Precise
data are unavailable for all of the approaches (Halpern et al., 2006), but
hotspots alone have mobilized at least $750m of funding for conservation
in these regions (Myers, 2003). More specifically, conservation funding
mechanisms have been established for several of the approaches, such
as the $100m, ten-year Global Conservation Fund focused on high-
biodiversity wilderness areas and hotspots, and the $125m, five-year
Critical Ecosystem Partnership Fund, aimed exclusively at hotspots
(Dalton, 2000). The Global Environment Facility, the largest financial
mechanism addressing biodiversity conservation, is currently exploring a

resource allocation framework that builds on existing templates. Both civil
society and government organizations often utilize the recognition given
to regions as global conservation priorities as justification when applying
for geographically flexible funding. In addition, the global prioritization
systems must have had sizeable effects in the cancellation, relocation or
mitigation of environmentally harmful activities, even in the absence of
specific legislation (Kunich, 2001). Unfortunately, resources still fall an
order of magnitude short of required conservation funding (James et al.,
1999). Nevertheless, the dollar amounts are impressive, and represent
dramatic increases in conservation investment in these regions.
Challenges facing global prioritization

Limitations of data have thus far generally restricted global conservation
prioritization to specialist estimates of irreplaceability, to habitat loss as a
measure of vulnerability, and to coarse geographic units defined a priori.
Over the last five years, spatial datasets have been compiled with the
potential to reduce these constraints (Baillie et al., 2004), particularly for
mammals (Ceballos et al., 2005), birds (Orme et al., 2005), and amphibians
(Stuart et al., 2004). When these maps are combined with assessment of
conservation status, they enable the development of threat metrics based
on threatened species directly (Sisk et al., 1994; Rodrigues et al., 2004a). So
far, the main advances to global prioritization enabled by these new data
are validation tests of existing templates (Fonseca et al., 2000; Burgess et
al., 2002). Encouragingly, global gap analysis of priorities for the repre-
sentation of terrestrial vertebrate species in protected areas (Rodrigues et
al., 2004a, b) and initial regional assessment of plants (Küper et al., 2004)
yield results similar to existing approaches (Figure 2.5).
A few have argued that global conservation priorities should be driven
solely by those vertebrates known and loved by society (Jepson and
Canney, 2001). However, invertebrates represent the bulk of eukaryotic
diversity on Earth with over a million known species (Baillie et al., 2004)
and many more yet to be described (Novotny et al., 2002). The conser-
vation status of only ~3500 arthropods has been assessed (Baillie et al.,
2004), and so even setting aside microbes as near-irrelevant to conserva-
tion (Nee, 2004), global conservation priority is still far from being able to
incorporate megadiverse invertebrate taxa (Mace et al., 2000; Brummitt
and Lughadha, 2003). While regional data can show little overlap between
priority areas for arthropods and those for plant and terrestrial vertebrate
taxa (Dobson et al., 1997), there are strong correlations between phytopha-
gous insects and plant species richness (Kelly and Southwood, 1999), and
preliminary global data for groups like tiger beetles and termites suggest
Table 2.1 Spatial overlap between the nine proposed global biodiversity conservation prioritizations
Crisis ecoregions
Biodiversity
hotspots
Endemic bird
areas
Centres of plant
diversity
Megadiversity
countries
Global 200
ecoregions
High-biodiversity
wilderness areas
Frontier forests
Last of the wild
% land area 30 16 10 9 35 37 8 9 24
§ §
Crisis ecoregions 33* 14 10 44 36 2 1 4§
18
High
ability
vulner-
Biodiversity hotspots 61* 33* 21* 46* 78* 0§ 5 6§
Endemic bird areas 43* 50* 24* 68* 70* 7 11 11§
Centres of plant diversity 34 40* 28* 48* 66* 18* 14 21
Megadiversity countries 38 21 19* 12 53* 18* 11 24
Global 200 ecoregions 28 33* 19* 15* 49* 16* 16* 28
High irreplaceability
High-biodiversity wilderness areas 6§ 0§ 15 19* 79* 72* 41* 53*
§ §
Frontier forests 4 8 11 13 39 64* 35* 73*
Low
§ §
Last of the wild 5 4 4 7 34 43 17* 28*
vulnerability
Note: Cells show the percentage overlap of the approach in a given row by the approach in the corresponding column. The three blocks of cells
outlined in bold correspond to overlaps between approaches within three categories, indicated in the bars on the left: reactive approaches, those
prioritizing high irreplaceability and proactive approaches. Significance was tested against overlap expected based on the percentage of the land
area covered by each approach, using a chi-squared test (P = 0.05; * significantly greater overlap than expected; § significantly less overlap than
expected). There are many significant overlaps between approaches within each of the three categories: those prioritizing highly vulnerable regions
19
(2/2 overlaps significantly more than expected), those prioritizing irreplaceable regions (24/30), and those prioritizing regions of low vulnerability
(6/6). The opposite is true when comparing approaches prioritizing vulnerable regions with those prioritizing regions of low vulnerability (11/12
overlaps significantly less than expected). The approaches that incorporate irreplaceability almost all show significantly more overlap than
expected with and by at least four other systems. In contrast, those that do not incorporate irreplaceability only overlap significantly more than
expected with or by three other approaches at the most.
Note: Global conservation prioritization templates have been based almost exclusively
on bioregional classification and specialist opinion, rather than primary biodiversity data.
Such primary datasets have recently started to become available under the umbrella of the
IUCN Species Survival Commission (Baillie et al., 2004), and they allow progressive testing
and refinement of templates. (A) Global gap analysis of coverage of 11 633 mammal, bird,
turtle and amphibian species (~40% of terrestrial vertebrates) in protected areas (Rodrigues
et al., 2004b). It shows unprotected half-degree grid cells characterized simultaneously by
irreplaceability values of at least 0.9 on a scale of 0–1, and of the top 5% of values of an
extinction risk indicator based on the presence of globally threatened species (Rodrigues
et al., 2004a). (B) Priorities for the conservation of 6269 African plant species (~2% of
vascular plants) across a 1-degree grid (Küper et al., 2004). These are the 125 grid cells with
the highest product of range-size rarity (a surrogate for irreplaceability) of plant species
distributions and mean human footprint (Sanderson et al., 2002). Comparison of these two
maps, and between them and Figure 2.4, reveals a striking similarity among conservation
priorities identified using independent datasets. The difference in taxonomic and geographic
coverage between A and B also highlights the challenge facing the botanical community
to compile comprehensive primary data on plant conservation in order to inform global
conservation prioritization (Callmander et al., 2005). Rectifying this is part of the Global
Strategy for Plant Conservation of the Convention on Biological Diversity.
Figure 2.5 Incorporating primary biodiversity data in global conservation

priority-setting
much higher levels of congruence (Mittermeier et al., 2004). Similarly,

pioneering techniques to model wholesale irreplaceability by combining
point data for megadiverse taxa with environmental datasets produce
results commensurate with existing conservation priorities (Ferrier et al.,
2004). These findings, while encouraging, in no way preclude the need to
use primary invertebrate data in global conservation prioritization as they
become available.
Aquatic systems feature poorly in existing conservation templates,

although there is evidence that marine biodiversity may be at least as
threatened as (Dulvy et al., 2003) and freshwater biodiversity even more
threatened (McAllister et al., 1997) than that on land (Baillie et al.,
2004). Only one conservation prioritization explicitly incorporates aquatic
systems (Olson and Dinerstein, 1998). The irreplaceability dimension has
been particularly overlooked in the seas, with the traditional emphasis of
marine conservation being on species richness (Briggs, 2002) despite little
correspondence between marine species richness and endemism (Hughes
et al., 2002; Price, 2002). Nevertheless, the most comprehensive study yet,
albeit restricted to tropical coral reef ecosystems, identified ten priority
regions based on endemism and threat (Roberts et al., 2002). Eight of
these regions lie adjacent to priority regions highlighted in Figure 2.4,
raising the possibility of correspondence between marine and terrestrial
priorities despite the expectation that surrogacy of conservation priorities
will be low between different environments (Reid, 1998). Efforts to iden-
tify freshwater priorities lag further behind, although initial studies reveal
a highly uneven distribution of freshwater fish endemism at regional
(Darwall et al., 2005) and global (Mittermeier et al., 2004) scales.
Most measurement of irreplaceability is species-based, raising the
concern that phylogenetic diversity may slip through the net of global
conservation priorities (Mace et al., 2000; Jepson and Canney, 2001;
Brummitt and Lughadha, 2003; Kareiva and Marvier, 2003). However,
analyses for mammals (Sechrest et al., 2002) and birds (Brooks et al., 2005)
find that priority regions represent higher taxa and phylogenetic diversity
better than would be predicted by the degree to which they represent
species. Islands such as Madagascar and the Caribbean hold especially
high concentrations of endemic genera and families (Mittermeier et al.,
2004). A heterodox perspective argues that the terminal tips of phyloge-
netic trees should be higher priorities than deep lineages (Erwin, 1991). In
any case, the balance of work implies that even if phylogenetic diversity
is not explicitly targeted for conservation, global prioritization based on
species provides a solid surrogate for evolutionary history.
That global conservation priority regions capture phylogenetic history
does not necessarily mean they represent evolutionary process (Mace
et al., 2000; Myers and Knoll, 2001; Smith et al., 2001). For example,
transition zones or ‘biogeographic crossroads’, frequently overlooked by
conservation prioritization, could be of particular importance in driving
speciation (Smith et al., 1997; Spector, 2002). On the other hand, the scale
of global conservation priorities is often coarse enough to capture the
transitions necessary in facilitating evolutionary processes, such as resili-
ence to climate change (Midgley et al., 2001). There is also evidence that
areas of greatest importance in generating biodiversity are those of long-

term climatic stability, especially where they occur in tropical mountains
(Fjeldså and Lovett, 1997), which are incorporated in most approaches to
global conservation prioritization. A related hypothesis states that regions
with particularly high percentage endemism – tropical mountains and
islands – necessarily capture evolutionary process (Fa et al., 2004). The
development of metrics for the maintenance of evolutionary process is in
its infancy, and represents an emerging research front (Araújo, 2002).
A final dimension that will prove important to assess in the context of
global conservation prioritization concerns ecosystem services (Jepson and
Canney, 2001; Kareiva and Marvier, 2003; Odling-Smee, 2005). Although
the processes threatening biodiversity and ecosystem services are likely
similar, the relationship between biodiversity per se and ecosystem ser-
vices remains unresolved (Loreau et al., 2001). Thus, while it is important
to establish distinct goals for these conservation objectives (Sarkar, 1999),
identification of synergies between them is strategically vital. This research
avenue has barely been explored and questions of how global biodiversity
conservation priorities overlap with priority regions for carbon sequestra-
tion, climate stabilization, maintenance of water quality, minimization of
outbreaks of pests and diseases, and fisheries, for example, remain unan-
swered. However, the correspondence between conservation priorities and
human populations (Cincotta et al., 2000; Balmford et al., 2001; Baillie
et al., 2004) and poverty (Balmford et al., 2003; Baillie et al., 2004) is an
indication that the conservation of areas of high biodiversity priority will
deliver high local ecosystem service benefits.
From global to local priorities

The establishment of global conservation priorities has been extremely
influential in directing resources towards broad regions. However, a
number of authors have pointed out that global conservation prioritiza-
tion has had little success in informing actual conservation implementa-
tion (Dinerstein and Wikramanayake, 1993; Mace et al., 2000; Jepson,
2001; Brummitt and Lughadha, 2003). Separate processes are necessary
to identify actual conservation targets and priorities at much finer scales
(Supriatna, 2001), because even within a region as uniformly important
as (say) Madagascar, biodiversity and threats are not evenly distributed.
Bottom-up processes of identification of priorities are therefore essential
to ensure the implementation of area-based conservation (Whittaker et
al., 2005).
Indeed, numerous efforts are underway to identify targets for conser-
vation implementation. Many focus on the site scale, drawing on two
decades of work across nearly 170 countries in the designation of
important bird areas (BirdLife International, 2004). There is an obvious

need to expand such work to incorporate other taxa (Eken et al., 2004),
and to prioritize the most threatened and irreplaceable sites (Ricketts et
al., 2005). Such initiatives have recently gained strong political support
under the Convention on Biological Diversity, through the development
of the Global Strategy for Plant Conservation and the Programme of
Work on Protected Areas. Both mechanisms call for the identification,
recognition and safeguarding of sites of biodiversity conservation signifi-
cance. Meanwhile, considerable attention is also targeted at the scale of
landscapes and seascapes, to ensure not just the representation of bio-
diversity but also of the connectivity, spatial structure and processes that
allow its persistence (Cowling et al., 2003).
Global conservation planning is key for strategic allocation of flexible
resources. Despite divergence in methods between the different schemes,
an overall picture is emerging in which a few regions, particularly in
the tropics and in Mediterranean-type environments, are consistently
emphasized as priorities for biodiversity conservation. It is crucial that the
global donor community channels sufficient resources to these regions, at
the very minimum. This focus will continue to improve if the rigour and
breadth of biodiversity and threat data continue to be consolidated, espe-
cially important given the increased accountability demanded from global
donors. However, it is through the conservation of actual sites that bio-
diversity will ultimately be preserved, or lost, and thus drawing the lessons
of global conservation prioritization down to a much finer scale is now the
primary concern for conservation planning.
Notes
1. This chapter is an expanded version of Brooks et al. (2006), ‘Global biodiversity conser-
vation priorities’, Science, 313(5783), 58–61, originally published by AAAS.
2. We thank G. Fabregas, D. Knox, T. Lacher, P. Langhammer, N. Myers, A. Sugden and
W. Turner for help with the manuscript, anonymous peer reviewers for comments, the
Gordon and Betty Moore Foundation for funding, D. Ockwell and J. Lovett for the invi-
tation for this contribution and for editorial help, and AAAS for permission to reprint
this expanded version of our original review paper.
References
Araújo, M.B. (2002), ‘Biodiversity hotspots and zones of ecological transition’, Conservation
Biology, 16(6), 1662–3.
Baillie, J.E.M., L.A. Bennun, T.M. Brooks, S.H.M. Butchart, J.S. Chanson, Z. Cokeliss, C.
Hilton-Taylor, M. Hoffmann, G.M. Mace, S.A. Mainka, C.M. Pollock, A.S.L. Rodrigues,
A.J. Stattersfield and S.N. Stuart (2004), A Global Species Assessment, Gland, Switzerland:
IUCN – International Union for Conservation of Nature.
Balmford, A. and A. Long (1994), ‘Avian endemism and forest loss’, Nature, 372(6507), 623–4.
Balmford, A. and A. Long (1995), ‘Across-country analyses of biodiversity congruence and
current conservation effort in the tropics’, Conservation Biology, 9(16), 1539–47.
Balmford, A. and T. Whitten (2003), ‘Who should pay for tropical conservation, and how
could the costs be met?’, Oryx, 37(2), 238–50.
Balmford, A., K.J. Gaston, A.S.L. Rodrigues and A. James (2000), ‘Integrating costs of con-
servation into international priority setting’, Conservation Biology, 14(3), 597–605.
Balmford, A., J. Moore, T. Brooks, N. Burgess, L.A. Hansen, P. Williams and C. Rahbek
(2001), ‘Conservation conflicts across Africa’, Science, 291(5513), 2616–19.
Balmford, A., G.J. Gaston, S. Blyth, A. James and V. Kapos (2003), ‘Global variation in
conservation costs, conservation benefits, and unmet conservation needs’, Proceedings of
the National Academy of Sciences of the U.S.A., 100(3), 1046–50.
Balmford, A., L. Bennun, B. ten Brink, D. Cooper, I. Côté, P. Crane, A. Dobson, N.
Dudley, I. Dutton, R.E. Green, R.D. Gregory, J. Harrison, E.T. Kennedy, C. Kremen,
N. Leader-Williams, T.E. Lovejoy, G. Mace, R. May, P. Mayaux, P. Morling, J. Phillips,
K. Redford, T.H. Ricketts, J.P. Rodriguez, M. Sanjayan, P.J. Schei, A.S. van Jaarsveld
and B.A. Walther (2005), ‘The Convention on Biological Diversity’s 2010 target’, Science,
307(5707), 212–13.
BirdLife International (2004), State of the World’s Birds 2004, Cambridge, UK: BirdLife
International.
Briggs, J.C. (2002), ‘Coral reef biodiversity and conservation’, Science, 296(5570), 1027.
Brooks, T.M., R.A. Mittermeier, C.G. Mittermeier, G.A.B. da Fonseca, A.B. Rylands,
W.R. Konstant, P. Flick, J. Pilgrim, S. Oldfield, G. Magin and C. Hilton-Taylor (2002),
‘Habitat loss and extinction in the hotspots of biodiversity’, Conservation Biology, 16(4),
909–23.
Brooks, T.M., J.D. Pilgrim, A.S.L. Rodrigues and G.A.B. da Fonseca (2005), ‘Conservation
status and geographic distribution of avian evolutionary history’, in A. Purvis, J.L.
Gittleman and T.M. Brooks (eds), Phylogeny and Conservation, Cambridge, UK:
Cambridge University Press, pp. 267–94.
Brooks, T.M., R.A. Mittermeier, G.A.B. da Fonseca, J. Gerlach, M. Hoffmann, J.F.
Lamoreux, C.G. Mittermeier, J.D. Pilgrim and A.S.L. Rodrigues (2006), ‘Global biodi-
versity conservation priorities’, Science, 313(5783), 58–61.
Brummitt, N. and E.N. Lughadha (2003), ‘Biodiversity: where’s hot and where’s not?’,
Conservation Biology, 17(5), 1442–8.
Bryant, D., D. Nielsen and L. Tangley (1997), Last Frontier Forests: Ecosystems and
Economies on the Edge, Washington DC: World Resources Institute.
Burgess, N.D., C. Rahbek, F.W. Larsen, P. Williams and A. Balmford (2002), ‘How much
of the vertebrate diversity of sub-Saharan Africa is catered for by recent conservation
proposals?’, Biological Conservation, 107(33), 327–39.
Callmander, M.W., G.E. Schatz and P.P. Lowry II (2005), ‘IUCN Red List assessment and
the Global Strategy for Plant Conservation: taxonomists must act now’, Taxon, 54(4),
1047–50.
Cardillo, M., G.M. Mace, J.L. Gittleman and A. Purvis (2006), ‘Latent extinction risk and
the future battlegrounds of mammal conservation’, Proceedings of the National Academy
of Sciences of the U.S.A., 103(11), 4157–61.
Ceballos, G., P.R. Ehrlich, J. Soberón, I. Salazar and J.P. Fay (2005), ‘Global mammal con-
servation: what must we manage?’, Science, 309(5734), 603–7.
Cincotta, R.P., J. Wisnewski and R. Engelman (2000), ‘Human population in the biodiver-
sity hotspots’, Nature, 404(6781), 990–92.
Cowling, R.M., R.L. Pressey, M. Rouget and A.T. Lombard (2003), ‘A conservation plan
for a global biodiversity hotspot – the Cape Floristic Region, South Africa’, Biological
Conservation, 112(1–2), 191–216.
Crosby, M.J. (1994), ‘Mapping the distributions of restricted-range birds to identify global
conservation priorities’, in R.I. Miller (ed.), Mapping the Diversity of Nature, London:
Chapman and Hall, pp. 145–54.
Dalton, R. (2000), ‘Biodiversity cash aimed at hotspots’, Nature, 406(6798), 818.
Darwall, W., K. Smith, T. Lowe and J.-C. Vié (2005), The Status and Distribution of
Freshwater Biodiversity in Eastern Africa, Gland, Switzerland: IUCN.
Dinerstein, E. and E.D. Wikramanayake (1993), ‘Beyond “hotspots”: how to prioritize

investments to conserve biodiversity in the Indo-Pacific region’, Conservation Biology,
7(1), 53–65.
Dobson, A.P., J.P. Rodriguez, W.M. Roberts and D.S. Wilcove (1997), ‘Geographic distri-
bution of endangered species in the United Space’, Science, 275(5299), 550–53.
Dulvy, N.K., Y. Sadovy and J.D. Reynolds (2003), ‘Extinction vulnerability in marine popu-
lations’, Fish and Fisheries, 4(1), 25–64.
Eken, G., L. Bennun, T.M. Brooks, W. Darwall, M. Foster, D. Knox, P. Langhammer,
P. Matiku, E. Radford, P. Salaman, W. Sechrest, M.L. Smith, S. Spector and J. Tordoff
(2004), ‘Key biodiversity areas as site conservation targets’, BioScience, 54(12), 1110–18.
Erwin, T.L. (1991), ‘An evolutionary basis for conservation strategies’, Science, 253(5021),
750–52.
Fa, J.E., R.W. Burn, M.E. Stanley-Price and F.M. Underwood (2004), ‘Identifying impor-
tant endemic areas using ecoregions: birds and mammals in the Indo-Pacific’, Oryx, 38(1),
91–101.
Ferrier, S., G.V.N. Powell, K.S. Richardson, G. Manion, J.M. Overton, T.F. Allnutt, S.E.
Cameron, K. Mantle, N.D. Burgess, D.P. Faith, J.F. Lamoreux, G. Kier, R.J. Hijmans,
V.A. Funk, G.A. Cassis, B.L. Fisher, P. Flemons, D. Lees, J.C. Lovett and R.S.A.R. Van
Rompaey (2004), ‘Mapping more of terrestrial biodiversity for global conservation assess-
ment’, BioScience, 54(12), 1101–9.
Fjeldså, J. and J.C. Lovett (1997), ‘Geographical patterns of old and young species in African
forest biota: the significance of specific montane areas as evolutionary centres’, Biodiversity
and Conservation, 6(3), 325–46.
Fonseca, G.A.B. da, A. Balmford, C. Bibby, L. Boitani, F. Corsi, T. Brooks, C. Gascon,
S. Olivieri, R. Mittermeier, N. Burgess, E. Dinerstein, D. Olson, L. Hannah, J. Lovett,
D. Moyer, C. Rahbek, S. Stuart and P. Williams (2000), ‘Following Africa’s lead in setting
priorities’, Nature, 405(6785), 393–4.
Ginsberg, J. (1999), ‘Global conservation priorities’, Conservation Biology, 13(1), 5.
Halpern, B.S., C.R. Pyke, H.E. Fox, J.C. Haney, M.A. Schlaepfer and P. Zaradic (2006),
‘Gaps and mismatches between global conservation priorities and spending’, Conservation
Biology, 20(1), 56–64.
Hobohm, C. (2003), ‘Characterization and ranking of biodiversity hotspots: centers of
species richness and endemism’, Biodiversity and Conservation, 12(2), 279–87.
Hoekstra, J.M., T.M. Boucher, T.H. Ricketts and C. Roberts (2005), ‘Confronting a biome
crisis: global disparities of habitat loss and protection’, Ecology Letters, 8(1), 23–9.
Hughes, T.P., D.R. Bellwood and S.R. Connolly (2002), ‘Biodiversity hotspots, centres of
endemicity, and the conservation of coral reefs’, Ecology Letters, 5(6), 775–84.
Humphries, C.J. (2000), ‘Hotspots: going off the boil?’, Diversity and Distributions, 7(1–2), 103–4.
Hunter, M.L., Jr. and A. Hutchinson (1994), ‘The virtues and shortcomings of parochialism:
conserving species that are locally rare, but globally common’, Conservation Biology, 8(4),
1163–5.
ICBP (1992), Putting Biodiversity on the Map, Cambridge, UK: International Council for
Bird Protection.
Innes, J.L. and K.B.H. Er (2002), ‘The questionable utility of the frontier forest concept’,
BioScience, 52(12), 1095–1109.
James, A., K. Gaston and A. Balmford (1999), ‘Balancing the earth’s accounts’, Nature,
401(6751), 323–4.
Jepson, P. (2001), ‘Global biodiversity plan needs to convince local policy-makers’, Nature,
409(6816), 12.
Jepson, P. and S. Canney (2001), ‘Biodiversity hotspots: hot for what?’, Global Ecology and
Biogeography, 10(13), 225–7.
Jepson, P. and R.J. Whittaker (2002), ‘Ecoregions in context: a critique with special reference
to Indonesia’, Conservation Biology, 16(1), 42–57.
Kareiva, P. and M. Marvier (2003), ‘Conserving biodiversity coldspots’, American Scientist,
91(14), 344–51.
Kelly, C.K. and T.R.E. Southwood (1999), ‘Species richness and resource availability: a phy-
logenetic analysis of insects associated with trees’, Proceedings of the National Academy of
Sciences of the U.S.A., 96(14), 8013–16.
Krupnick, G.A. and W.J. Kress (2003), ‘Hotspots and ecoregions: a test of conservation
priorities using taxonomic data’, Biodiversity and Conservation, 12(11), 2237–53.
Kunich, J.C. (2001), ‘Preserving the womb of the unknown species with hotspots legislation’,
Hastings Law Journal, 52(16), 1149–1253.
Küper, W., J.H. Sommer, J.C. Lovett, J. Mutke, H.P. Linder, H.J. Beentje, R.S.A.R. Van
Rompaey, C. Chatelain, M. Sosef and W. Barthlott (2004), ‘Africa’s hotspots of biodiver-
sity redefined’, Annals of the Missouri Botanical Garden, 91(4), 525–35.
Lamoreux, J.F., J.C. Morrison, T.H. Ricketts, D.M. Olson, E. Dinerstein, M.W. McKnight
and H.H. Shugart (2006), ‘Global tests of biodiversity concordance and the importance of
endemism’, Nature, 440(7082), 212–14.
Liu, J., D.C. Daily, P.R. Ehrlich and G.W. Luck (2003), ‘Effects of household dynamics on
resource consumption and biodiversity’, Nature, 421(6922), 530–33.
Long, A.J., M.J. Crosby, A.J. Stattersfield and D.C. Wege (1996), ‘Towards a global map
of biodiversity: patterns in the distribution of restricted-range birds’, Global Ecology and
Biogeography Letters, 5(4–5), 281–304.
Loreau, M., S. Naeem, P. Inchausti, J. Bengtsson, J.P. Grime, A. Hector, D.U. Hooper, M.A.
Huston, D. Raffaelli, B. Schmid, D. Tilman and D.A. Wardle (2001), ‘Biodiversity and eco-
system functioning: current knowledge and future challenges’, Science, 294(5543), 804–8.
Mace, G.M., A. Balmford, L. Boitani, G. Cowlishaw, A.P. Dobson, D.P. Faith, K.J.
Gaston, C.J. Humphries, R.I. Vane-Wright, P.H. Williams, J.H. Lawton, C.R. Margules,
R.M. May, A.O. Nicholls, H.P. Possingham, C. Rahbek and A.S. van Jaarsveld (2000),
‘It’s time to work together and stop duplicating conservation efforts. . .’, Nature,
405(6785), 393.
Margules, C.R. and R.L. Pressey (2000), ‘Systematic conservation planning’, Nature,
405(6783), 243–53.
McAllister, D.E., A.L. Hamilton and B. Harvey (1997), ‘Global freshwater biodiversity:
striving for the integrity of freshwater ecosystems’, Sea Wind, 11(2), 1–106.
McClean, C.J., N. Doswald, W. Küper, J.H. Sommer, P. Barnard and J.C. Lovett (2006),
‘Potential impacts of climate change on sub-Saharan Africa plant priority area selection’,
Diversity and Distributions, 12(16), 645–55.
Midgley, G.F., L. Hannah, R. Roberts, D.J. MacDonald and J. Allsopp (2001), ‘Have
Pleistocene climatic cycles influenced species richness patterns in the Greater Cape
Mediterranean Region?’, Journal of Mediterranean Ecology, 2(2), 137–44.
Millennium Ecosystem Assessment (2005), Ecosystems and Human Well-being: Biodiversity
Synthesis, Washington DC: World Resources Institute.
Mittermeier, R.A. (1988), ‘Primate diversity and the tropical forest’, in E.O. Wilson (ed.),
BioDiversity, Washington DC: National Academy Press, pp. 145–54.
Mittermeier, R.A., P. Robles Gil and C.G. Mittermeier (1997), Megadiversity, Mexico:
CEMEX.
Mittermeier, R.A., N. Myers, J.B. Thomsen, G.A.B. da Fonseca and S. Olivieri (1998),
‘Biodiversity hotspots and major tropical wilderness areas: approaches to setting conserva-
tion priorities’, Conservation Biology, 12(3), 516–20.
Mittermeier, R.A., N. Myers, P. Robles Gil and C.G. Mittermeier (1999), Hotspots, Mexico:
CEMEX.
Mittermeier, R.A., C.G. Mittermeier, P. Robles Gil, J.D. Pilgrim, W.R. Konstant, G.A.B.
da Fonseca and T.M. Brooks (2002), Wilderness: Earth’s Last Wild Places, Mexico:
CEMEX.
Mittermeier, R.A., G.A.B. da Fonseca, T. Brooks, J. Pilgrim and A. Rodrigues (2003a),
‘Hotspots and coldspots’, American Scientist, 91(5), 384.
Mittermeier, R.A., C.G. Mittermeier, T.M. Brooks, J.D. Pilgrim, W.R. Konstant, G.A.B. da
Fonseca and C. Kormos (2003b), ‘Wilderness and biodiversity conservation’, Proceedings
of the National Academy of Sciences of the U.S.A., 100(18), 10309–10313.
Mittermeier, R.A., P. Robles Gil, M. Hoffmann, J. Pilgrim, T. Brooks, C.G. Mittermeier,

J. Lamoreux and G.A.B. da Fonseca (2004), Hotspots: Revisited, Mexico: CEMEX.
Moran, D., D. Pearce and A. Wendelaar (1997), ‘Investing in biodiversity: an economic per-
spective on global priority setting’, Biodiversity and Conservation, 6(9), 1219–43.
Myers, N. (1988), ‘Threatened biotas: “hot spots” in tropical forests’, The Environmentalist,
8(3), 187–208.
Myers, N. (1990), ‘The biodiversity challenge: expanded hotspots analysis’, The
Environmentalist, 10(4), 243–56.
Myers, N. (1991), ‘Extinction “hot spots”’, Science, 254(5034), 919.
Myers, N. (2003), ‘Biodiversity hotspots revisited’, BioScience, 53(10), 916–17.
Myers, N. and A.H. Knoll (2001), ‘The biotic crisis and the future of evolution’, Proceedings
of the National Academy of Sciences of the U.S.A., 98(10), 5389–92.
Myers, N., R.A. Mittermeier, C.G. Mittermeier, G.A.B. da Fonseca and J. Kent (2000),
‘Biodiversity hotspots for conservation priorities’, Nature, 403(6772), 853–8.
Nee, S. (2004), ‘More than meets the eye’, Nature, 429(6994), 804–5.
Norris, K. and N. Harper (2003), ‘Extinction processes in hot spots of avian biodiversity
and the targeting of pre-emptive conservation action’, Proceedings of the Royal Society of
London, B 271(1535), 123–30.
Novotny, V., Y. Basset, S.E. Miller, G.D. Weiblen, B. Bremer, L. Cizek and P. Drozd (2002),
‘Low host specificity of herbivorous insects in a tropical forest’, Nature, 416(6883), 841–4.
O’Connor, C., M. Marvier and P. Kareiva (2003), ‘Biological vs. social, economic and politi-
cal priority-setting in conservation’, Ecology Letters, 6(8), 706–11.
Odling-Smee, L. (2005), ‘Dollars and sense’, Nature, 437(7059), 614–16.
Olson, D.M. and E. Dinerstein (1998), ‘The Global 200: a representation approach to con-
serving the earth’s most biologically valuable ecoregions’, Conservation Biology, 12(3),
502–15.
Olson, D.M. and E. Dinerstein (2002), ‘The Global 200: priority ecoregions for global con-
servation’, Annals of the Missouri Botanical Garden, 89(12), 199–224.
Olson, D.M., E. Dinerstein, E.D. Wikramanayake, N.D. Burgess, G.V.N. Powell, E.C.
Underwood, J.A. D’Amico, I. Itoua, H.E. Strand, J.C. Morrison, C.J. Loucks, T.F.
Allnutt, T.H. Ricketts, Y. Kura, J.F. Lamoreux, W.W. Wettengel, P. Hedao and
K.R. Kassem (2001), ‘Terrestrial ecoregions of the world: a new map of life on earth’,
BioScience, 51(11), 933–8.
Orme, C.D.L., R.G. Davies, M. Burgess, F. Eigenbrod, N. Pickup, V.A. Olson, A.J.
Webster, T.-S. Ding, P.C. Rasmussen, R.S. Ridgely, A.J. Stattersfield, P.M. Bennett, T.M.
Blackburn, K.J. Gaston and I.P.F. Owens (2005), ‘Global hotspots of species richness are
not congruent with endemism or threat’, Nature, 436(7053), 1016–19.
Ovadia, O. (2003), ‘Ranking hotspots of varying sizes: a lesson from the nonlinearity of the
species–area relationship’, Conservation Biology, 17(5), 1440–41.
Parmesan, C. and G. Yohe (2003), ‘A globally coherent fingerprint of climate change impacts
across natural systems’, Nature, 421(6918), 37–42.
Pimm, S.L. (2000), ‘Against triage’, Science, 289(5488), 2289.
Pimm, S.L., G.J. Russell, J.L. Gittleman and T.M. Brooks (1995), ‘The future of biodiver-
sity’, Science, 269(5222), 347–50.
Possingham, H.P. and K.A. Wilson (2005), ‘Turning up the heat on hotspots’, Nature,
436(7053), 919–20.
Prendergast, J.R., R.M. Quinn, J.H. Lawton, B.C. Eversham and D.W. Gibbons (1993),
‘Rare species, the coincidence of diversity hotspots and conservation strategies’, Nature,
365(6444), 335–7.
Pressey, R.L. and V.S. Logan (1994), ‘Level of geographic subdivision and its effects on
assessments of reserve coverage: a review of regional studies’, Conservation Biology, 8(4),
1037–46.
Pressey, R.L. and K.H. Taffs (2001), ‘Scheduling conservation action in production land-
scapes: priority areas in western New South Wales defined by irreplaceability and vulner-
ability to vegetation loss’, Biological Conservation, 100(33), 355–76.
Price, A.R.G. (2002), ‘Simultaneous “hotspots” and “coldspots” of marine biodiversity and
implications for global conservation’, Marine Ecology Progress Series, 241, 23–7.
Redford, K.H., P. Coppolillo, E.W. Sanderson, G.A.B. da Fonseca, E. Dinerstein,
C. Groves, G. Mace, S. Maginnis, R.A. Mittermeier, R. Noss, D. Olson, J.G. Robinson,
A. Vedder and M. Wright (2003), ‘Mapping the conservation landscape’, Conservation
Biology, 17(1), 116–31.
Reid, W.V. (1998), ‘Biodiversity hotspots’, Trends in Ecology and Evolution, 13(7), 275–80.
Ricketts, T.H., E. Dinerstein, T. Boucher, T.M. Brooks, S.H.M. Butchart, M. Hoffmann,
J.F. Lamoreux, J. Morrison, M. Parr, J.D. Pilgrim, A.S.L. Rodrigues, W. Sechrest, G.E.
Wallace, K. Berlin, J. Bielby, N.D. Burgess, D.R. Church, N. Cox, D. Knox, C. Loucks,
G.W. Luck, L.L. Master, R. Moore, R. Naidoo, R. Ridgely, G.E. Schatz, G. Shire, H.
Strand, W. Wettengel and E. Wikramanayake (2005), ‘Pinpointing and preventing immi-
nent extinctions’, Proceedings of the National Academy of Sciences of the U.S.A., 102(51),
18497–18501.
Roberts, C.M., C.J. McClean, J.E.N. Veron, J.P. Hawkins, G.R. Allen, D.E. McAllister,
C.G. Mittermeier, F.W. Schueler, M. Spalding, F. Wells, C. Vynne and T.B. Werner
(2002), ‘Marine biodiversity hotspots and conservation priorities for tropical reefs’,
Science, 295(5558), 1280–84.
Rodrigues, A.S.L., H.R. Akçakaya, S.J. Andelman, M.I. Bakarr, L. Boitani, T.M. Brooks,
J.S. Chanson, L.D.C. Fishpool, G.A.B. da Fonseca, K.J. Gaston, M. Hoffmann, P.A.
Marquet, J.D. Pilgrim, R.L. Pressey, J. Schipper, W. Sechrest, S.N. Stuart, L.G. Underhill,
R.W. Waller, M.E.J. Watts and X. Yan (2004a), ‘Global gap analysis – priority regions for
expanding the global protected area network’, BioScience, 54(12), 1092–1100.
Rodrigues, A.S.L., S.J. Andelman, M.I. Bakarr, L. Boitani, T.M. Brooks, R.M. Cowling,
L.D.C. Fishpool, G.A.B. da Fonseca, K.J. Gaston, M. Hoffmann, J.S. Long, P.A.
Marquet, J.D. Pilgrim, R.L. Pressey, J. Schipper, W. Sechrest, S.N. Stuart, L.G. Underhill,
R.W. Waller, M.E.J. Watts and X. Yan (2004b), ‘Effectiveness of the global protected area
network in representing species diversity’, Nature, 428(6983), 640–43.
Sanderson, E.W., M. Jaiteh, M.A. Levy, K.H. Redford, A.V. Wannebo and G. Woolmer
(2002), ‘The human footprint and the last of the wild’, BioScience, 52, 891–904.
Sarkar, S. (1999), ‘Wilderness preservation and biodiversity conservation – keeping divergent
goals distinct’, BioScience, 49(5), 405–12.
Schmidt, K. (1996), ‘Rare habitats vie for protection’, Science, 274(5289), 916–18.
Sechrest, W., T.M. Brooks, G.A.B. da Fonseca, W.R. Konstant, R.A. Mittermeier, A. Purvis,
A.B. Rylands and J.L. Gittleman (2002), ‘Hotspots and the conservation of evolutionary
history’, Proceedings of the National Academy of Sciences of the U.S.A., 99(4), 2067–71.
Shi, H., A. Singh, S. Kant, Z. Zhu and E. Waller (2005), ‘Integrating habitat status, human
population pressure, and protection status into biodiversity conservation priority setting’,
Sisk, T.D., A.E. Launer, K.R. Switky and P.R. Ehrlich (1994), ‘Identifying extinction
threats’, BioScience, 44(9), 592–604.
Smith, R.J., R.D.J. Muir, M.J. Walpole, A. Balmford and N. Leader-Williams (2003),
‘Governance and the loss of biodiversity’, Nature, 426(6962), 67–70.
Smith, T.B., R.K. Wayne, D.J. Girman and M.W. Bruford (1997), ‘A role for ecotones in
generating rainforest biodiversity’, Science, 276(5320), 1855–7.
Smith, T.B., S. Kark, C.J. Schneider, R.K. Wayne and C. Moritz (2001), ‘Biodiversity
hotspots and beyond: the need for preserving environmental transitions’, Trends in
Ecology and Evolution, 16(8), 431.
Spector, S. (2002), ‘Biogeographic crossroads as priority areas for biodiversity conservation’,
Stattersfield, A.J., M.J. Crosby, A.J. Long and D.C. Wege (1998), Endemic Bird Areas of the
World: Priorities for Biodiversity Conservation, Cambridge, UK: BirdLife International.
Stuart, S.N., J.S. Chanson, N.A. Cox, B.E. Young, A.S.L. Rodrigues, D.L. Fischman and
R.W. Waller (2004), ‘Status and trends of amphibian declines and extinctions worldwide’,
Science, 306(5695), 1783–6.
Supriatna, J. (2001), ‘All sectors of society must work together to save biodiversity’, Nature,
410(6824), 14.
Thomas, C.D., A. Cameron, R.E. Green, M. Bakkenes, L.J. Beaumont, Y.C. Collingham,
B.F.N. Erasmus, M. Ferreira de Siqueira, A. Grainger, L. Hannah, B. Huntley, A.S. van
Jaarsveld, G.F. Midgley, L. Miles, M.A. Ortega-Huerta, A.T. Peterson, O.L. Phillips and
S.E. Williams (2004), ‘Extinction risk from climate change’, Nature, 427(6970), 145–8.
Veech, J.A. (2000), ‘Choice of species–area function affects identification of hotspots’,
Veech, J.A. (2003), ‘Incorporating socioeconomic factors into the analysis of biodiversity
hotspots’, Applied Geography, 23(1), 73–88.
Werner, U. and J. Buszko (2005), ‘Detecting hotspots using species–area and endemics–area
relationships: the case of butterflies’, Biodiversity and Conservation, 14(8), 1977–88.
Whittaker, R.J., M.B. Araújo, P. Jepson, R.J. Ladle, J.E.M. Watson and K.J. Willis (2005),
‘Conservation biogeography: assessment and prospect’, Diversity and Distributions, 11(1),
3–23.
Wilson, K.A., R.L. Pressey, A. Newton, M. Burgman, H. Possingham and C. Weston (2005),
‘Measuring and incorporating vulnerability into conservation planning’, Environmental
Management, 35(5), 527–34.
Wilson, K.A., M. McBride, M. Bode and H.P. Possingham (2006), ‘Prioritizing global con-
servation efforts’, Nature, 440(7082), 337–340.
WWF and IUCN (1994–97), Centres of Plant Diversity: A Guide and Strategy for their
Conservation, Gland, Switzerland: WWF and IUCN.
3. Integrated conservation and development
projects: a positive role for forest
conservation in tropical Africa?1
Neil Burgess, David Thomas,
Shakim Mhagama, Thomas Lehmberg,
Jenny Springer and Jonathan Barnard
Background
A debate has been going on for a number of years on the best ways to
achieve conservation in Africa (and elsewhere). Two elements of the
debate involve those espousing ‘fortress conservation’ and those promot-
ing ‘people-focused conservation’. In some circles this debate has become
highly polarized, with a considerable divide on the best ways to achieve
conservation opening between biologists (Spinage, 1996, 1998; Kramer et
al., 1997; Oates, 1999; Attwell and Cotterill, 2000; Bruner et al., 2000) and
social scientists (Grove, 1995; Neumann, 1996, 1998; Borrini-Feyerabend
and Buchan, 1997; Ghimire and Pimbert, 1997; Hackel, 1999; Leach et
al., 2002). However, for those involved with implementing conservation
projects on the ground in the developing world, the polarized views often
represent impractical extremes. Moreover, for the people living in the
rural areas of developing countries, the divide between ‘development’
and ‘conservation’ is often quite artificial. The third element of the debate
involves attempts to merge human development and wildlife conservation
issues within a single integrated programme, ideally where all sides benefit,
the basis of Integrated Conservation and Development Projects (ICDPs).
These kinds of projects can be considered to fall between ‘fortress conser-
vation’ and ‘sustainable resource use for rural development’. The ICDP
has become one of the dominant approaches to field implementation
of conservation in the developing world over the past 30 years. In this
chapter we look at where ICDPs have come from, what their successes and
failures have been and where they are heading.
Where have ICDPs come from?

In the earlier part of the twentieth century, national governments in
African countries (often the colonial power, but in some cases African
royalty) tended to take a preservationist approach to the conservation of
30
Integrated conservation and development projects 31
wildlife and forestry resources (Neumann, 1996; Schrijver, 1997). Areas of

land were set aside for colonial and royal hunting and local uses were pro-
hibited. In other cases, areas of mountain forests were reserved to ensure a
water supply for people over a broader area (see Rodgers, 1993).
In the later part of the twentieth century the preservationist models
came under significant criticism from social scientists working primarily
with human development issues (Anderson and Grove, 1987; IIED, 1994;
Pimbert and Pretty, 1995; Alpert, 1996; Chambers, 1997). In an attempt to
create a more socially equitable model for conservation in poor develop-
ing countries, conservation agencies borrowed ideas from development
practitioners and created a single integrated model for conservation and
development – the ICDP. The model proved extremely popular, not least
amongst the development assistance agencies from developed countries,
which were charged with assisting developing countries to solve their human
development and environmental problems. This popularity provided signif-
icant new funding opportunities, and the ICDP model was rapidly adopted
by many of the larger conservation NGOs working in developing countries.
The first operational projects were established in the middle 1980s (Hannah,
1992; Stocking and Perkin, 1992; Sanjayan et al., 1997; Larsen et al., 1998;
MacKinnon, 2001; Jeanrenaud, 2002a, b; Franks and Blomley, 2004; Wells
and McShane, 2004). Since that time, hundreds of ICDPs have operated
across the world, especially in Africa where people live side by side with
biological resources (Brandon and Wells, 1992; Brown and Wyckoff-Baird,
1993; Barrett and Arcese, 1995; Fisher, 1995; Caldecott, 1998; Margolius
and Salafsky, 1998; Wainwright and Wehrmeyer, 1998; Newmark and
Hough, 2000; Adams and Hulme, 2001; Hughes and Flintan, 2001).
The ICDP differs from protectionist approaches to conservation in
that the local people surrounding the areas of high natural resource value
also form a focus for project attention (Franks, 2001). In the protected
area management systems of the past, people were often excluded by
force and regarded as ‘poachers’ who should be punished or even shot if
they ventured into the protected area without government permit. If they
suffered costs from living close to the protected area these were ignored
and it was often suggested that they should move elsewhere. This caused
much antagonism between protected area staff and local residents. Most
protected area management agencies have softened their stance in recent
years and many agencies share some of their revenues, assist with local
development projects, or allow local populations to take some resources
from the reserve. Such changes have improved relations between protected
areas and people in many places, and such changes bring the park closer
to the ICDP model.
The ICDP differs from community-based natural resource management
because it always contains a core area protected for conservation reasons

(either government or community-managed), whereas community-focused
approaches may or may not contain core conservation areas (Dubois and
Lowore, 2000; Songorwa et al., 2000; Roe and Jack, 2001). However, in
some of the developing models of community-based forest or community-
based wildlife management the villagers are taking over the management
of former government forest or wildlife reserves. Hence, there is a blend-
ing of the ICDP model and that of community-based natural resource
management – which is perhaps not surprising as many of the management
and implementation issues are the same.
The advantages of ICDPs

There are a number of ways in which ICDPs have been seen as advanta-
geous implementation models in the context of poor African societies. A
major advantage to the poor is that these projects aim to be socially just
(Carney, 1999; Koziell, 2000; Franks, 2001). They do not aim to work to
further the interests of elites, but instead are trying to achieve a long-term
solution to poverty alleviation and natural resource management with the
participation of local communities.
The fact that ICDPs have aimed to address the needs of the rural poor
has also been a major advantage in terms of their acceptability to agencies
engaged in poverty alleviation (Wells and Brandon, 1992). The attention
of ICDPs to development issues has allowed the approach to be main-
streamed within the portfolio of development assistance agencies, which
provide by far the largest source of external funding to poor develop-
ing countries (see, for example, Wells et al., 1999). Stricter conservation
approaches can be funded using private foundation money, or by money
collected by NGOs from wealthy individuals in northern countries, but
this represents a significantly smaller pool of resources.
The declarations from the World Summit on Sustainable Development
in South Africa in September 2002 focused heavily on the alleviation of
poverty and the benefits of good environmental management to achieve
this aim (WSSD, 2002). This was regarded as especially important in
Africa where there remains the greatest levels of global poverty (UNDP,
2001). The potential of the ICDP approach to tackle both poverty and
natural resource management issues provides significant advantages over
other conservation approaches.
The problem with ICDPs

Over the past ten years there have been increasing critiques of ICDPs,
from both the development and the conservation communities. These cri-
tiques arise from a perception that ICDPs have not performed as well as
expected in delivering either conservation or development results. ICDPs

have also been criticized for their focus on defining problems and under-
taking solutions at local levels, when many local problems are driven by
powerful social, economic and political forces external to project sites
(Sanjayan et al., 1997; Larson et al., 1998). Critics have also noted the high
level of funding required for ICDPs, and difficulties in achieving long-term
sustainability. The fact that ICDPs place natural resource conservation
as a central goal is often seen as privileging an international conservation
agenda while the local communities bear the local costs of that conserva-
tion (Leach et al., 2002). For those involved with human development
work, projects should focus exclusively on the needs of poor local people,
rather than find ways in which development activities can be devised to
maintain the status quo of established protected areas (for wildlife or
forest conservation).
On the other side, some conservationists have advocated returning to
the ‘core values’ of conservation – that of protecting wild habitat in offi-
cial conservation areas (generally government owned in Africa), that will
ensure the survival of species and habitats valued at a national or interna-
tional level. It is stated that ICDPs put too much effort into the develop-
ment interests of local communities, to the detriment of the conservation
work (Oates, 1995, 1999).
Among economists there have also been critiques of the ICDP approach
as not being sufficiently direct to create incentives for conservation
(Ferraro and Kiss, 2002). It is argued that direct payments to local com-
munities for the costs they incur from conserving natural resources would
create a direct link between the receipt of money and the conservation
of a resource value. ICDPs do not make such a direct link as the natural
resource benefits are a more indirect product of multiple programme
interventions, including human development.
One difficulty with much of the debate on the value of the ICDP
approach is that it reflects the political and social perspectives of those
framing the debate, what their goals are, and where in the world they
live (Adams and Hulme, 2001). Management approaches for conserva-
tion interventions exist along a gradient from strict protection through
to community management (see matrix on p. 34). Strict Nature Reserves
and most National Parks would come on the left, and community-based
management approaches to the right. As ICDPs have a philosophy of
integrating their management approach, they fall somewhere near the
centre of this gradation. For those people favouring a protectionist
approach to conservation, ICDPs are too ‘social’ in their approach, and
for those focusing on community management, ICDPs represent a means
to support the established government and ‘outsider’ elites.
CATEGORIES OF INTERVENTION
Protectionist Integrated Communal Management

EXAMPLES
Strict Nature Reserve Joint Forest Management Community-based
Forest Management
National Park Community-based
Wildlife Management
Forest Wilderness Area
Improving the ICDP model

We believe that recent experiences with the ICDP model, despite all
the harsh criticism, do not argue for the cessation of the approach. The
blending of poverty alleviation with conservation goals (or conservation
with poverty alleviation) is in line with many international conserva-
tion agreements, and is evidently morally correct in poor regions of the
world. ICDPs are not aiming for conservation at the expense of lives and
livelihoods – but instead are trying to find a solution to this dual goal
through a multidisciplinary approach. This is not only an issue of rel-
evance to conservation staff, as it has been shown that the environment
does matter to poor people in Africa (Posey, 1999; Songorwa, 1999), and
that maintaining a strict divide between ‘development’ and ‘conservation’
is often quite artificial in the minds of the people living in the sites where
conservation projects are implemented.
Instead of abandoning the overall ICDP approach, we believe that the
limitations identified in ICDP design and performance need to be addressed,
new interventions tested and performance of these interventions evaluated
in a rigorous way. Failure to build upon the ICDP model and to truly
create the conditions where livelihood enhancement is linked carefully with
natural resource conservation will almost certainly result in a significant
decline of funding for conservation-related activities. This becomes more
likely as funding from northern governments is targeted to the improve-
ment of livelihoods and the reduction of poverty across Africa.
Through our experience, and from discussions with colleagues and a
review of the literature, we have identified seven issues that need more
attention during the design and the implementation phases of future
ICDPs in Africa.
Scale is appropriate
Recent discussion has focused on whether the ICDP model only applies
to small sites of isolated habitat, or whether it can be used as a model
BOX 3.1 DEFINING GOALS, FOCAL BIOLOGICAL

ELEMENTS AND TARGETS
● A goal is essentially the long-term conservation outcome

you wish to achieve. For example, in a forested landscape
of south-west Cameroon, the goal might be, ‘Conserve
sufficient connected habitat to maintain viable long-term
populations of lowland gorilla, forest elephant, and forest
buffalo’.
● Focal elements refer to the set of biological characteris-
tics that make an area significant for conservation. Focal
elements include species (such as elephants, endemic
birds and plant species), habitats (cloud forest, wetlands,
miombo woodlands) and processes (colonial nesting sites
for birds, elephant migration routes). For example, in the
Cameroon Highlands the endemic birds would be focal
species; the montane forests would be focal habitat; and
the hydrological function of water catchment to lowland
areas would be a focal ecological process.
● A target is the amount, type, and configuration of the
land needed to conserve the focal elements. It might also
contain, in the case of species, a determination of viable
population levels. Following the example above, a target
for, for example, mountain gorillas might be, ‘Conserve
minimum of 500 km2 of interconnected forests ranging
in altitude from 1400 metres to 3000 metres, and access
to water’. A full set of targets should maintain the focal
elements of an area.
for interventions across larger landscapes (Franks, 2001; Burgess et al.,

2002b). In recent years, a landscape approach to the design and imple-
mentation of ICDPs has been in favour, for example in the Participatory
Environment Management Programme in Uganda and Tanzania.2 The
design of ICDPs at the landscape scale is perhaps even more challenging
than at smaller scales. Landscape-level design can use biological features
(forest cover, area-dependent species, ecological processes) to derive the
conservation targets (Box 3.1). Alternatively, the targets for the ICDP
can be derived from analyses of the threats (or pressures) to the natural
resource values (Box 3.2), or can use human socioeconomic indicators to
design the conservation interventions. Whether using biological features,
BOX 3.2 POTENTIAL SOCIOECONOMIC ISSUES

THAT CAN FORM THREATS TO
BIOLOGICAL DIVERSITY
● Patterns of land and resource use:

damaging land and resource use (including forest, water,
wildlife);
damaging development plans (roads, dams, and so on)
or projected changes in land use;
lack of existing zoning regulations;
lack of protected areas.
● Governance and land/resource ownership and man-
agement:
political boundaries (provinces, districts);
insecure or conflicting land tenure (private, public,
ancestral/communal areas);
conflicting responsibilities for management (for example,
Forestry vs. Agriculture Departments).
● Population data:
high population density and growth rates, population
centres;
unhelpful human migration patterns (in- and out-
migration);
high levels of poverty.
threats, or a combination of attributes including human development

needs will make a more targeted ICDP able to function across landscapes
is not yet known, but this thinking may have helped to clarify some of the
structural issues of ICDP design.
There is real local ownership

Experience indicates that ICDPs are more likely to succeed if they have
two levels of support. First, in African countries, it is important that
the government authorities support the project, as it will be impossible
to implement if their support is lacking. Once operational, the issue of
ownership moves to the people who live in close proximity to the natural
resource values the ICDP is seeking to conserve. In general the involvement
of many different groups of people (stakeholders) during the design and
implementation stages of a project is fundamental to making the project
work and for it to become sustainable. However, this is often difficult to
achieve as different groups of local people can have highly conflicting

points of view about the best use of the natural resources (for example,
government forestry/wildlife staff, pitsawyers, commercial agriculturalists,
traditional healers, hunters, rich people/poor people). Making all of these
groups feel that they are involved in the ICDP and that it is working on
their behalf is extremely difficult, and some trade-offs are inevitable.
Protection components are included

It is important for a conservation project to define what it is trying to con-
serve and for whose benefit. In most cases ICDPs have listed the species
that the core habitat area is important for, and perhaps some from the
surrounding lands as a part of the project justification. Many projects
proceed no further, and do not know if the habitat is large enough, suit-
able enough, or in the right spatial configuration to maintain the bio-
logical values. In recent years attempts have been made to structure these
questions, so that an ICDP could establish targets for species, habitats and
important ecological processes as a part of its planning phase. Box 3.1 out-
lines simple approaches for defining focal biological elements and setting
targets, as core parts of the project design. These targets could be built as
indicators into a logical framework for an ICDP, typically at the level of
overall project goal or objectives.
The management of any form of natural resource implies a need for
rules, regulations, boundaries and enforcement mechanisms. It is there-
fore not surprising that supporting protection and regulation is a common
part of ICDPs. Indeed, it has been concluded that protection of the core
natural resource values is essential for all ICDPs (Hannah, 1992; Rodgers,
1993; Abbot et al., 2001). In many African ICDP locations, there is a
government-managed protected area at the core of the ICDP. In such
cases, the government officers traditionally undertake the protection
component. This system works well in some places and poorly in others.
Reasons for a poor performance range from a lack of staff or equipment
to allow them to do the job, through to their personal exploitation of the
resource for their own profit, or for the profit of the government agency
managing the area.
Poor governance by the official managers of the core area (or percep-
tions of it) can result in negative perspectives of the local people towards
the official managers, which makes ICDP models difficult to operate
without considerable work on building trust and cooperation. In other
ICDPs a locally controlled and managed area of natural habitat may be
found instead of a government reserve. In many places these local systems
are long-standing and can provide good protection. If there is no gov-
ernmental protected area, and no traditional management systems, some
ICDPs have attempted to create management systems that take on their

own institutional form over the long term.
A related issue is whether there is an appropriate and functioning legal
framework for the ICDP, and for actions related to conservation and
development. If there are no legal frameworks (either from the national
government, or from traditional leaders) then it is difficult to design
project interventions that require the ‘control’ of use of natural resources.
In many parts of Africa, projects start by working with the government-
authorized village leadership, and gradually discover that there may be an
equally (or sometimes more) powerful system of traditional leadership.
Working with the right people from the outset of an ICDP has obvious
benefits, but is often difficult to achieve.
Project targets are clearly formulated

Importantly, within the context of an ICDP there needs to be a clear
and strong linkage between the development activities of the project and
the conservation objectives of the project. Often the ICDP development
activities are executed separately to the conservation activities and have
no obvious linkage to the overall conservation goals, for example the
improvement of roads, purchase of sewing machines, or improvements in
local health care facilities. In such cases it is difficult to link the develop-
ment interventions to the conservation, which may cause the project to
fail. However, this strategy may work if the linkages have been developed
carefully over a long period of time, or people know that they are receiving
these benefits as a form of ‘payment’ for not damaging species, habitats or
ecological processes.
Most ICDPs are developed by teams of consultants, generally from
outside the area. There is a tendency for these consultant teams to build
assumptions about the state of the local community and the natural
resource management systems into the targets for the ICDP, without
having the data to validate or refute these assumptions. They also bring
their own biases on the ‘best’ ways to achieve conservation in a particular
situation. One kind of assumption that is often inadequately tested is the
source of threats to natural resources. Over-simplistic analyses are often
made of the threats to the system, and to the natural resource value that
the ICDP is targeting. The main threat to a mountain forest may come
from agricultural expansion, but this may not be aimed at providing more
food to the local populations, but rather a commercial crop for export to
townspeople. The clearance of natural habitats to expand the commercial
agriculture may be organized by a powerful person based well away from
the site of the ICDP and who is simply paying local people to work on his/
her behalf.
BOX 3.3 DEVELOPMENT ACTIVITIES THAT ARE

WHOLLY CONSISTENT WITH FOREST
CONSERVATION OBJECTIVES
Activity Benefit
Tree planting Reduces dependence on extraction
from natural forest
Community management Empowers local populations to protect
forest (generally outside reserves)
Agricultural improvement Reduces shifting cultivation, con-
trols soil erosion and reduces loss of
natural forest
Ecotourism Provides a value to the natural forest
resource
Sustainable extraction Provides a value to the natural forest
resource
Small animal husbandry Can reduce hunting pressure in
natural forest (but care has to be
taken to maintain these links)
Alternative incomes Can take pressure from forest (but
care has to be taken to maintain the
beneficial links between the utilization
of the resource and the conservation
of the habitat and species of concern
Assumptions about threats need to be validated before the project

starts, or at least early in its lifespan. Targeted studies using appropriate
specialists can explore different assumptions, and in cases where assump-
tions are shown to be false, this can be a painful experience. Failure to
understand assumptions is another major cause of project failure. Linked
to understanding assumptions is the idea of undertaking better analyses of
the threats to the natural resource value in question. Methodologies exist
to analyse both the surface and ‘root-cause’ threats that are impacting on
a system.
A number of development actions are entirely consistent with the objec-
tives of, for example, forest conservation (Box 3.3). These and activities
like them form suitable interventions for future ICDPs.
One of the best ways to ensure that project targets are well structured
and logically arranged is to use the Logical Framework Approach,3 a
methodology that is popular in both government development agencies
and conservation NGOs. This approach starts with an analysis of threats

to develop a problem tree for an area. By reversing the problem tree a set
of solutions is developed, which forms the basis for the project interven-
tions in the area. The log frame structures the project in terms of broad
goals, narrower objectives and the activities that need to be done to
achieve the objectives (and ultimately the goal). Although some criticize
the framework as too restrictive and not permitting adaptive management
approaches, the ‘log frame’ provides a very effective way to organize a
project. This method has the additional advantage that it can be designed
using participatory methods in workshops of the relevant stakeholders. It
also takes consideration of the assumptions made when proposing actions
to solve threats, and can be used to develop a framework to measure
actual conservation success (a monitoring and evaluation framework)
(Caldecott, 1998; Margolius and Salafsky, 1998).
During the design phase of an ICDP, and throughout its lifespan,
people coming from a natural resource management background (biolo-
gists and foresters and so on) and those coming from a human develop-
ment background (social scientists and development scientists) need to
pool their ideas and approaches to achieve the conservation of an area of
high importance for biological diversity. Compromise over the idealized
scenarios of a number of different stakeholders is an essential component
of the development and long-term management of such projects. The
theme of compromise and negotiation continues into the implementation
phase, and without it the project becomes difficult to manage. The lack
of compromise and a failure to appreciate the points of view of others
involved in ICDP design and management is, we believe, the fundamental
cause of the conflict that has arisen around these projects in recent years.
But it is also the nature of the compromises that need to be made that have
prevented ICDPs achieving as much conservation or development as was
originally envisaged – in most cases a neat win-win situation is not possible
and success is much vaguer and difficult to measure.
Specialized computer software is now available that allows biological
and social data to be integrated easily and cheaply and that might provide
some assistance to the spatial design of ICDP interventions and to achiev-
ing ‘consensus’ over different possible conservation scenarios (Box 3.4).
This process is possible at a remote computer lab, but can also be done
using participatory methods in the field. Stakeholder negotiations can
form a part of this form of ‘land use planning’ and may assist in resolving
some of the conflicting stakeholder interests, or at least exposing them
clearly, as the ICDP is designed and interventions for different geographi-
cal areas are proposed.
BOX 3.4 USE OF SPATIAL DECISION SUPPORT

SYSTEM (SDSS) SOFTWARE TO
INTEGRATE CONSERVATION TARGETS
AND THREATS
What is an SDSS?
A decision support software (DSS) can be generally defined as
an interactive computer-based software to help decision-makers
utilize data and models to solve problems. For our purposes, we
further define a DSS to include a spatial (mappable) component,
commonly provided through a GIS. Therefore a spatial decision
support system (SDSS) is a software program that uses a variety
of spatial data and analytical and statistical modelling capabilities
to answer problems in a map or graphic display. Most SDSSs can
be adapted to meet decision-makers’ needs to solve problems,
formulate alternative displays, interpret and select appropriate
implementation options and modify or include new data.
Why use an SDSS?

SDSS allows users to input a variety of spatial data (that is, species
ranges, future land use plans, cost of land parcels, development
zones) into a computer system and specify a set of requirements
(for example, select areas of forests that maximize tiger habitat
but minimize cost and distance from roads). These requirements
can be modified or changed and therefore are useful in develop-
ing a series of conservation planning options. A common output
of an SDSS is a map or series of maps that indicate various land
use configuration options that will meet conservation, develop-
ment, or combined goals. The outputs of an SDSS can provide
compelling visualization of the conservation landscape that can
aid in communicating the plan with broad stakeholder groups.
However, as with all computer-generated solutions, your answer
is only as good as the data used in the program.
Currently, a range of SDSS software is being used by some

conservation organizations, including SITES (Marxan and
Spexan), C-PLAN, IDRISI, NatureServe DSS, TAMARIN
and GeoNetWeaver. All of these programs can accept both
conservation-orientated and development-orientated information.
Clearly identify livelihoods opportunities (productive potential of different

habitats is taken into account)
In Africa the ICDP approach has perhaps been most successful in savanna
woodland regions with large mammal populations that generate tourist
revenues, and/or can be used for hunting and meat production for local
people. There are numerous examples of successful or nearly successful
savanna woodland ICDPs from southern and eastern Africa (Leader-
Williams et al., 1996; Hulme and Murphree, 1999). In the dense forest
habitats there is a lower density of large mammals, a lower rate of biomass
production that can be hunted for food and a lower potential for tour-
ists to visit the area (see Lukumbuzya, 2000; Wily and Mbaya, 2001).
Most direct forest values have been realized by logging timber trees, or
hunting animals as bush-meat, and conservationists argue that these uses
are unsustainable everywhere they have been attempted in Africa. The
difficulties of designing sustainable management approaches with poverty-
stricken people whose short-term survival is likely to override long-term
management opportunities are well known (Hackel, 1999). Far more of
the forest values are indirect and realized by people away from the forest
edge, for example clean water supply, reliable water flows to downstream
users (both people and industry), carbon sequestration, genetic resource
conservation and so on. Exceptions are rare, but include some of the
montane forests of the Albertine Rift in Central Africa; here the presence
of mountain gorillas makes the conservation of the forests financially
viable. There is a large direct economic benefit to the local people from
tourists and a large international interest in the conservation of the gorilla
as a species.
A sustainable end point is defined

All ICDPs have the ultimate aim of solving resource management prob-
lems and leaving a sustainable system in place that can carry on the work
of the project at the local level, forever. One of the first problems is decid-
ing how to measure sustainability. Ecological sustainability is often dif-
ferent from agricultural sustainability, or economic sustainability. Most
experiences so far indicate that in poor developing countries the problems
are not solved entirely, even after a decade of ICDP interventions, and cre-
ating the funding and institutional mechanisms to sustainably manage the
resources is complex. Typically, an ICDP will remain in an area working
to solve these issues for as long as funding is in place. Once the funding
ends, or is set to end, then attempts are made to leave as sustainable a
system as possible. It is a common perception that interventions quickly
disappear as soon as the project leaves the area and the situation returns
to how it was before the project was operational.
Although hard to achieve, it is perhaps best to try to define a sustainable

end point for the ICDP from the outset, and work towards this end point.
If the natural resource values are in a government reserve then building
the strength of that agency to achieve its mandate in an equitable manner
should assist long-term management sustainability. If the resources are
found in community-controlled areas, then building the capacity and
management authority of locally constituted bodies (village committees or
similar) to take on the role of managing the resource may be the best way
to achieve sustainability.
If the natural resource requires funding to manage, then working to
achieve financial sustainability is critical. Most sources of sustainable
funding for long-term conservation interventions in Africa come from
tourism (for example, ecotourism, see Ashley and Roe, 1998), or use
the natural resource to generate funds. One other source of long-term
funding is the trust fund, where a capital sum is used to generate some
interest to manage the resource in perpetuity. Another mechanism is the
national or local taxation system that might – for example – generate
funds from water users and then provide these funds to the managers of
the relevant watersheds – in particular in some relatively dry countries like
Kenya, Zimbabwe, Tanzania and Malawi. Considerable discussion now
takes place around the issue of whether schemes that entail ‘Payments
for Environmental Services’ can provide a real solution to the matter of
sustainably financing conservation around ICDP sites. The most popular
schemes – and those where there has been the greatest amount of work –
are those related to carbon and water payments.4
A number of other issues can make achieving a sustainable end point
to project interventions challenging. Changing politics and changes in the
local/global economies can alter an apparently ‘sustainable’ outcome to
one that is unsustainable. There may also be local differences in interpret-
ing a sustainable end point. Local farmers may accept a balance between
habitat conservation and agricultural development, but a powerful local
politician may have different aims for use of these resources. Finally,
valued biodiversity sites may need a subsidy (in perpetuity) to have a sus-
tainable conservation outcome, that is, if in economic terms the benefits
from conversion greatly outweigh the benefits from conservation.
Monitoring and evaluation confirms project success

Most ICDPs have failed to measure their conservation impact (Kremen et
al., 1994; Salafsky and Margolius, 1999; Klieman et al., 2000; Danielsen et
al., 2001; Wells and McShane, 2004). Without such measures it has been
relatively easy for them to be heavily criticized (for example, Oates, 1999;
Leach et al., 2002).
A major challenge for ICDPs is to devise systems to collect data system-

atically that show that they have delivered conservation better than doing
nothing (the null hypothesis), or when compared with other conservation
(or development) approaches. ICDPs, especially those set up using devel-
opment assistance funds, were in the past often only required to measure
their activities, such as numbers of meetings held, numbers of people
trained, newsletters produced, study tours completed, accounts produced
on time and audited correctly and so on. Such measurements illustrate
that the project is working according to its schedule and that its activities
are being done. However, this level of measurement does not illustrate
real conservation impact in terms of reducing the threats to the natural
resources within the ICDP area, and does not measure the biodiversity
and societal state of the region before, during and after the project inter-
vention. These fundamental problems are harder to solve. Successful
changes would generally be regarded as a successful ‘outcome’ of the
project, and much effort is now placed on devising monitoring systems to
look at project impact and the achievement of the relevant outcomes.
In addition to assisting project design, the Logical Framework Approach
can also be used to assist in the development of a scheme for measuring
the effectiveness of an ICDP. At all levels of the log frame (but especially
at the higher levels of the goal and objectives) the development of meas-
urable indicators of pressure (threat), state (biodiversity and people) and
response (biological and people) would provide a real mechanism to assess
ICDP success. Monitoring of activities and budgets and reports produced
on time and so on, would continue to be important as well, but if conser-
vation success were measured across a wide suite of ICDPs this would
provide the data to answer many currently unanswerable questions.
In addition to ICDPs themselves seeking to measure their conservation
success by looking at how effectively they address threats, preserve or
change the biodiversity state, there is also a clear need for multiple ICDPs
to roll up their results into a more formal assessment of the success of the
approach in general. Currently most of the statements of project success
or failure are opinion- or case-study-driven. We are not aware of any study
that uses quantitative data to test the success of ICDPs, when compared
with any other conservation mechanism, either in monetary terms, or in
terms of conservation delivery or livelihood improvements. This lack of
hard data allows considerable unproductive debate to occur.
We believe that monitoring and evaluation schemes that look at con-
servation impact need to be built into every ICDP. For species, surveys of
endemic, rare or focal species can provide the evidence that these elements
of value are maintained, or increased. Satellite, aerial photographic or
ground surveys of habitats (and reserve boundaries) provide an additional
measure whereby the success of the project at conserving biological

elements of value can be assessed. Rates of extraction, levels of distur-
bance and habitat ‘condition’ provide a further tranche of data whereby
conservation impact can be determined. Measuring the effect the project
has, such as reduction of threatening activities, increased protection, fewer
purchases of illegally hunted animals, and so forth links the response of
society to the state of the resources. Surveys of the knowledge, attitudes
and practices of the local population provide an indirect measure of the
success of a project at changing practices and perceptions – which can be
repeated to assess change over time.
Measuring development impact is also possible, for example using
measures such as numbers of people within the project area who are short
of food, do not send their children to school, or have very poor levels of
agricultural productivity. As income levels increase, then the number of
people making a living from the sustainable use of the forest resource,
or forest-related activities (for example, tourism) can also be used as a
measure of development impact. Measuring the combination effect of
conservation and development interventions within an ICDP is the most
challenging issue of all. We are not aware of any attempt to measure the
synergistic effects of different interventions.
Building success in ICDPs in the forests of Africa

Here we look at two case-study ICDPs from African forests and assess
how well they have addressed, or are addressing the issues we outline
above as fundamental concerns in improving the performance of ICDPs in
the region. These projects are ones with which we have had a long personal
involvement, hence our comments are certainly not unbiased. However,
due to this involvement we also know some of the failures that the projects
have experienced, and have seen in some cases how these have later been
solved.
The Uluguru Mountains of Tanzania

The Uluguru Mountains are located in eastern Tanzania (Figure 3.1)
and comprise a forest-capped area covering 1500 sq km of highlands,
with forested areas found discontinuously from 150 m up to 2630 m
above sea level. The main ridge runs almost north–south, and there are
a number of smaller outlying hills that are broadly regarded as a part of
the Uluguru Mountains landscape. Administratively the area falls within
two Tanzanian rural districts and a part of one urban municipality. More
than 50 villages have a border with the larger forests of the area, which are
found in 22 government Forest Reserves (Figure 3.2a). Over 100 villages
cover the entire Uluguru Mountains landscape, and these often have their
Note: The dark line denotes the boundaries of the river basin that are linked to the
Eastern Arc.
Source: Valuing the Arc Programme, Cambridge University, UK.
Figure 3.1 The location of the Uluguru Mountains in eastern Tanzania in

relation to other mountain forest blocks in the ‘Eastern Arc’
biogeographical region of Africa
own smaller patches of forest, often for traditional spiritual purposes, and
50 of these villages touch the forest margins (Figure 3.2b). Population
density is somewhat variable, but typically there are high densities of
people up to the borders of the Forest Reserves and in some parts of the
area human populations increase at higher altitudes where the farming
Morogoro town
Uluguru North
Forest Reserve
North
20 km
North
Forest
reserves
Village Uluguru South
boundaries Forest Reserve Boundaries between
villages
Ward Roads and tracks
a) boundaries b)
Forest outside the
reserves in this area
has been extensively
converted to banana
plantation
Pitsawing
Fire wood
collection
Commercial
agriculture
Mining for
rubies
c) Roads d)
Note: a) location of government-owned Forest Reserves, b) human population density in

the wards around the Ulugurus in 1988, c) main threats facing the forests of the area, and d)
main area of project intervention during Phase II DANIDA support.
Source: WCST Uluguru Mountains Biodiversity Project.
Figure 3.2 The Uluguru Mountains landscape, showing geographical

features of importance when designing an ICDP
potential is better (Figure 3.2c). Most people living on the mountains are
poor (Hartley and Kaare, 2001). For an outsider it is quite a daunting
region and a car can only reach a few higher regions of the mountain.
However, for the people of the Ulugurus, paths link nearby villages and
there are paths crossing the mountains to facilitate longer-distance travel-
ling. It is often quicker to walk (or run) over the mountain than it is to
drive around it. Distances are measured in hours of walking.
The biological importance of the Uluguru Mountains is well known,

with more than 135 species of strictly endemic plants and 16 species of
strictly endemic vertebrates (Lovett and Pócs, 1993; Lovett and Wasser,
1993; Svendsen and Hansen, 1995; Burgess et al., 1998, 2002a; Burgess
and Clarke, 2000; Doggart et al., 2005). The mountains are also impor-
tant hydrologically as they provide the source for the main water supply
to the largest city in Tanzania, Dar es Salaam (see for example, Pócs,
1974, 1976), grow considerable quantities of food for export to towns and
are home to over 100 000 people – primarily from the Luguru tribe. The
Ulugurus therefore have high international values for biodiversity conser-
vation, high national values for water supply to the national and regional
capital cities, and high local importance as a living place for the Luguru
people.
Externally funded ICDP activities have been undertaken in the
Ulugurus for over a decade (see for example, Bhatia and Buckley, 1998;
Burgess et al., 2002b), but have tended to cover small geographical areas
(Figure 3.2d). Three project phases are recognized here:
● Phase 1. The planning phase where socioeconomic surveys were

undertaken to try to understand the interventions that were appro-
priate and needed in the area (Bhatia and Ringia, 1996; Bhatia and
Buckley, 1998). This phase was funded by the European Union
and the Royal Society for the Protection of Birds and worked with
Tanzanian NGO and government partners.
● Phase 2. This phase (after a gap of three years) started to imple-
ment the findings of the preparatory studies. Danish International
Development Agency (DANIDA) funded this part of the work,
which involved Danish (Dansk Ornitologisk Forening – DOF) and
Tanzanian (Wildlife Conservation Society of Tanzania – WCST)
NGOs and Tanzanian government staff.5
● Phase 3. This phase represented an expansion of Phase 2, where
the existing Danish government-supported programme was joined
by a second project funded by GEF and managed by the devel-
opment agency, CARE. The DANIDA-supported programme
started in 2002 and the GEF one started in 2003. Both ran for
five years.
The focal area for intervention by these projects is the region of highest
forest biodiversity and greatest threats to the forests. Until 2002, this
caused project actions to concentrate in the forest remnants outside
the official government Forest Reserves on the northern end of the
range (Figure 3.2d). The area was heavily forested in 1955 (from aerial
Public land forest
Uluguru North
Forest Reserve
Boundary
Uluguru North
Catchment
Forest Reserve
Forest 1955
Tandai Village
Forest 1977
Forest 2000
Note: This area is closely co-incident with that of the focal project area as it is here where
most conflicts between biodiversity conservation and human use of the land occur.
Source: WCST Uluguru Mountains Biodiversity Conservation Project.
Figure 3.3 Loss of the non-reserved forest outside the Uluguru North
Catchment Forest Reserve, 1955–2000
photographs), but had been largely deforested by 2000, apart from some
remnants (Burgess et al., 2002a; Figure 3.3). The project attempted to
preserve these fragments by building them into the village land use pattern
as ‘Village Forest Reserves’. In the Phase 3 project there will be a greater
spread of interventions around the mountain, both north (DOF/WCST)
and south (GEF/CARE), and more funding for the conservation of the
Forest Reserves that contain the highest levels of biodiversity.
Scale is appropriate The Ulugurus are a large and topographically

complex range of mountains and are occupied mainly by one tribe, the
Luguru, although an influx of other tribes is noted. Biologically there is a
considerable similarity in the flora and fauna of the forests across the entire
mountain range, with the major differences being due to altitude. Due to
these factors it makes sense to consider a large-scale ICDP design across
the entire Uluguru landscape. During Phase 2 support there was a mis-
match between the scale of the project planning and monitoring (covering
the entire landscape) and the scale of project interventions (Figure 3.2d).
Project interventions were focused in a smaller part of the mountain that
had the greatest problems in terms of forest and biodiversity loss (Figure
3.3), and where there was the highest rate of commercial agriculture and
impact from outside interests. During the design of the third phase of
activities on the mountain, efforts were made to expand activities to cover
the entire landscape. However, funding remains insufficient to fully tackle
the whole area.
There is real local ownership The earliest phases of project input into
the Ulugurus were designed and implemented in collaboration with
regional government partners, who managed the programme as their own.
However, these earlier phases did not adequately involve the people living
around the mountains and apart from studies, training, and some tree
planting efforts, no significant impacts of project interventions were felt in
the villages on the mountain.
The second and third phases of this ICDP engaged with district govern-
ment officers, village governments, farmers’ groups, tree planting groups,
women’s groups and government forestry officers based in the mountains.
By working at these levels the project has found greater resonance with
local people, but has also experienced some problems of ‘partnership’,
especially within the elements being coordinated by CARE. By working at
the local level, the communities agreed to set aside patches of both natural
forest and human-made woodlots as Village Forest Reserves. Towards
the end of the second phase, the project started to work closely with the
traditional chiefs as well as with a broader range of government-appointed
staff in the villages (Figure 3.4). This has further improved the perception
and ownership of project interventions among the local communities, and
has also raised awareness of the project and the conservation issues of
these mountains at the national level. Changes to a more people-focused
approach to forest conservation in the Ulugurus have been facilitated
by the publication of a new Forest Policy (GOT, 1998), Guidelines for
Community-based Forest Management (GOT, 2001), and a new Forest Act
(GOT, 2002). These legal changes provide the mechanisms to empower
local populations to manage forest resources, either in collaboration with
the central government, or alone.
Project targets are clearly formulated The initial project, starting in

1993, was designed using the traditional approaches – teams from outside
working with government authorities in the nearby towns developed
project proposals based on their best-available understanding of the situa-
tion on the ground, and some targeted case studies (Lyamuya et al., 1994).
Output 2.1, Jan 1999 onwards

Output 1.1, Jan 1999 onwards
400
Implementation
2500
Implementation
300
2000
scores
scores 1500 200
1000
100
500
0 0
1 2 3 4 5 6 7 1 2 3 4 5 6 7
Implementation periods Implementation periods
Output 2.2 January 1999 onwards Output 3.2, January 1999 onwards
Implementation scores
6 200
5
150
4
3 100
2
50
1
0 0
1 2 3 4 5 6 7 1 2 3 4 5 6 7
Output 4.2, June 1999 onwards Output 4.4, January 1999 onwards
600
90
500 80
70
400 60
300 50
40
200 30
20
100
10
0 0
1 2 3 4 5 6 7 1 2 3 4 5 6 7
Note: Implementation periods refer to six month reporting periods to DANIDA, with
the first reporting period (1) being the baseline when the project started. Scores relate
to a variety of measures easily gathered by the project management, often with several
measures added to produce this index of change. The scheme was not continued for the
full three years of the project phase due to changes in staff and because political changes in
government priorities in Denmark resulted in much additional work for staff in Tanzania.
Explanation of outputs:
Output 1.1 Administration of WCST and skills of its staff enhanced.
Output 2.1 Activities provided to involve Tanzanian members and volunteers in the work
of the Society.
Output 2.2 Training provided to develop skills of members and volunteers as a resource
to promote the work of the Society.
Output 3.2 Uluguru project assists the formulation of agreements between District
Forestry and local people on sustainable uses of the Uluguru Public Land.
Output 4.2 Uluguru project supports sustainable agricultural practices in villages adjacent
to Uluguru Forest Reserves.
Output 4.4 Uluguru project produces written materials to promote its work within
Tanzania and internationally.
Source: WCST Uluguru Mountains Biodiversity Conservation Project.
Figure 3.4 Simple measures of success in the implementation of Logical

Framework Outputs during part of the second phase of the
Uluguru ICDP (with DANIDA funding)
Between 1993 and 1995 extensive further studies were undertaken to feed
into the development of further ICDP activities in the Ulugurus (Bhatia
and Ringia, 1996; Bhatia and Buckley, 1998). The results of these studies
were used to design the Phase 2 DANIDA support on this mountain. At
this point a logical framework approach was used to capture the links
between the inputs and outputs of the project, and the ways to measure
conservation achievement. Planning of the Phase 3 project involved
Tanzanian and foreign development experts, social scientists and a series
of consultative meetings with relevant local people living on the mountain
to further refine the series of project interventions (Hartley and Kaare,
2001). A further detailed logical framework was developed, but despite
the increased stakeholder involvement it was quite similar to that from
the second phase. Designing the third phase did, however, highlight the
importance of the traditional leadership on the mountain. The village gov-
ernment that was part of the Tanzanian government structure held some
power, but it was not in charge of land allocation, which passed through
the female line of different clans of the Luguru people. By looking closely
at different stakeholders, the Phase 3 proposals were able to capture the
facts that:
● Almost all government officers (national, regional, district, divi-

sion, ward and village) wished to retain the reserves in their present
form to ensure the maintenance of national (water flow) and local
(better climate for farming) values. As the reserves had been under
government control for a long time (almost 100 years) most of the
local population also accepted their existence as a part of the local
pattern of land use.
● The traditional chiefs wanted to retain the Forest Reserves (and
other non-reserved forest areas) as places where their ancestors
could remain undisturbed, and also because of some sacred sites and
so animals (in particular a mythical giant snake) could find refuge.
● Some groups of people within the villages also wanted to retain the
Forest Reserves as sources of clean water, medicinal herbs (tra-
ditional healers), for hunting, to provide fuel and building wood
(which was often not available in farmland areas), or for spiritual
reasons (a place for the ancestors to live).
However, other stakeholder groups wished to obtain greater freedom of

access to the forests to pursue a number of activities:
● For some people, growing bananas has become a lucrative activity

(Hymas, 1999). This crop grows best on recently cleared forest soil,
and has resulted in the loss of over 20 sq km of non-reserved forest

on the Ulugurus over the past 20 years (Burgess et al., 2002a, 2002b).
Some of those people involved in this activity would like to expand
their production into the reserved forest, which would provide even
greater financial rewards.
● For most people farming on the Ulugurus, their crop yields are very
low, typically only one to three bags of maize per acre of cultivated
land (Hymas, 2000, 2001). Much better yields can be obtained from
newly cleared areas of forest – at least for a few years. Hence many
farmers have a desire to obtain ‘fresh’ land from the forest, and
some manage to obtain ‘permissions’ to cultivate on the edge of the
Forest Reserve from village authorities.
● Pitsawyers regarded their extraction of high-value timber from the
forest as beneficial as it generated money, did not destroy the forest
and was a better option for using the forest than converting it to
farmland. In areas being cleared for banana farming the pitsawyers
were actively sawing up felled trees, and expressed regret at the loss
of their future work opportunities through the clearance. However,
they also wished for greater legal access to the reserve so that they
could extract more trees and sell them – activities that were currently
illegal.
● Other villagers sought the right to access the forest to collect fire-
wood and other woody products that they need (and already obtain
illegally) from the forest (Hymas, 2000). Even the issue of crossing
the mountain through the forest was a vexed one – the paths across
the forests are essential lifelines for villagers, saving days of travel-
ling in some cases, but they do not have the official right to use the
paths, even though they are centuries old in some cases.
The biggest compromise during the design process from the side of
those wishing to retain the reserves was to agree to explore methods of
collaborative management of the forest resources with forest-adjacent vil-
lages. Devolution of power and authority from district and national gov-
ernment level to village and traditional authorities would be required for
this, and its effective implementation will need a period of negotiation and
formalization through signed agreements (bye-laws). Currently ICDPs on
the Ulugurus have only been gathering experience on the potential ways to
achieve this goal by working in remaining patches of non-reserved forests
to establish Village Forest Reserves. The experience provides some models
that can be applied to the village communities living around the official
government Forest Reserves.
Protection components are included The Uluguru ICDP is working at

three levels to ensure that forest protection components are included. Most
of the remaining natural forest and the highest biodiversity values are
found within large Forest Reserves on the top of the mountain managed by
the Catchment Forest Project, a branch of the Forestry Department that is
under central government control. These reserves are maintained for their
water catchment functions. Smaller patches of forest are also found in a
number of Local Authority Forest Reserves managed by the district, gen-
erally on the slopes of the mountains. These smaller forest patches are of
lower biological importance when compared with the Catchment Forest
Reserves. Finally, there are traditionally protected areas of forest on the
Ulugurus, which are under the authority of the traditional chief (one large
area) or individual villages or clans (smaller patches). The current Uluguru
ICDP is working with Catchment Forestry, with the district Forest Officer
and with the traditional chief and the relevant village authorities to try to
enhance the protection of these different categories of forest.
However, the funding allocation within the Uluguru ICDP for protec-
tion activities is modest. Most funding has been provided for developing
arrangements between the local village authorities and the government
agencies over some kinds of collaborative management of the forest on
the Ulugurus. The extent that nationally important Catchment Forest
Reserves should be managed collaboratively with communities, and what
the community benefits will be, remains a major issue of debate. The
current view is that collaborative management should go ahead, but that
much more work is required to identify the benefits that communities can
get. Another reality is that with only three Forest Officers stationed on the
mountain, but with more than 50 villages and several tens of thousands of
people, there remains a strong need to involve villagers in the management
of these reserves. Since 2005, it has also been proposed that the Uluguru
Forest Reserves be upgraded to the status of Nature Reserve. This status
would not prevent collaborative management agreements, but might
bolster the government management capacity.6
Productive potential of different habitats is taken into account Resource

extraction agreements have not been formulated for the Uluguru
Mountains, as legally the majority of the forest (within reserves) is pro-
tected and utilization is not allowed. However, the local villagers do utilize
the forests as a source of firewood, building materials and medicines.
There is also some pitsawying of valuable timber species. If and when
agreements are formulated between the local villages and the catchment
forestry managers, they will need to consider the potential of the forest
to supply materials sustainably. Hunting pressure has already removed
large mammals and reduced populations of smaller ones, and some

timber species are no longer of harvestable size due to past logging. The
true sustainable level of resource utilization on the mountain has not
been assessed, and will be difficult to achieve because of the centuries of
exploitation that have already occurred.
A sustainable end point is defined In the Ulugurus the maintenance of all

remaining forest cover, the reconnection of separated forest patches, and
full local participation in forest management could represent a sustain-
able end point. Having the funding in place to manage these areas forever
would also represent a sustainable end point. But such sustainability has
not been defined by the existing projects and given the scale of the area
and the problems that it faces in terms of population growth and resource
needs, reaching a sustainable conclusion is highly problematic. In simple
financial terms, the various government offices with a responsibility for
the management of the reserves on these mountains lack the funds to
effectively undertake their jobs. Moreover, the local populations are poor
and the forests do not provide them with much in terms of cash revenues.
It’s much more profitable for them to convert the forests to banana plan-
tations or other forms of agriculture, at least in the short term. But the
forests have huge national cash value as a source of water to the capital
city and much of the industry of the country. They also have an interna-
tional value as the home of hundreds of unique species including species of
commercially valuable genera (African violet, busy lizzie, begonia, coffee
and so on). If these indirect values could be captured using tax on water
users, or through other kinds of monetary system then the management
of the Ulugurus over the longer term could be ensured. The World Bank
has recognized the indirect monetary values to the Tanzanian economy
and started a trust fund mechanism for the region during 2002, with
initial capitalization of $7m. In its initial phase the trust fund will not
fund projects in the Uluguru Mountains, but may in future years. Current
projects through WWF-CARE-IIED are investigating whether the estab-
lishment of water payment schemes (Payments for Water Environmental
Services) is a viable conservation funding mechanism for the Ulugurus.
Such a mechanism could provide the required management funds, while
at the same time ensuring that the natural values required for Tanzanian
development (water supply) continue to be provided.
In many ways these forests epitomize the problems of ending an ICDP
programme – there is no end point because management interventions will
be needed forever to assist agriculture to improve, to resolve disputes over
land use, to maintain the existing reserves and so on. Working to develop
sustainable funding mechanisms is perhaps the only way to achieve the
sustainability of project interventions and thus end the ICDP cycle of

donor support.
Monitoring and evaluation confirms project success The Uluguru ICDP

has undertaken various forms of biological, social, habitat, disturbance
and attitude surveys since 1993.7 Most of these studies are baselines
against which future trends can be measured. A few studies have also tried
to look backwards in time to assess what has changed over time, in terms
of habitat and species values (see for example, Burgess et al., 2002b). The
WWF-World Bank Management Effectiveness Tracking Tool has been
recently used to gauge the impact of project activities on the Uluguru
Forest Reserves, showing an improvement over time. The other system
that might have been used to measure the impact of conservation interven-
tions is the ‘Threat Reduction Assessment’ methodology, which measures
the extent to which the projects have reduced pressures/threats in the area
(Salafsky and Margolius, 1999).
Conservation and development in the Bamenda Highlands, Cameroon

The Bamenda Highlands support remnant areas of montane forest of
global biodiversity importance for endemic species, within a landscape
dominated by farmland and high human population densities. People are
poor and derive most of their livelihoods from farming and exploiting
natural resources. Most of the natural forests remaining in this area are
managed either by traditional authorities, or by the more remote govern-
ment forest department.
The Kilum-Ijim forest (Figure 3.5) forms the core area of the community-
based conservation in the Bamenda Highlands Programme. This forest
extends over about 17 300 hectares on the slopes of Mount Oku (3011
metres) and the adjoining Ijim Ridge and is the last significant remnant
of Afro-montane forest in West Africa. Mount Oku lies in the Bamenda
Highlands, part of the Cameroon Mountain chain. The forest is a globally
important centre of endemism: 15 bird species endemic to the Cameroon
Mountains can be found at Kilum-Ijim, of which Bannerman’s turaco
(Tauraco bannermani) and the banded wattle-eye (Platysteira laticincta)
are restricted to the Kilum-Ijim forest and a few other forest remnants
within the Bamenda Highlands. Both species are threatened and the forest
represents the only possibility of conserving viable populations of these
two species. Surveys of other taxa also demonstrate very high endemicity,
not only of the Bamenda Highlands, but of the Kilum-Ijim forest.
The alternative forest conservation strategies suggested at the start
of the project in the 1980s were either to advocate the establishment of
a protected area (an option that had been proposed and attempted, by
0 100 km Lake
0 10 km
NIGERIA
CHAD
CAR
CAMEROON
KUMBO
ATLANTIC
OCEAN Mt. Oku
EQU L GABON CONGO
GUINEA s
n d
BAMENDA a
l
h
g
i
H
a
d
n
e
m
a
B
KEY
Forest cover
Land over 1800m
Source: BirdLife International – Cameroon Highlands Project.
Figure 3.5 Location of forest patches that form part of the Bamenda
Highlands ICDP network, within the geographical extent of
the Cameroon Highlands
government, for several decades) or to support conservation without an

official protected area that was based instead on community agreements
and sustainable use of forest resources. When the project started in 1987,
the first step was to agree and demarcate a forest boundary. This was
urgently needed to avoid further loss to encroaching farmland, and the
project worked closely with community leaders to agree a boundary that
was then marked with Prunus africana trees. In these earlier phases, the
project focused on education and awareness, sustainable use of forest
products (especially honey) and a ‘livelihoods’ programme that aimed to
assist farmers to improve production on their farmlands bordering the
forest. Following the passing of a new Forestry Law in 1994, the Kilum-
Ijim Forest Project started a major new phase, in which the focus became
support to decentralized forest management by adjacent communities
(under management plans drawn up by the community and approved by

the Ministry of Forests and Wildlife (MINFOF)).8
Scale is appropriate Initially the project focused only on the Kilum-Ijim

forests. But over time the approach was requested by more and more
surrounding villages with their own patches of similar forest and similar
development issues to be solved. Therefore, whilst the Kilum-Ijim forest is
still the top conservation priority, the project was expanded to a regional
scale of approach, and the scale of operations became the wider Bamenda
Highlands. Under this enlarged programme, communities in the region
were invited to approach the project for assistance – but support came with
conservation-related conditions attached. This aimed to ensure genuine
commitment on behalf of the village before any work commenced. The
project also comprised components focused on building a local constitu-
ency in support of forest conservation (a diverse education and awareness
programme) and components focused on capacity-building for supporting
community forest management within government (MINFOF) and local
NGOs. Indications are that this approach had significant success in some
components – especially the education and awareness programme.9 Possibly
as a result of the success of this awareness campaign, many more communi-
ties approached the project for support for forest management than had
been anticipated (the project planned to work with 30 communities, and
focus on supporting the eight highest priorities, but received over 60 appli-
cations for support). The project was reluctant to turn communities away
and so initially attempted to respond to all inquiries. However, it was clear
that limited resources required activities to be more focused, and the project
subsequently prioritized ten key communities who received high levels of
direct support, 15 communities who received limited direct support, and
supported the remaining communities indirectly through local NGOs.
Within the context of national and local constraints (for example, a
poorly resourced and motivated civil service, political interference, inad-
equate levels of funding), the regional, landscape approach was judged a
success. It allowed cost-effective use of resources and materials (trained
staff, methodologies, awareness materials), encouraged engagement of a
wider constituency both in government and civil society (since Bamenda,
the Provincial capital, lies at the heart of the project area) and has the
potential, through the institutional structures built through the project,
to justify a mechanism for the long-term management of the forests (see
below). The attention to the wider Bamenda Highlands landscape has also
revealed the biodiversity importance of many of the smaller forest frag-
ments, which had previously not been well surveyed, and helped to put
Kilum-Ijim into a wider context. This approach would seem to be suited to
similar landscapes, where habitats of high biodiversity value are scattered

in an agricultural landscape, but of course solutions need to be situation
specific.
There is real local ownership When BirdLife International (then ICBP)

began working at the Kilum-Ijim forest in 1987 one of the first steps was to
assist local communities in the demarcation of a forest boundary beyond
which no further clearance for agriculture would take place. Nineteen
years on that boundary remains largely intact, and traditional authorities
have dealt with the few infringements that have taken place promptly and
effectively. Each of the communities surrounding the forest has a working
and legally recognized Forest Management Institution (FMI) and 11 com-
munities have produced forest management plans and now hold legal title
to the resources of their forest (with the management responsibilities that
implies). The project has supported communities in the process of devel-
oping the necessary institutional capacity, and with the highly complex
legal process of forest registration and management plan approval. This
has been combined with a development programme that has demonstrated
methods of improving productivity and sustainability of land use outside
the forest, as well as enhanced production and value of products harvested
from inside the forest (such as honey).
There is real overlap at Kilum-Ijim between long-term community
interests from forest conservation (especially watershed protection but
also non-timber forest products like honey, medicinal plants, fuel wood
and bamboo) and biodiversity conservation. Most of the remaining forest
is on very steep slopes and conversion to agriculture is almost certainly
not sustainable (as landslides in adjacent areas testify) and would lead to
loss of watershed protection functions (again, as demonstrated by many
adjacent areas where springs have dried up). However, whilst most people
recognize the convergence of biodiversity conservation interests with their
own social and economic concerns, and therefore support the project,
the process continues to be undermined by individuals or groups with no
immediate interest in forest conservation – especially graziers (most of
whom are wealthy individuals who are no longer living in the area) and
Prunus africana exploiters (external, and very often armed). Regrettably
these interest groups have been able to influence the judicial process, and
their illegal activities (which are also unwanted by the community) have
been extremely difficult to control. What this shows is that the interests
and wishes of the majority, who have organized themselves into demo-
cratic community institutions, and who have legally backed rights, can
still be frustrated by powerful or stubborn individuals in situations where
capacity (or commitment) in state authorities is lacking.
Project targets are clearly formulated Over the 19 years that BirdLife
International has been working in the Bamenda Highlands, the plan-
ning mechanisms have been constantly developed. Project interventions
were based on a detailed logical framework that was itself created from
a detailed problem tree developed with project staff and villagers around
the forests.
Detailed logframes and workplans have proved critical tools in the
regular cycle of planning, implementation, monitoring and evaluation, and
have been used in a flexible way, with managers prepared to modify objec-
tives (with the agreement of donors) if circumstances made that appropri-
ate. However, what the full problem tree demonstrates is an immense web
of problems that can only partly be solved by a project approach of fixed
duration (and budget). The Bamenda Highlands project aimed to put in
place a sustainable outcome within the constraints of politics and national
economics (the context – national debt and an underpaid civil service).
These factors are not within a single project’s ability to influence and yet
clearly they significantly affect what can be achieved – and the strategies
that must be put in place to try to accomplish a project’s objectives.
Protection components are included A 1000 hectare plantlife sanctuary

forms a core protection area in the centre of the Kilum-Ijim forests, and
the people-centred approach at Kilum-Ijim also takes place within a legal
context (provided by a Provincial Decree) that imposes (state) controls on
land use (banning inappropriate activities such as burning and grazing).
The project additionally chose to support a traditional protected area in
the form of an ancestor living place (forest) under the authority of the
traditional chief (fon). This traditional protected area has been formal-
ized through the creation of local forest management institutions around
the forest who ensure that the boundaries are respected and that resource
utilization agreements are adhered to. Effective implementation of the
management plans they produce is in turn monitored and enforced by
MINFOF. Although different from the model of a government-managed
Forest Reserve or National Park, the proximity of the people to the
resource and their acceptance of traditional authority make this model
work. Although no detailed comparisons have been made it is believed
that the level of protection of the forests under the Bamenda Highlands
programme is stronger than those under government protection. The
forest is highly valued culturally, and many taboos and regulations exist
that are enforced by the fon and his ruling council (kwifon). De jure gov-
ernment ‘control’ over forests had eroded this traditional authority and
management system, without replacing it with an effective alternative
(MINFOF Forest Officers were too few, poorly resourced and often not
motivated to manage the forest). The project has helped to rebuild the
traditional authority of the fon and kwifon within a parallel and integrated
system of management and control by community-based institutions of
users (FMIs). However, experience shows that government support is still
essential as a back-up and endorsement of FMIs and traditional authority
rules, especially where powerful outside (and internal) interests threaten
to destabilize the situation. For example, the valuable bark of Prunus
africana is illegally exploited by outside gangs that are often armed, and
illegal livestock graziers in the forest enter with the backing of wealthy
and influential livestock owners. This experience suggests that a three-
pronged approach (FMIs–traditional authorities–government) may be the
most effective combination, and that each group has its role to play in an
effective overall protection system.
Productive potential of different habitats is taken into account The project

has made efforts to assess the sustainable levels of offtake for key forest
products – those studies examined harvesting and production of fuel
wood, honey and small mammals (mainly rodents), and future work
will attempt to assess the sustainable offtake for Prunus africana bark.
These studies suggest that the forest is already exploited at levels equal
to or exceeding its productive potential. For resources such as mammals
this is self-evident – it is very rare now to see duikers, monkeys or other
mammals in the forest, and most hunters have resorted to trapping a
range of small rodent species. Although the reproductive potential of these
species is probably very high, their detailed biology in this respect is not
known and in most cases studies have not been carried out. Since rodents
are likely to be key seed dispersers for some tree species the long-term
impact of over-harvesting on ecosystem functioning could be severe. The
project’s approach of integrating conservation and development aimed
to address this issue, by finding alternatives (supplements) to harvest of
products from the forest.
A sustainable end point is defined The main projects in the Bamenda

Highlands lasted for 17 years and cost a total of over US$3m with the
objective of conserving montane forests with a total area of about 30 000
ha. In its very early stages the project was ‘fire-fighting’ – its priority was
to try to halt the rapid encroachment of farmland into the forest. The
strategy then shifted to education and awareness and building sustainabil-
ity of local livelihoods through support to agricultural development and
income-generating activities linked to the forest. However, it is now recog-
nized that the conservation of the forest is a long-term enterprise, and that
to be sustainable key elements are going to be strong institutions to carry
the process forward (first and foremost) possibly combined with some
form of ‘subsidy’ to enable national government and local communities to
conserve a resource of global significance into the future. As a result the
present project strategy for sustainability is based on achieving sustainable
institutions for the long-term management of the forest. This will work
at three levels – that of community institutions (FMIs), the traditional
authorities and the government. As suggested above in relation to pro-
tection activities, each of these levels needs to work effectively for forest
management to succeed. It was envisaged that these three institutional
elements would be supported financially in the long term through a trust
fund (CAMCOF – CAmeroon Mountains COnservation Foundation) to
support sustainable management and conservation of forests through-
out the Cameroon Highlands, with an initial focus on the Kilum-Ijim
Forest and Mount Cameroon. However, despite considerable investment
from the Global Environment Facility and the UK’s Department for
International Development towards project development, CAMCOF
failed to get the necessary support and the Foundation has been closed.
Sustainable financing for conservation and sustainable use in the region
remain unresolved issues.
Monitoring and evaluation confirms project success For much of the

past 19 years there has been no formal monitoring scheme in place to
measure the conservation success of project interventions in the Bamenda
Highlands. An initial survey of the extent of the montane forest carried
out in 1983/84 noted the rapid rates at which montane forest in the region
was being cleared for cultivation, and concluded that without rapid
intervention the remaining Kilum-Ijim forest would be destroyed by the
year 2000. The fact that the forest remains is an important measure of
success, especially considering the continued loss and degradation of
other montane forests, including Forest Reserves, over the same period.
Monitoring of forest extent using satellite images and aerial photogra-
phy has shown the impact of the project’s interventions. They show that
between 1958 and 1988 (that is, pre-project) more than 50 per cent of the
forest was lost. This was followed by a period of forest regeneration and
in the period 1988–2001 the forest extended by 7.8 per cent of the 1988
area. This extension of forest cover has taken place largely through forest
regeneration on cleared areas within the boundary that was demarcated
in 1988.
In 1994 a conscious change in approach was made to ensure that the
success of the project in terms of conserving the key bird species in the
forest (a key conservation target) could be reported. A thorough analysis
of the conservation objectives helped to define what the project hoped
to achieve in terms of key species (especially endemics and threatened

species of plant and animal), habitats and ecosystem functions (Maisels,
1998). The monitoring system then evolved to comprise a comprehensive
monitoring system designed to operate on a number of scales (Maisels
and Forboseh, 1999) and combining point counts of birds, permanent
vegetation quadrats and occasional sightings of mammals with fixed point
photography and satellite image analysis. It aimed to combine species
monitoring with ‘functional’ monitoring (monitoring of keystone species,
such as pollinators (sunbirds, bees) and seed dispersers (frugivorous birds
and mammals). However, this level of monitoring is very labour intensive
(and therefore costly) and proved not to be sustainable when major-donor
funding of the project came to an end in 2004. However, initial results sug-
gested some stability in forest condition (although the effect of ‘lag’ is not
known). For example, survey data from the census of Bannerman’s turaco
from 1991, 1995–97 and 1999–2001 failed to detect significant changes
in the population of this ‘flagship’ species (Kilum-Ijim Forest Project,
2002).
It was recognized by the project that measurement of ‘biodiversity’
rather than ‘resources’ is mainly of interest to conservationists (be it
MINFOF, or national or international NGOs) rather than local commu-
nities. In order to address this the project also put in place a programme
of institutional monitoring and natural resource monitoring designed to
be implemented by FMIs. Based on discussions with forest users, these
participatory measures allow FMIs to measure the state of their institu-
tion (by asking questions on certain indicators such as ‘What percentage
of the population are members of the FMI’, ‘What proportion of FMI
members participate in collective forest management activities such as fire-
tracing?’ and ‘How representative is the FMI (number of women, men,
youth, elders, Fulani and so on)?)’. Another set of indicators are designed
to allow FMIs to assess and monitor the condition of the forest resources,
and the threats it faces, using measures that are relevant to them and their
livelihoods. This uses indicators such as ‘number of streams flowing in
the dry season’, ‘annual yield from bee hives’ and ‘number of seedlings of
selected species (mainly species of economic importance such as food trees
and those whose timber is used for carving) that are over knee height’.
This community-based monitoring is at an early stage, and its effectiveness
in directing management by FMIs has not yet been assessed.
The project has also undertaken specific (one-off) exercises in order to
measure both the uptake of technologies introduced by the project and
the attitudes of people towards the forest and its conservation. An uptake
survey in 1999 interviewed 950 farmers in villages around the forest to
determine their uptake of technologies promoted by the project through
its ‘livelihoods’ programme (such as contour ridging, tree production,

live fencing, erosion control and improved livestock feeding). This dem-
onstrated levels of outreach (that is, percentage of respondents who had
received information on the technology) ranging from 24 per cent (for
training in raising fruit trees) to 93 per cent (for training in methods of
controlled burning and alternative land preparation methods) and uptake
levels (percentage of farmers who had received training who then adopted
the technology) of between 49 per cent and 98 per cent (for uptake of con-
trolled burning) (Tsongwain, 1999; Kilum-Ijim Forest Project, 2002).
Assessment of changes in behaviour have been complicated by the
lack of any baseline statistics. However, using qualitative, participatory
methods and a comparative approach (comparing communities that
had little contact with the project, with those that had had high levels of
contact and uptake) has demonstrated the significant positive impact of
the project on people’s attitudes to the forest and its conservation (Abbot
et al., 1999, 2001).
Discussion
Looking at our two ICDP examples from the forests of Africa, a number
of issues emerge. First, it is clear that neither of these ICDPs is perfect,
even to those who have been involved with them over a period of years.
In this regard we accept some of the criticisms that the ICDP approach
does not deliver as much conservation or development as some originally
proposed it would. But, both of these projects are working in some of the
least developed regions of the world, where billions of dollars of develop-
ment assistance and the best economists in the world have largely failed to
improve national economic fortunes (UNDP, 2001). We therefore believe
that any conservation successes need to be set in the context of greater
economic failures.
Scale and ownership/benefits

In both of our examples, the projects operate over relatively small parts
of the landscape of biological importance, and focus on maintaining
critical patches of forest cover for their biodiversity values – principally
endemic species with no known economic importance. The values of the
biodiversity are uncosted. The local values for villagers living adjacent to
the forest habitats, in terms of direct cash generated from these forests or
their endemic species, are small. However, in both of our case examples
the local perception of the value of the forest in terms of providing a water
supply through the dry season is high. Numerous villagers in both regions
tell stories of the drying of their local stream when its catchment area
was deforested. The cultural value of the remaining forest is also high in
both of these areas, as places for the ancestors to live in. Moreover, in the
Uluguru Mountains of Tanzania the national cash values of the forests in
terms of maintaining water flow to the capital city and most of the nation’s
industry are huge. The appreciation of these values and the development
of systems that capture some of this value and use it for long-term man-
agement, offers one of the best ways to achieve sustainable management
of the forest resources and some assistance to the development of local
human populations.
Ownership
Both of our two example projects illustrate a gradual shift in emphasis from
forest ownership by the national government to a situation where local
people have a role in the management of government-owned reserves (for
example, in the Ulugurus), or more promisingly perhaps where they estab-
lish and manage Village Forest Reserves on their own land and where the
government, or a local NGO, is a facilitating and problem-solving agency.
The legal mechanisms that allow these community approaches to forest
ownership and conservation may provide a boost to forest conservation
efforts more widely across Africa in coming decades. An evolving issue in
development thinking involves attempts to ensure that issues of equity and
rights are considered fully in development projects. This novel thinking,
especially on rights, is likely to become an important aspect of ICDP design
and implementation, especially given the global re-emphasis on solving
some of the fundamental issues of poverty and inequality across the devel-
oping world, especially in Africa. One of the risks of this approach is that
the choices that local people make might not be ones that ‘conservationists’
want, but equally there is a potential for people to reject centralized govern-
ment models (for example, large-scale concessions, clearance for industrial
agriculture – oil palms and so on) and choose their own pathway.
Planning
In our opinion one of the most helpful mechanisms for designing and
managing an ICDP to ensure that it works towards its goals, and has the
chance to periodically review and alter these goals, has been the Logical
Framework Approach (or related planning systems). By building a set
of interventions around solving the environmental (and human develop-
ment) problems in an area, and monitoring these carefully over time, best
practice conservation management can result. However, even in cases
where log frames have been developed, many are poorly formulated and
do not attempt to measure conservation impact – instead they monitor
whether activities are progressing and funding is being used. The major-
ity of ICDPs in operation have this flaw, which is one reason why it has
been easy for them to be criticized – most cannot provide data to answer
what their conservation achievements have been. In this regard we would
endorse exploring ways to measure the pressure/threat to an area and
how it has changed, the state of the system (both biodiversity/habitat and
socioeconomic) and what responses occur (reserves declaration, attitude
changes, activity changes, degradation changes, governance changes and
so forth). If logical frameworks could be used to develop and help main-
tain monitoring programmes that capture information on these broad
areas, this would be a significant step forward. If a sufficiently large sample
of ICDPs were operating similar monitoring schemes then some statisti-
cally analysed conclusions about the success or failure of the ICDP model
might be developed.
Sustainability
Long-term sustainability has rarely been achieved by ICDPs, even those
that have operated for a decade or more. It may be that this model of
project cannot achieve a sustainable end point, especially when working in
developing countries where the economic situation is stagnant and there
are few funds from government or local leaders to support conservation
activities. Solving the macroeconomic problems of many African nations
would enhance the chances of long-term sustainability for projects operat-
ing in poor rural areas. In the absence of economic improvements, working
to establish mechanisms that ensure the supply of modest funding over the
long term, and the availability of reliable and trained staff, may offer the
best chances to reach a sustainable end point. If there are no funds then
good staff will move to other areas, or will start to engage in other activ-
ities to support themselves. If there is too much funding for a known short
period, then people may seek to capitalize on it while it is there.
Neither of our two example ICDPs are close to being economically
sustainable. The models used by these projects have involved significant
sums of money, expatriate support, purchase of transport and fuel, inter-
national travel and some foreign management costs. The local economy
of these areas involves small amounts of money, local decision-making, a
lack of mechanized transport, few income-generating opportunities and
traditional beliefs and norms. In order for the ICDP approach to become
more sustainable it has to become financially, technically and operation-
ally closer to the normality for the areas concerned, which is a serious
challenge using current project implementation models.
Protection components
We endorse the criticism that ICDPs do not expend as much effort as
required on natural resource conservation. Our experience with ICDPs
funded by development agencies shows how their focus on poverty allevia-

tion, while providing valuable funds, can force the ICDP to expend nearly
all its effort on development with only weak and insufficient components
targeting the protection of natural resource values. Part of the current
criticism that ICDPs are trying to achieve conservation using indirect
mechanisms comes from this mismatch of agendas – on the one hand the
funding agency wants to see conservation by development, whereas the
conservation agency wants to see conservation gain and will try to design
development interventions to achieve this goal. However, there need not
always be this dichotomy, and there are many situations where conserva-
tion and development agendas directly overlap, especially in poor rural
communities in Africa.
Implementation constraints
Our two example ICDPs, and especially the Uluguru Mountains ICDP
managed by CARE have been dogged by implementation constraints.10
This has affected the efficiency of the projects, and has resulted in some
significant time delays and failures to achieve some of the planned targets.
This issue has already been explored in some detail in the review by Wells
and McShane (2004) where they note that even a well-designed project
can fail because of problems with its implementation strategy – including
problems of obtaining suitable staff, and so on.
Learning components
One general observation is that ICDPs would benefit from a more scien-
tific approach, including designing interventions to test hypotheses about
different kinds of interventions. The learning component of many ICDPs
is generally small and often given a low priority. However, this means that
ICDPs will continue to fail to show what they have achieved, or indicate
what interventions give the best value for money.
Local people can collect much of the data required to monitor conser-
vation impact. This can both simplify the methods, allow local under-
standing of what is required and what the project is trying to achieve, and
also keep costs low (Danielsen et al., 2001). However, as with many other
elements of ICDPs, the data collected by local investigators will need peri-
odic field validation to ensure that standards are maintained. These locally
based methods of monitoring can tell a lot about resource use and man-
agement actions, but they might not satisfy the requirements of conserva-
tion biologists interested in the number of particular species or details of
ecological services. In these cases more professionally based monitoring
may be required, involving trained scientists.
In addition to measuring biological and threat-related measures within
an ICDP, measuring perceptions and attitudes of the local population

over time can also indicate conservation trends. Local populations are
often reported to have a negative perception of protected areas. However,
our Bamenda Highlands case study provides evidence that ICDPs can
change these negative perceptions over time. Such changes go a long way
to making the interventions sustainable in the long term.
Other management options

Given the long-term funding input to our two example ICDPs, and the
relatively low level of impact, it is relevant to consider whether other
forms of conservation might have been more successful. In many parts of
the world National Parks or Strict Nature Reserves are regarded as the
most secure means for conserving wild areas. Would these have worked
in the case of the Ulugurus or the Bamenda Highlands? The possible sce-
narios for these two sites are highly different. For the Ulugurus, most of
the forest remains in two large central government Forest Reserves and
the boundaries are fairly well marked. As the national government already
controls the forest area, then changing the Forest Reserves into a National
Park would only involve a transfer of control within the government, from
the Forest Department to Tanzania National Parks. An alternative that
is now being pursued in Tanzania is to upgrade the status of these Forest
Reserves to a Nature Forest Reserve, which would make the reserves more
secure but keep them under the authority of the same part of government.
For Kilum-Ijim in the Bamenda Highlands, any protected area would be
strongly resisted by local people – for example, previous attempts to des-
ignate a Forest Reserve failed. The creation of a National Park or other
form of government reserve would be highly problematic, strongly resisted
by the local people, and difficult to implement in a meaningful way.
Other models of conservation management that have been advocated
in recent years include direct land purchase, purchase of concessions for
conservation purposes and direct payments to communities. Direct land
purchase and concession purchase models cannot easily work in these two
project areas, although legal changes are making this more possible (see
Wily and Mbaya, 2001 for Tanzania). Current ideas about making direct
payments to communities in return for specific conservation targets (area
of forest, population of a species, number of annual fires and so on) may
help to focus attention on making conservation interventions as direct as
possible (Ferraro and Kiss, 2002). A relevant question is whether if all the
funds placed in ICDPs had instead been used for direct payments, would
the communities have been better off? The total funds used in the Uluguru
Mountain ICDPs since 1993 are around US$2.5m. If all that funding
had been available in Tanzania at the project site it would represent less
than US$2 per locally based person (assuming around 100 000 people
on the mountain) per annum over eight years. Although this is not a
major financial inducement, it is a more tangible sum than most received
from the project activities. However, we predict that the direct payments
model would be difficult to implement in poor rural communities. The
mechanism of delivering funds would need to be carefully worked out and
monitoring this would require considerable effort – and would therefore
rapidly become expensive. Funds would also need to be available over
the long term if this was the only mechanism employed to conserve the
forests. The direct payment approaches may work best linked to a trust
fund mechanism that provides small funding to local communities over
long periods.
In conclusion, the ICDP model has certainly been challenged in recent
years but we believe that the jury is still out and nothing is yet in place/
proposed that can realistically (proven to) replace it. Few would argue
that it makes sense (moral, practical) to conserve areas that are rich
in biodiversity without involving the people that are living in the area.
Falling back totally on the strict ‘protection’ approach with the exclusion
and marginalization it implies is not a direction most conservationists in
Africa want to take. The integration of conservation at high biodiversity
sites with alleviation of social and economic problems is a goal that has
popular support, but the challenge is to find ways of making it work,
to learn from the successes (and failures) of existing projects and not to
dismiss the approach without proposing viable alternatives backed by
detailed scientific analysis.
Notes
1. BirdLife International managed the Kilum-Ijim Forest Project and Bamenda Highlands
Forest Project in collaboration with the Cameroon Ministry of the Environment and
Forestry and forest-adjacent communities. The projects received funding from the
Global Environment Facility (GEF/UNDP), the UK Community Fund (National
Lottery Charities Board), the British Department for International Development
(DFID) through the Civil Society Challenge Fund, the Joint Funding Scheme
(DFID), the Global Environment Facility (GEF-World Bank) through the Cameroon
Biodiversity Conservation and Management Programme, the Dutch Ministry of
Agriculture, Nature Management and Fisheries through the Programme International
Nature Management (PIN) and WWF Netherlands. Current work at Kilum-Ijim is
funded by the JJ Charitable Trust and the Dutch Ministry of Foreign Affairs.
Conservation Projects in the Uluguru Mountains landscape have been funded
by the European Union, the Royal Society for the Protection of Birds (RSPB), the
Government of Tanzania, Danish International Development Agency (DANIDA),
Dansk Ornitologisk Forening (DOF) and the Wildlife Conservation Society of
Tanzania (WCST), Christian Aid, the Spanish NGO INTERMON and will soon
receive funds from the GEF. We would like to the thank the following Tanzanian staff
from the various parts of government and projects working around the Ulugurus for
their support and assistance: Lameck Noah, Cheyo Mayuma, Elisha Mazengo, Elisa
Pallangyo, Sia Aroko, Phanuel Bwimba, John Mejissa, Eliakim Enos, Ernest Moshi,
Prof. Amon Mattee, Athman Mgumia. The following also collected data in the field in
Tanzania, some of which is used here: Andrew Perkin, Tom Romdal, Marcel Rahner,
Anders Tøttrup, Klaus Tølbøl Sørensen and Olivier Hymas.
2. See http://www.care.org/careswork/projects/TZAO38.asp, accessed 11 August 2009.
3. See http://www.wikipedia.org/wiki/logical_framework_approach, accessed 11 August
2009.
4. See www.ecosystemmarketplace.com, accessed 12 August 2009.
5. See www.africanconservation.com/uluguru, accessed 12 August 2009.
6. See www.easternarc.or.tz for details of the Nature Reserve Process, accessed 12 August
2009.
7. See www.easternarc.or.tz, accessed 12 August 2009.
8. Formerly the Ministry of Environment and Forestry – MINEF.
9. A recent (2003) independent mid-term review of the project wrote that ‘Regarding
general awareness and support for natural resource conservation the work of the past
has paid off. The Northwest Province is without doubt the most conservation minded
in the entire land, and past efforts by the Kilum-Ijim Forest Project and more recently
the Bamenda Highlands Forest Project have contributed immensely to this achieve-
ment through its hard work in the field of Environmental Education. This has been so
successful, that the output may be considered achieved at this point in time. An impres-
sive list of Environmental Education material has been developed and disseminated,
ranging from e.g. 22 000 posters, courses for 60 biology teachers, to 32 newspaper
articles, 25 radio programmes, etc. The conservation constituency is so overwhelmingly
present at all levels of North-western society, that it would not be irresponsible for the
Project to consider winding down its activities in this domain, unless in direct support
of achievement of other outputs, and reallocate the available resources in time, money
and staff to the implementation of other activities’ (Bamenda Highlands Forest Project,
2003, independent mid-term review, BHFP, Cameroon).
10. See progress reports on www.easternarc.or.tz, accessed 13 August 2009.
References
Abbot, J., S.E. Neba and M.W. Khen (1999), ‘Turning our eyes from the forest: the role of
the livelihoods programme at Kilum-Ijim Forest Project, Cameroon in changing attitudes
and behaviour towards forest use and conservation’, Cambridge, UK: unpublished report
to BirdLife International.
Abbot, J.I.O., D.H.L. Thomas, A.A. Gardner, S.A. Neba and M.W. Khen (2001),
‘Understanding the links between conservation and development in the Bamenda
Highlands, Cameroon’, World Development, 29(7), 1115–36.
Adams, W.M. and D. Hulme (2001), ‘If community conservation is the answer in Africa,
what is the question?’, Oryx, 35(3), 193–200.
Alpert, P. (1996), ‘Integrated Conservation and Development Projects: examples from
Africa’, BioScience, 46(11), 845–55.
Anderson, D. and R. Grove (eds) (1987), Conservation in Africa: Peoples, Policies and
Practice, Cambridge, UK: Cambridge University Press.
Ashley, C. and D. Roe (1998), Enhancing Community Involvement in Wildlife Tourism: Issues
and Challenges, London: IIED Wildlife and Development Series No. 11.
Attwell, C.A.M. and F.P.D. Cotterill (2000), ‘Postmodernism and African conservation
science’, Biodiversity and Conservation, 9(5), 559–77.
Barrett, C.B. and P. Arcese (1995), ‘Are Integrated Conservation-Development Projects
(ICDPs) sustainable? On the conservation of large mammals in sub-Saharan Africa’,
World Development, 23(7), 1073–84.
Bhatia, Z. and P. Buckley (1998), ‘The Uluguru Slopes Planning Project: promoting com-
munity involvement in biodiversity conservation’, Journal of East African Natural History
Society, 87(1), 339–49.
Bhatia, Z. and O. Ringia (1996), Socio-economic Survey of Selected Villages in the Uluguru
Mountains, Tanzania. Uluguru Slopes Planning Project Report No. 3, Sandy, UK: Royal
Society for the Protection of Birds.
Borrini-Feyerabend, G. and D. Buchan (1997), Beyond Fences. Seeking Social Sustainability
in Conservation. Volume 1: A Process Companion, Gland, Switzerland: IUCN.
Brandon, K. and M. Wells (1992), ‘Planning for people and parks: design dilemmas’, World
Development, 20(4), 557–70.
Brown, M. and B. Wyckoff-Baird (1993), Designing Integrated Conservation and Development
Projects, Washington DC: Biodiversity Support Program.
Bruner, A., R.E. Gullison, R.E. Rice and G.A.B. da Fonseca (2000), ‘Effectiveness of parks
in protecting tropical biodiversity’, Science, 291(5501), 125–8.
Burgess, N.D. and G.P. Clarke (eds) (2000), The Coastal Forests of Eastern Africa, Gland,
Switzerland and Cambridge, UK: IUCN Forest Conservation Programme.
Burgess, N.D., M. Nummelin, J. Fjeldså, K.M. Howell, K. Lukumbuzya, L. Mhando,
P. Phillipson and E. Vanden Berghe (eds) (1998), ‘Biodiversity and conservation of the
Eastern Arc Mountains of Tanzania and Kenya’, Special Issue: Journal of the East African
Natural History Society, 87(1–2), 1–367.
Burgess, N.D., N. Doggart and J.C. Lovett (2002a), ‘The Uluguru Mountains of eastern
Tanzania: the effect of forest loss on biodiversity’, Oryx, 36(2), 140–52.
Burgess, N.D., T. Lehmberg, C. Loucks, J. D’Amico and J. Morrison (2002b), ‘Some aspects
of designing conservation landscapes: with an example from eastern Africa’, in P. Schultz
and D. Noppen (eds), Integrating Conservation and Development with a Landscape
Approach, Copenhagen, Denmark: Environment and Development Network.
Caldecott, J. (1998), Designing Conservation Projects, Cambridge, UK: Cambridge University
Press.
Carney, D. (1999), ODI Poverty Brief: Approaches to Sustainable Livelihoods, London:
Overseas Development Institute.
Chambers, R. (1997), Whose Reality Counts? Putting the First Last, London: Intermediate
Technology Publications.
Danielsen, F., D.S. Balete, M.K. Poulsen, M. Enghoff, C.M. Nozawa and A.E. Jensen (2001),
‘A simple system for monitoring biodiversity in protected areas of a developing country’,
in H. Agersnap and M. Funder (eds), Conservation and Development: New Insights
and Lessons Learnt, Copenhagen: Environment and Development Network, pp. 231–
72.
Doggart, N., J. Lovett, B. Mhoro, J. Kiure and N.D. Burgess (2005), ‘Biodiversity surveys in
eleven Forest Reserves in the vicinity of the Uluguru Mountains, eastern Tanzania’, Dar
es Salaam, Tanzania: Wildlife Conservation Society of Tanzania, and Tanzania Forest
Conservation Group, www.easternarc.or.tz, accessed 12 August 2009.
Dubois, O. and J. Lowore (2000), The Journey towards Collaborative Forest Management in
Africa: Lessons Learned and Some Navigational Aids, London: IIED, Forestry and Land
Use Series No. 15.
Ferraro, P.J. and A. Kiss (2002), ‘Direct payments to conserve biodiversity’, Science,
298(5599), 1718.
Fisher, R.J. (1995), Collaborative Management of Forests for Conservation and Development.
Issues in Forest Conservation Series. Gland, Switzerland: IUCN and WWF.
Franks, P. (2001), ‘Poverty and environmental degradation in the context of Integrated
Conservation and Development Projects: what makes an ICDP integrated?’, in
H. Agersnap and M. Funder (eds), Conservation and Development: New Insights and
Lessons Learnt, Copenhagen: Environment and Development Network, pp. 95–104.
Franks, P. and T. Blomley (2004), ‘Fitting ICD into a project framework: A CARE per-
spective’, in T.O. McShane and M.P. Wells (eds), Getting Biodiversity Projects to Work:
Towards More Effective Conservation and Development, New York: Columbia University,
pp. 77–97.
Ghimire, K.B. and M. Pimbert (eds) (1997), Social Change and Conservation, London:
Earthscan.
Govt. of Tanzania (GOT) (1998), Tanzanian Forest Policy, Dar es Salaam, Tanzania:
Division of Forestry and Beekeeping.
Govt. of Tanzania (GOT) (2001), Community-based Forest Management Guidelines, Dar es
Salaam, Tanzania: Division of Forestry and Beekeeping.
Govt. of Tanzania (GOT) (2002), Forest Act, Dar es Salaam, Tanzania: Division of Forestry
and Beekeeping.
Grove, R. (1995), Green Imperialism, Cambridge, UK: Cambridge University Press.
Hackel, J.D. (1999), ‘Community conservation and the future of Africa’s wildlife’,
Hannah, L. (1992), African People, African Parks: An Evaluation of Development Initiatives
as a Means of Improving Protected Area Conservation in Africa, Washington DC:
Conservation International.
Hartley, D. and S. Kaare (2001), ‘Institutional, policy and livelihoods analysis of communi-
ties adjacent to the Uluguru Mountains Catchment Reserves, Eastern Arc Mountains’,
Dar es Salaam, Tanzania: CARE Tanzania, www.africanconservation.com/uluguru,
accessed 12 August 2009.
Hughes, R. and F. Flintan (2001), Integrated Conservation and Development Experience:
A Review and Bibliography of the ICDP Literature, London: IIED, Biodiversity and
Livelihoods Issues Series No. 3.
Hulme, D. and M. Murphree (1999), ‘Communities, wildlife and the “new conservation” in
Africa’, Journal of International Development, 11(2), 277–86.
Hymas, O. (1999), ‘Bananas in the hills’, MSc dissertation, University College London,
www.africanconservation.com/uluguru, accessed 12 August 2009.
Hymas, O. (2000), ‘Assessment of the remaining forests on the Uluguru Mountains and the
pressures they face, Morogoro, Tanzania: Uluguru Mountains Biodiversity Conservation
Project’, www.africanconservation.com/uluguru, accessed 12 August 2009.
Hymas, O. (2001), ‘People on the hills: an assessment of forest disturbance of the Ulugurus
in relation to access, Morogoro, Tanzania: Uluguru Mountains Biodiversity Conservation
Project’, www.africanconservation.com/uluguru, accessed 12 August 2009.
IIED (1994), Whose Eden? An Overview of Community Approaches to Wildlife Management,
Nottingham, UK: Russell Press.
Jeanrenaud, S. (2002a), ‘Changing people/nature representations in international conserva-
tion projects’, IDCS Bulletin, 33(11), 111–22.
Jeanrenaud, S. (2002b), People-orientated Approaches in Global Conservation: Is the Leopard
Changing its Spots?, London: IIED, Institutionalizing Participation Series.
Kilum-Ijim Forest Project (2002), Report from the Ecological Monitoring Programme,
July 2001–October 2002, Cameroon: Kilum-Ijim Forest Project, BirdLife International/
MINEF.
Kleiman, D.G., R.P. Reading, B.J. Miller, T.W. Clark, J.M. Scott, J. Robinson, M.L.
Wallace, R.J. Cabin and F. Felleman (2000), ‘Improving the evaluation of conservation
programmes’, Conservation Biology, 14(2), 356–65.
Koziell, I. (2000), Diversity not Adversity: Sustaining Livelihoods with Biodiversity, London:
IIED.
Kramer, R., C. van Schaik and J. Johnson (eds) (1997), Last Stand. Protected Areas and
Defense of Tropical Biodiversity, New York and Oxford, UK: Oxford University Press.
Kremen, C., A.M. Merenlender and D.D. Murphy (1994), ‘Ecological monitoring: a vital
need for Integrated Conservation and Development Programs in the tropics’, Conservation
Biology, 8(2), 388–97.
Larson, P.S., M. Freudenberger and B. Wyckoff-Baird (1998), WWF Integrated Conservation
and Development Project: Ten Lessons from the Field 1985–1996, Washington DC: WWF.
Leach, M., J. Fairhead and K. Amanor (eds) (2002), ‘Science and the policy process: perspec-
tives from the forest’, IDS Bulletin, 33(1), 1–126.
Leader-Williams, N., J.A. Kayera and G.L. Overton (1996), Community-based Conservation
in Tanzania, Gland, Switzerland and Cambridge, UK: IUCN.
Lovett, J.C. and T. Pócs (1993), Assessment of the Condition of the Catchment Forest
Reserves, a Botanical Appraisal, Dar es Salaam, Tanzania: Catchment Forestry Report

No. 93.3, Forest and Beekeeping Division, Ministry of Tourism, Natural Resources and
Environment.
Lovett, J.C. and S.K. Wasser (eds) (1993), Biogeography and Ecology of the Rain Forests of
Eastern Africa, Cambridge, UK: Cambridge University Press.
Lukumbuzya, K. (2000), ‘The effects of different management regimes on the diversity and
regeneration of woody vegetation under a Joint Forest Management model in Tanzania’,
in P. Virtanen and M. Nummelin (eds), Forests, Chiefs and Peasants in Africa: local man-
agement of natural resources in Tanzania, Zimbabwe and Mozambique, Silva Carelica, 34,
University of Joensuu, Finland, pp. 159–77.
Lyamuya, V.E., L.G. Noah, E.J., Kilasara and N.D. Burgess (1994), Socio-economic and
Land Use Factors Affecting the Degradation of the Uluguru Mountains Catchment in
Morogoro Region, Tanzania, Morogoro, Tanzania: Regional Natural Resources Office.
MacKinnon, K. (2001), ‘Integrated Conservation and Development Projects – can they
work?’, Parks, 11(2), 1–4.
Maisels, F. (ed.) (1998), Conservation Objectives of the Kilum-Ijim Forest, Ecological
Monitoring Programme, Kilum-Ijim Forest Project, BirdLife International/MINEF.
Maisels, F. and P. Forboseh (1999), Methods Manual and Biodiversity List. First Draft,
Cameroon: Kilum-Ijim Forest Project, BirdLife International/MINEF.
Margolius, R. and N. Salafsky (1998), Measures of Success: Designing, Managing and
Monitoring Conservation and Development Projects, Washington DC: Island Press.
Neumann, R.P. (1996), ‘Dukes, earls, and ersatz edens: aristocratic nature preservationists in
colonial Africa’, Environment and Planning, 14(1), 79–98.
Neumann, R.P. (1998), Imposing Wilderness: Struggles over Livelihood and Nature
Preservation in Africa, Berkeley, CA: University of California Press.
Newmark, W.D. and J.L. Hough (2000), ‘Conserving wildlife in Africa: Integrated
Conservation and Development Projects and beyond’, BioScience, 50(7), 585–92.
Oates, J. (1995), ‘Challenges to nature conservation with community development – a case
study from the forests of Nigeria’, Oryx, 29(2), 115–22.
Oates, J.F. (1999), Myth and Reality in the Rain Forest: How Conservation Strategies are
Failing in West Africa, Berkeley, USA: University of California Press.
Pimbert, M.C. and J.M. Pretty (1995), ‘Parks, people and professionals: putting “par-
ticipation” into protected area management’, Geneva: UNRISD Discussion Paper No.
DP57.
Pócs, T. (1974), ‘Bioclimatic studies in the Uluguru Mountains (Tanzania, East Africa) I’,
Acta Botanica Academiae Scientarium Hungaricae, 20(1–2), 115–35.
Pócs, T. (1976), ‘Bioclimatic studies in the Uluguru Mountains (Tanzania, East Africa)
II. Correlations between orography, climate and vegetation’, Acta Botanica Academiae
Scientarium Hungaricae, 22(1–2), 163–83.
Posey, D.A. (1999), Cultural and Spiritual Values of Biodiversity, London: UNEP and
Intermediate Technology Publications.
Rodgers, W.A. (1993), ‘The conservation of the forest resources of eastern Africa: past influ-
ences, present practices and future needs’, in J.C. Lovett and S. Wasser (eds), Biogeography
and Ecology of the Rain Forests of Eastern Africa, Cambridge, UK: Cambridge University
Press, pp. 283–331.
Roe, D. and M. Jack (2001), Stories from Eden: Case-studies of Community-based Wildlife
Management, Evaluating Eden Series No. 9, London: IEED.
Salafsky, N. and R. Margolius (1999), ‘Threat reduction assessment: a practical and cost-
effective approach to evaluating conservation and development projects’, Conservation
Biology, 13(4), 830–41.
Sanjayan, M.A., S. Shen and M. Jansen (1997), ‘Experiences with conservation-development
projects in Asia’, Washington, DC: World Bank Technical Paper No. 388, The World
Bank.
Schrijver, N. (1997), Sovereignty over Natural Resources: Balancing Rights and Duties,
Cambridge, UK: Cambridge University Press.
Songorwa, A. (1999), ‘Community-based wildlife management (CWM) in Tanzania: are the

communities interested?’, World Development, 27(12), 2061–79.
Songorwa, A.N., T. Buhrs and K.F.D. Hughey (2000), ‘Community-based wildlife manage-
ment in Africa: a critical assessment of the literature’, Natural Resources Journal, 40(3),
603–43.
Spinage, C.A. (1996), ‘The rule of law and African game – a review of some recent trends and
concerns’, Oryx, 30(3), 178–86.
Spinage, C.A. (1998), ‘Social change and conservation misrepresentation in Africa’, Oryx,
32(4), 265–76.
Stocking, M. and S. Perkin (1992), ‘Conservation-with-development; an application of the
concept in the Usambara Mountains, Tanzania’, Transactions of the Institute of British
Geographers, 17(3), 337–49.
Svendsen, J.O. and L.A. Hansen (eds) (1995), Report on the Uluguru Biodiversity Survey
1993. Parts A and B, Morogoro, Tanzania: Tanzania Forestry Research Institute.
Tsongwain, D.V. (1999), ‘Conservation and development: agricultural intensification and the
uptake of project initiated alternative livelihood activities around the Kilum-Ijim Forest
(Cameroon)’, unpublished MSc dissertation, Department of Geography, University of
Cambridge, UK.
UNDP (2001), World Development Report, New York: UNDP.
Wainwright, C. and W. Wehrmeyer (1998), ‘Success in integrating conservation and develop-
ment? A study from Zambia’, World Development, 26(6), 933–44.
Wells, M. and K. Brandon (1992), ‘People and parks – linking protected area management
with local communities’, Washington, DC: The World Bank, WWF, USAID.
Wells, M. and T.O. McShane (2004), ‘Integrating protected area management with local
needs and aspirations’, Ambio, 33(8), 513–519.
Wells, M., S. Guggenheim, A. Khan, W. Wardojo and P. Jepson (1999), Investing in
Biodiversity: A Review of Indonesia’s Integrated Conservation and Development Projects,
Washington, DC: The World Bank.
Wily, L.A. and S. Mbaya (2001), Land, People and Forests in Eastern and Southern Africa at
the Beginning of the 21st Century: The Impact of Land Relations on the Role of Communities
in Forest Future, Nairobi, Kenya: IUCN Eastern Africa Programme Office.
World Summit on Sustainable Development (WSSD) (2002), www.un.org/jsummit/html
basic_info/basicinfo.html, accessed 11 August 2009.
4. Biodiversity conservation in managed
landscapes
Tom M. van Rensburg and Greig A. Mill
Introduction
Biodiversity conservation has emerged as one of the most important and
controversial global environmental issues in recent years (UNEP, 1995).
First, it has been suggested that we are on the verge of mass extinctions,
the like of which have not been observed in the fossil record (Wilson,
1985). Second, it is argued that biodiversity loss matters because it is of
fundamental importance to human society. It provides food, shelter, fuel,
supports recreation and tourism and is thought to play an important part
in global life support and in the functioning of ecosystems (Lindberg,
1991; Raven et al., 1992; Brown et al., 1994).
A decline in habitat is thought to be one of the most significant causes
of the loss in terrestrial biodiversity (Wilson, 1985). A large proportion
of the earth’s fertile land has been converted into managed agricultural,
forest and urban landscape. Recent estimates by the FAO (2004) indicate
that some 38 per cent of land globally is now utilized for agriculture. One
solution proposed by ecologists is to expand reserves and protected areas.
However, there are a number of problems with this approach: protected
areas cover a limited area – approximately 11 per cent of the earth’s
surface (WRI, 2005); protected areas generally exclude economic activities
and they can impose costs on land managers and prevent future economic
opportunities from taking place. Consequently, it is unlikely that the pro-
portion of land allocated to protected areas will be sufficient to maintain
all biodiversity.
In recent years a number of studies indicate that biodiversity conserva-
tion must focus on managed human-dominated ecosystems (Miller, 1996;
Reid, 1996; Daily et al., 2001; Rosenzweig, 2003; Polasky et al., 2005).
Economically valuable managed landscapes do not necessarily have to
exclude biodiversity conservation goals. A wide range of species occur in
the presence of human activities and much of the world’s biodiversity is
found in human-dominated ecosystems (Pimmental et al., 1992). Instead
of threatening biodiversity, many managed systems may actually enhance
biodiversity because of, rather than in spite of, the day-to-day manage-
ment activities carried out by land managers. Indeed, the phenomenon
75
of land abandonment has become a subject of major concern in many

countries because it results in the loss of biodiversity. It is also thought
that land managers may conserve biodiversity because it supports the
productivity and resilience of the ecosystems they manage and there is
now a significant literature on the functional and ecosystem service values
associated with biodiversity (Ellis and Fisher, 1987; Daily, 1997; Barbier,
2000; Daily et al., 2000).
However, the intensity and mode of disturbance clearly play an impor-
tant part in the management of ecosystems. Heavily disturbed agricultural
and forest ecosystems in many parts of the world are threatened by human
intervention that has resulted in a loss of biodiversity and resilience.
Clearly it is important to distinguish between managed landscapes that
undergo disturbance in which biodiversity appears to be thriving and
those in which it is threatened. In this chapter we explore the relationship
between human-induced disturbance and biodiversity. We also consider
the relationship between disturbance regimes and the properties of a
managed ecosystem – its productivity, resilience and stability.
All too frequently the application of ‘good science’ is not in itself
sufficient to guarantee desirable environmental outcomes with respect
to biodiversity conservation. This also requires an understanding of
socioeconomic and policy considerations: how markets allocate scarce
resources, how they influence decisions taken by land managers and why
they frequently fail to protect biodiversity and other non-market values.
To a degree, most land managers are affected by the market. Thus, the
chapter also explores the economic linkages between markets and natural
resources, and the possibilities of exploiting the powerful creative forces
of the market in a manner that conserves biodiversity whilst also provid-
ing useful marketable goods and services. Finally the chapter concludes
with some policy recommendations on making markets account for the
benefits of biodiversity conservation in managed landscapes. We begin by
distinguishing between genetic and functional diversity.
Genetic and functional diversity in managed landscapes

Several definitions have been proposed to capture the multifaceted nature
of biodiversity (ecosystems, species and genes). This is acknowledged in
the definition developed in the Convention on Biodiversity as follows:
‘“Biological diversity” means the variability among living organisms from
all sources, including inter alia, terrestrial, marine and other aquatic eco-
systems and the ecological complexes of which they are part; this includes
diversity within species, between species and among ecosystems’ (UNEP,
1995. p. 8).1
A distinction is made in the literature between genetic diversity and
Biodiversity conservation in managed landscapes 77
functional diversity. Genetic diversity usually refers to the genetic variation

that exists within a species (the gene pool). Genes are the fundamental unit
of biodiversity and the ultimate source of all variation among all animal
and plant species (Dobzhansky, 1970; Soulé and Wilcox, 1980). Genetic
diversity has been proposed as the basis on which to make conservation
decisions using the evolutionary distinctiveness of taxa when assigning
them priorities for preservation (Vane-Wright et al., 1991; Crozier, 1992;
Solow et al., 1993; Weitzman, 1998). Here, the relative ecological value is
based on how far away species are from one another genetically and an
objective value is assigned to the taxonomic distinctiveness or degree of
‘independent evolutionary history’ (IEH) that is vested in a species (Vane-
Wright et al., 1991).
Conservation organizations frequently employ descriptions and meas-
ures of ecosystem diversity based on genetic diversity and they tend to
place great emphasis on species and their populations (IUCN, 1988).
Although considerable sums have been allocated towards species preser-
vation, there is frequently a bias towards charismatic species, large birds
or mammals that are very familiar to the public. Conservation groups and
professional conservationists often exploit certain species and ecosystems
to further their own conservation goals. Conservationists have called
the charismatic species that win the hearts of the general public ‘flagship
species’. This may be at the expense of less well-known species that may be
critical for the functioning of ecosystems (Metrick and Weitzman, 1994).
Understanding the value of biodiversity to land managers requires a
different perspective that is linked to the functional value of biodiversity.
How then does functional diversity differ from genetic diversity and if
so, why do these differences matter to managers? First, genes are after
all just chemicals that have no value in and of themselves. Instead, genes
have value in what they do – control the structure and function of life –
instead of in what they are. Measures of genetic distance may not capture
the relative values of species such as the complex functional relationships
embodied in ecosystems. Two species might be very similar with respect to
genetic distance but they may perform very different functions within the
ecosystem. One might be a keystone species that is vital to the well-being
of the managed ecosystem whilst the other is ‘functionally redundant’.
Species diversity is relevant to land managers because some species appear
to play a more important functional role than others. An individual who is
evaluating a species in terms of its functional role would be more sensitive
to a change in the ecosystem’s productivity than would a person focusing
on biodiversity. A person assessing an ecosystem from an ecosystem func-
tion perspective would be more likely to focus on key species and processes
and might overlook the disappearance of a rare species. There is greater
emphasis on the biological integrity of the system than simply ensuring

that all the biotic elements are present.
Functional diversity refers to the characteristics of ecosystems and
includes ecosystem complexity at different levels of organization such as
trophic levels (Cousins, 1991). This approach uses trophic-level analysis
to relate species diversity to functional ecosystem parameters such as food
web structure or the transfer of energy, water and chemicals between dif-
ferent trophic levels. Functional diversity can be interpreted as the number
of species required for a given ecological process.
In managed ecosystems, although every species may have a particular
role, it does not follow that these roles are of equal importance. Ecologists
acknowledge that some species have a greater ecological impact than
one might expect from their abundance or biomass and these have been
referred to as keystone species (Power et al., 1996). Some ecosystem
studies indicate that only a small number of the numerous species found
in ecosystems perform key functions or so-called keystone roles and that
most species perform a perfunctory role (Holling, 1992). For example,
beavers have been shown to have a profound impact on streams, forests
and wetlands through dam construction.
Many species may play keystone roles that involve interdependencies
with other species (Daily et al., 1993). The elimination of any single com-
ponent of an ecosystem could lead to an unanticipated unravelling of com-
munity structure and to local extinctions of some species. Bird dispersers
such as the blackcap (Sylvia atricapilla) interact with shrub species to
influence floristic diversity, spatial patterns of vegetation development and
plant dynamics in patchy Mediterranean vegetation in southern France
(Debussche and Isenmann, 1994). Seeds deposited by blackcaps under-
neath pioneer shrubs (which have positive ‘nurse’ effects on other plants)
were more likely to survive than in the open field and birds may actually
trigger dynamic successional processes initiated by pioneer woody peren-
nials in Mediterranean grasslands and shrublands.
The ‘keystone role’ of a species may also depend on whether a number
of other species can assume its functional role within the ecosystem
(Schindler et al., 1989). Functional redundancy is known to occur if
other species can perform similar roles (Hutchinson, 1961: Walker, 1992).
Although there is evidence that the deletion of some species has very little
effect on ecosystem functioning, in many diverse tropical forests there are
so many rare species that collectively they may have an important impact
on the ecosystem. Here, top predators in ecosystems are relatively scarce
because they are large in size and at the top of the food chain but may
nevertheless be important in terms of ecosystem structure.
Most research has focused on which species are important here and now.
However, millions of species have not even been identified let alone evalu-
ated for their potential values to humans. There are difficulties in predict-
ing which species will be important in the future since the present functions
performed by a species may provide no clues as to its role when environ-
mental conditions change (Main, 1982; Lovejoy, 1988). Tree species that
colonize gaps in tropical forests and species that require fire to enhance
germination provide such examples. The population of Cactoblastis
cactorum, which is relatively rare in Australia today, would not provide
an accurate description of its importance in controlling Opuntia in that
country in previous decades. However, studies on ecosystem function may
reveal clues as to the most sensitive components of food webs, and nutri-
ent and energy flows. Research reveals that the most sensitive components
of ecosystems are those in which the number of species performing a
particular function is thought to be very small (Schindler, 1990).
Most ecologists recognize that some species play a more important
functional role than others. But what does this imply in terms of the prop-
erties of an ecosystem? Land managers are principally concerned with
ecosystem productivity and its variability. The next section provides a
review of the relationship between biodiversity and the stability, resilience
and productivity of ecosystems.
Importance of biodiversity in managed landscapes

The importance of biodiversity is associated with a controversial theoreti-
cal debate amongst ecologists that began in the early 1950s: does biodiver-
sity affect the stability of an ecosystem? Holling (1973) refers to stability
as a characteristic of the individual populations of an ecosystem. For
example, stability is defined as the propensity of a population to return to
some kind of equilibrium following a disturbance. The stability of ecosys-
tems may be linked to their biodiversity and it has long been hypothesized
that more diverse ecosystems are more stable. A clue as to why this may
be the case is illustrated by a natural disturbance that deleted some species
from the ecosystem. A diverse system might be little affected by the impact
because other species with similar niches could perform similar functions
to the missing species. Early advocates of this theory include MacArthur
(1955) who postulated that a highly diverse ecosystem would change less
upon the removal or addition of a species than would an ecosystem with
fewer species. Elton (1958) also suggested that less diversity resulted in less
ecological stability.
However these theories were not without their critics. May (1973)
challenged this argument and showed that a highly connected system
(higher biodiversity) may be less stable than simpler ones and more vul-
nerable to disturbance because all its components closely interact and are
therefore subject to the effects of perturbations. A drought that eliminates

key species in a complex ecosystem, for example, will have widespread
repercussions on the animals that depend on them.
More recent work (Tilman, 1996) has shown that there exists an impor-
tant distinction between the properties of a community and its individual
species, so although diverse ecosystems are more stable than less diverse
ecosystems the populations within them can have great variability. From
this perspective, what matters is the stability of the community or eco-
system, not their individual populations. There is some experimental
evidence to support these assertions. Tilman and Downing (1994) have
shown that an ecosystem with many species is more likely to be stable even
though the populations of individual species may experience considerable
fluctuations.
Resilience is a further factor that refers to properties of the stability of
a system. The traditional concept of resilience is a measure of the speed of
return to an equilibrium state after an ecosystem has been disturbed (Pimm,
1984; O’Neill et al., 1986). Alternative definitions have been proposed by
Holling (1973). He describes resilience as the propensity of an ecosystem to
retain its functional and organizational structure following a disturbance.
Expressed another way, resilience is the amount of disturbance that can
be absorbed before the system changes its structure by changing the vari-
ables that control how the system behaves (Holling, 1973). A characteristic
feature of ‘Holling-resilience’ then is that though the system parameters
(net primary production, or system growth rates, species composition)
may change after disturbance, a resilient community will return quickly
to equilibrium after disturbance is removed. A resilient ecosystem does
not necessarily imply that all of its component populations are stable.
Environmental perturbation may result in the extinction of an individual
species without affecting ecosystem function or resilience. Holling (1973)
distinguishes between stability as a property associated with individual
populations of an ecosystem, and resilience as a property of an ecosystem.
Early work by Holling (1973) has suggested that, in general, the resili-
ence of an ecosystem is an increasing function of the diversity of that
system. There is some empirical evidence to support this view. In a series
of field experiments in drought-affected grasslands in Minnesota, Tilman
(1996) has shown that species-poor plots were less productive in terms of
biomass than species-rich plots (see Figure 4.1A). He also demonstrated
that species-poor plots were more greatly harmed by drought (less resist-
ant), took longer to return to pre-drought conditions (less resilient) and
were less stable than species-rich plots. Tilman et al. (1997) also demon-
strated that plots with lower functional diversity had lower productivity in
biomass terms than plots with high functional diversity (see Figure 4.1B).
200 A
180
160
Plant biomass (g/m2)

140
120
100
80
60
40
20
0
0 5 10 15 20 25 30 35
Species diversity (number of species added)
220 B
200
180
Plant biomass (g/m2)
160
140
120
100
80
60
40
20
0
0 1 2 3 4 5
Functional diversity
(number of functional groups added)
Note: Curves shown are simple asymptotic functions fitted to treatment means.
Source: Tilman et al. (1997).
Figure 4.1 (A) Dependence of 1996 above-ground plant biomass (that

is, productivity) (mean and standard error) on the number
of plant species seeded into the 289 plots. (B) Dependence of
1996 above-ground plant biomass on the number of functional
groups seeded into each plot
Having considered some reasons why biodiversity might be important

we now provide some evidence as to why managed landscapes might be
so significant in supporting biodiversity conservation goals. Managed
landscapes that are vital economically in supporting human populations
can also make a very important contribution to biodiversity conservation.
Biodiversity can coexist alongside human activities and economically
valuable managed landscapes do not necessarily have to exclude biodiver-

sity conservation goals. Indeed, there is a significant literature that sug-
gests that many managed systems may actually augment biodiversity and
that land abandonment or a decline in management activities can actually
threaten biodiversity. We now turn to some examples of how land manag-
ers from around the world manage systems that conserve biodiversity. We
also consider what can happen to biodiversity when land and management
activities are abandoned.
Landscape management and biodiversity

In many managed landscapes good conservation practice succeeds because
it is perceived to coincide with the interests of land managers whose
support is vital for conservation initiatives. Such conservation practices
may also have been developed to avoid over-utilization of the resource on
which the human population depends. Consequently most biodiversity
exists in human-dominated ecosystems (Pimmental et al., 1992).
First, some examples of where good conservation practice is coinciden-
tal with the interests of land managers are provided, and second, we look
at abandonment. In areas where human populations have long been an
integral part of the landscape and had much to do with its recent evolution,
species may have adapted to ‘managed’ landscapes. For example, human
impacts on biodiversity in the Mediterranean basin may play a positive
role where current levels of biodiversity are in part maintained by contin-
ued human influence. Pignatti (1978) reports that domestic livestock, and
an opening up of evergreen oak forests in the Mediterranean, provided
new opportunities for speciation of herbaceous annual flora. For example,
the dehesas of south-west Spain have evolved around a distinct and long
history of anthropogenic influence. These open wood pasture systems are
derived from ancient Mediterranean forests and are managed to support
livestock production with some accompanying arable cultivation and
silviculture but are widely recognized as being of high conservation value
(Baldock et al., 1993; Telleria and Santos, 1995; Díaz et al., 1996).
Floristic diversity is high and dehesa grasslands are remarkable for
maintaining some of the most species-rich grasslands outside the tropics,
with as many as 60 plant species per square metre having been recorded
(Marañon, 1986). A number of explanations have been advanced for the
high floristic diversity associated with dehesas. The Mediterranean basin
acts as a transitional biogeographical location. It has been suggested that
its flora, which comprises several different genetic elements, has been
enriched by historical climatic fluctuations during the Quaternary, by
complexity of mountain relief and by altitudinal heterogeneity and histori-
cal human disturbance (Zohary, 1973; Whittaker, 1977; Marañon, 1986).
Defoliation by domestic herbivores and the occurrence of frequent fires

in association with periodic droughts are also thought to have promoted
plant diversification particularly of annual species and initiated adapta-
tions to drought, fire and grazing (Pignatti, 1978; Naveh, 1994).
Many bird species from northern Europe overwinter in the dehesas and
are reliant upon the dehesas as a food source. Telleria et al. (1992) provide
evidence that suggests that dehesas may support more diverse communi-
ties of passerines than neighbouring stands of high forest. Dehesa habitat
supports 64 per cent of the population of common cranes wintering in
Spain (50 000–60 000 birds), which represents 70–85 per cent of the western
European population (Alonso et al., 1990). The crane population is not
widespread in other habitats as cranes rely on acorns, so holm oak dehesas
are of great importance for this species, considered vulnerable because
of their decreasing population trend (Tucker and Heath, 1994). It is also
thought that the distribution of white stork is most strongly associated
with open holm oak wood pasture and it has been suggested that conver-
sion of this habitat could lead to a decline in its populations (Carrascal et
al., 1993).
Managed woodland and grazing systems elsewhere in Europe are also
important for maintaining biological diversity. The bird community of
western oak woods in the United Kingdom, particularly the abundance of
the wood warbler (Phylloscopus sibilatrix), pied flycatcher (Ficedula hypo-
leuca) and redstart has long been recognized as unique and some grazing
may help to create the open conditions in the understorey and field layer
favoured by these species. Some studies that focus on grazing even report
that subspecies of grasses may develop according to the specific ecological
conditions that occur in a grazed or mowed sward (Reinhammar, 1995).
The development of species-rich raised coastal dune and bog habitats in
north-western Europe, known as machairs, is also thought to be strongly
associated with agriculture and human activity, particularly fire and
grazing (Mate, 1992; Edwards, et al., 2005). Machairs, which are priority
habitats under the European Habitats Directive, are unique ecosystems
confined, in the northern hemisphere, primarily to west and north-west
coasts of Ireland and Scotland. Machairs are priority habitats because
of the high plant species richness which contain elements of calcareous
grassland and sand dune plant communities. We now turn to the issue of
abandonment.
Londo (1990) reports that, in the absence of management, semi-natural
grassland communities revert by processes of natural succession to natural
woodland and forest and the diversity of herbaceous species falls. Many
traditional extensive farming practices have been shown to maintain plant
and animal diversity (González Bernáldez, 1991; Naveh, 1994), and where
these activities cease, susceptibility to disturbances, especially fire, can be

increased. Fire in turn can have a negative effect on biodiversity (Faraco
et al., 1993). Landscape homogenization can also result from the aban-
donment of agricultural/pastoral land (Alés et al., 1992). Without human
management diverse plant communities in the Mediterranean basin, for
example, become overgrown, and displaced by relatively few, shrubby
unproductive species. Livestock may play a positive role in influencing the
system.
Bokdam and Gleichman (2000) have suggested that abandonment is
a major threat to traditional pastoral landscapes and their wildlife in
Europe. They report that increased labour costs have undermined tradi-
tional herding systems, which are being replaced by free-ranging grazing
systems leading to a decline in species-rich open heathland. The manage-
ment of Mediterranean woodland has become an important issue in many
areas because of the abandonment of large areas that were previously
exploited by grazing. In many cases impenetrable thickets have developed
with continuous accumulation of fuel leading to catastrophic wildfires.
Valderrábano and Torrano (2000) evaluate goats as a potential manage-
ment tool for controlling encroachment of Genista scorpius in black pine
stands in the Spanish Pyrenees. They report that as a consequence of goat
browsing and thinning, dense woodland was opened up and desirable tree
growth and development was stimulated.
In traditional land husbandry, maintenance of biodiversity and eco-
nomic outputs are closely intertwined. For example, the relationship
between habitat characteristics, weather and spatial variation in animal
behaviour was investigated by De Miguel et al. (1997). They suggest that
shrub areas provide shelter and represent an important browse resource
during winter and that this leads to the occurrence of a diversified land-
scape with different successional stages (from pastures to clear and dense
woodlands) that occur in close proximity, which in turn leads to high
levels of flora and fauna.
Clearly there is evidence to suggest that biodiversity can coexist in land-
scapes of economic importance but can land managers from around the
world actually use biodiversity to support the productive process? This is
the subject of the next section, which considers how land managers may
conserve biodiversity because it supports the productivity, stability and
resilience of the ecosystems they manage.
Landscape management and ecosystem properties

Maintenance of biodiversity may be coincident with management goals
such as improved agricultural productivity under highly variable environ-
mental and socioeconomic conditions. For example, pastoralists in Africa
deliberately maintain as many as a dozen breeds of camel in the Sudan

because they are able to exploit the vegetation of extreme environments,
including deserts and other uncultivated land. The loss of these hardy
animal breeds therefore means a reduction in the area of human habitat
(Köhler-Rollefson, 1993).
In Mediterranean dehesa systems good conservation practices that
promote biodiversity have arisen because local farmers recognize that a
diverse system helps reduce variation in productivity from year to year.
Local farmers do not necessarily have biodiversity conservation goals
in mind as a management aim. Nevertheless, biodiversity coincides with
certain production goals such as improved stability of production under
unpredictable environmental conditions. Large differences in climatic,
geological and topographical gradients contribute to a considerable
degree of variation in productivity across the regional landscape. Climatic
factors are instrumental in dictating plant and animal dynamics and prod-
uctivity. Consequently, a long history of anthropogenic influence has led
to the development of a high level of management and functional com-
plexity as a means of ensuring stability. Attempts to reach an understand-
ing of the relationship between management practices and environmental
variability have prompted research effort into the interactions between the
individual components that comprise this complex ecosystem including
tree, herbaceous and shrub, and livestock components.
For example, holm oak is preferred by dehesa farmers because it
favours the growth of highly productive perennial herbaceous species
through improved retention of soil moisture, modification of micro-
climate, improved nutrient availability and improved soil properties
(Marañon, 1986; Joffre and Rambal, 1993). Marañon (1986) has observed
a much higher phenological diversity in dehesa systems that include a tree
component (Figure 4.2).
Groups of perennial herbaceous species may be significant for maintain-
ing productivity because they are able to utilize nutrients and moisture
more effectively. These include Agrostis castellana, Dactylis glomerata,
Lolium perenne and Phalaris aquatica, which were all found more fre-
quently beneath tree canopies than in the open field (Joffre et al., 1988).
Joffre et al. (1988) hypothesized that differences in nitrogen utilization
occurred between annual and perennial species and that the efficiency of
nitrogen utilization by herbaceous species was affected by the tree canopy.
They report higher nitrogen mineralization in grasslands with perennials
compared with annuals and greater nitrogen mineralization below the tree
canopy.
Farmers have evolved complex farming systems specifically to be able
to exploit resources in species-rich environments. Diverse multispecies
50
Species richness
40
(no. sp./4 m2)
30
20
10
Oak understorey Canopy edge Open grassland
Note: Mean standard error and range have been drawn.
Source: Marañon (1986).
Figure 4.2 Average species richness (number of species in 4m2 plots)

in three habitats: oak understorey, canopy edge and open
grassland
herbivore systems, such as game ranches on the savannas of Africa, may

include up to 20 different mammal herbivore species (Cumming, 1993).
Managers exploit differences in grazing habits that can lead to a degree of
complementarity in the use of forage resources where the total productiv-
ity of the range is seen to increase. Short grass (concentrate) grazers benefit
from the modification of sward structure brought about by long grass
(bulk) grazers, for example, sheep generally perform better when grazed in
mixed systems than when grazed alone (Nolan and Connolly, 1977). This
is usually only the case when large quantities of unpalatable poor-quality
fodder are available. McNaughton (1984) reports that in the Serengeti,
the larger bulk grazers consume long grass and these are then followed by
smaller ungulates that create ‘grazing lawns’. These lawns are sources of
high-quality forage and so herbivores are seen to influence the quality and
productivity of the grazing resource. Mixed species grazing systems may
also be preferred to single species systems because they improve yields and
do not over-exploit productive herbaceous species. It has been reported
that sheep and cattle may affect the plant community in different ways.
Bedell (1973) has shown that sheep can reduce the abundance of clover in
a sward but they also increase the amount of Poa trivialis (Conway et al.,
1972). In contrast, a high proportion of cattle has been shown to increase
the amount of clover relative to grass. In this manner combined cattle

and sheep grazing systems may be more productive than single species
systems.
Examples of rangeland management systems that attempt to encourage
diversity in herbivore populations to enhance resilience include replac-
ing monocultures of domestic livestock with multispecies game systems
and combined cattle/game ranches such as the CAMPFIRE (Communal
Areas Management Programme for Indigenous Resources) programme in
Zimbabwe (Cumming, 1993). Scholes and Walker (1993) have suggested
that events such as fire and herbivory play an important role in maintain-
ing the diversity and resilience of such systems. Here, the reduction of such
perturbations is thought to reduce landscape diversity and the ability of
the system to survive similar shocks in the future.
Moreover, in savannas (Walker et al., 1981) and Agrostis-Festuca grass-
land in Britain (Hulme et al., 1999) groups of grass species are important
in maintaining the system’s productivity. In other habitats, such as boreal
and deciduous forests in North America and Europe, where insectivorous
bird species are considered to be instrumental in controlling outbreaks of
forest insect pests, overall species diversity is important for maintaining
stability (Morris et al., 1958; Tinbergen, 1960; Campbell and Sloan, 1976;
Holmes et al., 1979; Takekawa et al., 1982; Holling, 1988; Maquis and
Whelan, 1994). There may also be indirect effects of diversity with some
species influencing the survival of other species, such as key plant species
determining the course of successional processes through the provision of
so-called ‘nurse effects’. Several studies have observed a greater number of
seedlings beneath mature trees compared with more open areas (Griffin,
1971; Borchert et al., 1989; Espelta et al., 1995). Similarly, shrub species
may influence seedling establishment, acorn consumption and the extent of
browsing by herbivores (Morgan, 1991; Callaway, 1992; Herrera, 1995).
Despite the importance of biodiversity in contributing towards eco-
logical services many ecosystems are undergoing profound change due
to economic development. Heavily disturbed agricultural ecosystems in
many parts of the world are threatened by human intervention, which
has resulted in a loss of biodiversity, productivity and resilience. Land
managers frequently need to know how biodiversity is affected by the
level and intensity of management and what will be the result of that
change. The significance of human-induced disturbance and environmen-
tal perturbation is the subject of the next section.
Ecosystem disturbance and biodiversity

Considerable insight into the understanding of conservation biology
has been gained through knowledge of the effects of human-induced
disturbance on biodiversity (Wilson and Johns, 1982). There is a substan-

tial literature that shows that human-induced disturbance and habitat
degradation can result in a decline in biodiversity and species extinction.
Highly intensive agricultural practices that reduce spatial habitat com-
plexity leading to homogenization of the landscape may lead to biodiver-
sity loss. The decline in most of Europe’s SPECS (Species of European
Conservation Concern) has been linked to land use and management
changes with agricultural intensification being cited as the most significant
threat to bird populations (Tucker and Heath, 1994). Arable farming
systems in parts of Europe are thought to have played a part in the decline
of many species. For example, as a consequence of changing conditions in
agricultural fields in Britain many bird species have undergone significant
population declines. Fuller et al. (1991) report that many British farmland
birds have declined dramatically over the last three decades as agricultural
land use has altered, hedgerows have declined and farms have developed
to form larger contiguous areas. A reduction in forest area due to agricul-
tural expansion can also reduce species diversity. Studies of the avifauna
of fragmented forests have shown that some species are absent or infre-
quent in very isolated sites and that smaller woodland size gives rise to
less bird species diversity (Lynch and Whigham, 1984; Opdam et al., 1985;
Ford, 1987; van Dorp and Opdam, 1987).
Data from censuses of domestic animals collected for tax purposes as
well as from hunting statistics have been combined with palynological
reconstructions of vegetation to demonstrate the long-term ecological
effects of management practices. For example, hunting statistics for moose
and roe deer in Sweden suggest dramatic recent population increases that
have probably contributed to the decline of deciduous tree species (Ahlén,
1975). Peterken and Tubbs (1965) related fluctuating grazing regimes of
horses, pigs and cattle in the New Forest, England, to waves of regenera-
tion based on the age structure of existing trees. In Poland pollen data has
enabled reconstructions of vegetation successions (Mitchell and Cole,
1998). This has been combined with data on herbivore densities for forests
in eastern Poland over the last 200 years (Faliñski, 1986) and shows that
the proportion of conifers, principally Picea abies, increased dramati-
cally at the expense of broadleaved species during the period of intensive
grazing. Tree regeneration in the subsequent low-intensity grazing period
was dominated by broadleaved taxa, initially Betula, Populus and Caprinus
and, subsequently, Tilia and Quercus.
Jorritsma et al. (1999) used a dynamic simulation model FORGRA
to evaluate the impact of grazing on Scots pine regeneration in the
Netherlands. They showed that even low densities of ungulates could have
a significant impact on forest development. Their model indicates that
the presence of one cow per ha virtually eliminates recruitment entirely.

Simulations of the model described above by Kienast et al. (1999) confirm
these results since they also demonstrated that high browsing pressure
does reduce recruitment and does alter the forest structure considerably,
leading to high rates of tree mortality and more open forests. The spatial
model developed by Weber et al. (1998) was used to determine the effects
of grazing intensity and grazing heterogeneity applied to the southern
Kalahari and shows that high levels of grazing lead to shrub invasion.
Jeltsch et al. (1997) also reported that when grazing intensity reaches a
critical level, shrub cover increases, drastically lowering the productivity
of the range.
High levels of disturbance by wild or relatively unmanaged introduced
animals are also thought to affect ecosystem productivity. The study of
long-term grazing–vegetation interactions using palaeovegetation data in
Ireland show that reasonably high populations of giant Irish deer imposed
a high pressure on shrubby vegetation and had a profound effect on the
change in vegetation communities from juniper scrub to grassland in
Ireland during the Late-glacial Interstadial (11 000–12 000 BP) (Bradshaw
and Mitchell, 1999). In Galicia, Spain, Hernandez and Silva-Pando
(1996) report a decline in the abundance and diversity of shrub species
after a period of three years’ grazing by red (Cervus elaphus) and roe deer
(Capreolus capreolus). Jane (1994) considered the long-term effects of
browsing by red deer (Cervus elaphus) on mountain beech (Nothofagus
solandri) in New Zealand and concluded that the impact of high deer
densities on vegetation remains and can persist for many decades. She sug-
gests that in critical high altitude areas, large reductions in deer densities
were required to trigger the regeneration necessary for tree replacement.
Intensive levels of herbivory may reduce plant productivity, survival,
reproduction and growth (Fay and Hartnett, 1991; Fox and Morrow,
1992; Relva and Veblen, 1998). In a long-term experiment to evalu-
ate the impact of domestic livestock on tree species Hester et al. (1996)
manipulated sheep stocking density and season in an upland broadleaved
woodland in Cumbria, UK. They observed that growth and survival to
the sapling stage was negatively correlated with grazing intensity, and
suggest that, apart from plots grazed at the lowest animal densities, only
a small proportion of saplings will attain canopy height. Other studies
from around the world implicate browsing by domestic livestock as a
cause of poor tree species recruitment (Kingery and Graham, 1991). Van
Hees et al. (1996) employed an exclosure to determine the impact of roe
(Capreolus capreolus) and red deer (Cervus elaphus) on beech (Fagus silvat-
ica), pedunculate oak (Quercus robur) and silver birch (Betula pendula) in
the Netherlands. They showed that browsing reduced sapling abundance,
height and above-ground biomass of all three species. Some studies in the
UK report a high number of seedlings within fenced enclosures compared
with unfenced areas (Sykes and Horrill, 1979; Marrs and Welch, 1991;
Staines, 1995). Historical records have also been used that suggest that
deer may prevent natural regeneration of Scots pine (Pinus silvestris) in
the United Kingdom. With respect to stocking densities, studies inves-
tigating the impact of ungulates foraging on upland heaths in Scotland
suggest that red deer (Cervus elaphus) at stocking densities of .1 deer/20
ha can prevent tree regeneration (Staines et al., 1995).
Persistent high levels of disturbance are also thought to affect ecosystem
function, particularly where these eliminate important functional groups
that affect ecosystem processes. Groups of grass species may be significant
in maintaining the productivity of savanna ecosystems (Walker et al.,
1981). Walker et al. found that grasslands with persistent intensive grazing
by settled peasant farmers had lower levels of productivity than moderate
opportunistic grazing practices employed by nomadic pastoralists. In the
former case, productive functional groups declined because herbivores
showed a preference for the most palatable species, whilst in the latter
case these preferred species were able to persist in the sward and adapt to
change and instabilities caused by grazing and drought, thereby maintain-
ing structural resilience.
Overgrazing may exacerbate the high inter-annual variation in produc-
tivity on many rangelands. Walker (1988) has observed a much higher
phenological diversity in semi-arid systems not subject to heavy grazing
compared with those that are intensively grazed. On lightly grazed areas
he noted an even mix of early, mid- and late season grasses that were
able to respond to rainfall wherever it occurred in the season. Heavy
grazing leads to an absence of highly palatable early season species that
are replaced by later growing species (Silva, 1987). The implication being
that forage production was lower and more unstable on heavily grazed
areas compared with lightly grazed land because the sward was not able
to respond to early season rains. In the Serengeti, McNaughton (1985)
has also shown that forage production was more stable where the number
of species contributing to biomass was high compared with swards where
relatively few species contributed to forage production.
Many complex ecosystems that aim to maximize heterogeneity (such as
non-equilibrium systems) may be threatened by intensive grazing regimes
that attempt to restrict livestock movements. This may have a negative
impact on the stability and sustainability of the system. In areas where the
fodder resource is widely dispersed seasonally and spatially, restrictions
on stock movements by using paddocks can lead to land and vegetation
degradation (Hoffman and Cowling, 1990). Increasing the connectivity
of an under-connected system may also cause the system to change to a

new stable state. An example of such a change has occurred in semi-arid
shrublands, where the erection of fences has restricted animal move-
ments, leading to localized feeding and degradation of vegetation and soil
resources (Hoffman, 1988).
There is evidence that overgrazing may trigger the transition from one
ecosystem to another, for example from forest to grassland or grassland to
a shrubby semi-desert (Holling, 1973; Westoby et al., 1989). This evidence
suggests that state and transition models (STM) are appropriate in many
rangeland situations where vegetation is best described by a set of discrete
‘states’ and a set of discrete ‘transitions’ between states. During a transi-
tion the system jumps to another state if a threshold is exceeded. This can
be triggered by fire, rainfall or grazing and the system never rests halfway
through a transition. They use an STM in eastern Australia to show how
once ecological thresholds are exceeded the system shifts from a woodland
with a grass understorey to a less productive shrubby state. The concept of
ecological thresholds suggests that there are limits to the ability of ecosys-
tems to withstand environmental perturbation. If such limits are exceeded,
ecosystems may shift to a less productive phase.
In some circumstances human management and spatial landscape
change may undermine ecosystem processes. For example, some empirical
studies have demonstrated that habitat fragmentation may reduce para-
sitism rates on herbivorous insects at different spatial scales (Kruess and
Tscharntke, 1994; Roland and Taylor, 1997). Similarly, silvicultural prac-
tices may reduce parasitism rates of spruce bark beetle (I. typographus) in
central Sweden (Weslien and Schroeder, 1999).
Clearly persistent intensive use of resources can degrade ecosystems
and impact negatively on key functional groups. However, the level and
nature of disturbance appears to be an important factor. Some examples
of managed ecosystems indicate that an element of disturbance caused
by human intervention may actually enhance biodiversity. In the next
section we explore the impact of low and moderate levels of disturbance
on biodiversity, productivity and resilience.
Moderate ecosystem disturbance and biodiversity

MacArthur and Wilson (1967) suggest that some disturbance can promote
diversity because different species respond to disturbance in different ways.
They first characterized species as either r- or K- strategists, which have
evolved mechanisms to optimize resources in quite different environments.
The former (r-strategist) refers to species that attempt to maximize growth
in an unconstrained environment, reproduce quickly, disperse widely and
are of smaller size and shorter lifespan. On the other hand K-strategists
include species that optimize growth in a climax successional phase or

a crowded environment, are highly adapted to stable equilibrium condi-
tions, are less flexible, more vulnerable to change, are generally longer
lived and do not disperse as well. High levels of disturbance may lead to
species-poor habitats since they favour the persistence of competitive,
opportunistic r species better adapted to cope with disturbance (Miller,
1982). Conversely, undisturbed environments that do not undergo change
may support less diversity because they favour the persistence of domi-
nant K-strategists. Linder et al. (1997) examined the effects of fire history
on stand structure and plant diversity in Swedish forest reserves. They
concluded that the reintroduction of fire represents an important means
of disturbance that was necessary to promote diversity of flora and fauna
in the area. Continued fire suppression has changed successional patterns
and altered stand structure. Late successional species such as spruce domi-
nate due to lack of fire, and pioneer species such as pine, silver birch and
aspen are decreasing in number because they require fire disturbance to
regenerate. This appears to accord with MacArthur and Wilson’s (1967)
theory where undisturbed environments may therefore support less diver-
sity because climax species are favoured. Linder et al. (1997) recommend
prescribed burning to ensure a relatively wide range of successional stages
to promote biodiversity over the longer term.
Higher habitat diversity due to moderate disturbance can also be
explained by niche relations and the manner in which species divide up
limited resources for their survival (Schmida and Wilson, 1985). They may
divide up the available space (for example, by selecting different habitats)
or energy resources (for example, by adopting different diets). Some studies
serve to demonstrate that moderate levels of human activity may enhance
biodiversity by opening up new niches, providing new food or protection
from predators and by diversifying micro-habitats. For example, struc-
tural heterogeneity is thought to be important for bird species diversity
and vegetation indexes have been developed to quantify structural diver-
sity particularly in relation to bird species (MacArthur and MacArthur,
1961; Willson, 1974; Erdelen, 1984). Several studies indicate that a decline
in structural diversity (James and Wamer, 1982; Terborgh, 1985; Ratcliffe,
1993; Telleria and Carrascal, 1994) and floristic diversity (Lynch and
Whigham, 1984) leads to less bird species diversity. This is confirmed by
Casey and Hein (1983) and Dambach (1944) who all reported that wood-
land heavily browsed by deer supported fewer bird species than woodland
that was not grazed (although see DeGraaf et al., 1991).
Schemske and Brokaw (1981) provide empirical evidence to show
that moderate disturbance in tropical forests caused by natural tree falls
resulted in the greatest diversity of bird species. Clout and Gaze (1984) in
New Zealand found that the highest levels of bird diversity were recorded
in disturbed productive forests while undisturbed mature forests contained
less bird diversity, though they were populated predominantly by native
bird species. Sternberg et al. (2000) conducted a four-year study on the
response of a Mediterranean herbaceous community to grazing manage-
ment in north-eastern Israel. Contrasting different grazing treatments they
found that low and high grazing regimes reduced herbaceous diversity but
that moderately grazed areas increased diversity.
The study of long-term grazing–vegetation interactions (102–106 years)
using palaeovegetation data permits the reconstruction of vegetation and
herbivore abundance and associations. Data from Jutland in Denmark
from the Holocene about 5000–7000 years ago suggests that large forest
herbivores did not have a significant influence on regional forest structure
(Bradshaw and Mitchell, 1999). This is because either large predators
held populations at modest levels, or the diversity of grazing species held
populations at stable, low populations of individual species.
Moderate levels of insect herbivory may actually increase productivity.
For example, Holling (1978) carried out an experiment on the defoliation
of balsam fir (Abies balsamea) by spruce budworm. The larvae result in the
death of mature trees aged 55–60 years though young trees are unaffected.
Saplings grow rapidly after mature forest is damaged, and the forest is
restored by its juvenile population. In the short term there is a shortfall
in the production of timber, but over the longer term wood production
remains unaffected. In fact, production rates of the juvenile forest remain
above that of the mature forest because in a mature stand, most trees
have passed their rapid-growth phase. Mattson and Addy (1975) reached
similar conclusions in their study on the effects of forest tent caterpillars
on aspen.
French et al. (1997) conducted a study on the development of Scots
pine in the Cairngorm mountains in Scotland and found that recruitment
is possible provided grazing/browsing pressure remains at a low level.
Similarly Sun et al. (1997) evaluated the effects of cattle grazing and seed-
ling size on the establishment of Araucaria cunninghamii in a silvo-pastoral
system in north-east Australia. They report that grazing did not cause
unacceptable mortality due to the fact that the tree has prickly needles,
which prevented browsing by cattle. They recommend that grazing does
not affect recruitment and can begin immediately after tree planting pro-
vided that a moderate stocking rate is used. Elsewhere, modelling has
also shown that plant populations may be little affected by low levels of
herbivory. Kienast et al. (1999) used a succession model – FORECE − to
assess the long-term dynamics of alpine forests in Central Europe. They
report that moderate levels of browsing posed no threat to the long-term
survival of these forests and did not alter the successional sequence of
forest development.
Recent developments on the functional complexity of ecosystems show
that small disturbances may actually enhance ecosystem function and
increase resilience. Holling (1986) and Holling et al. (1994) suggest that
some natural disturbances initiated by fire, wind and herbivores are an
inherent part of the internal dynamics of ecosystems and in many cases
set the timing of successional cycles.These natural perturbations are part
of ecosystem development and evolution, and seem to be crucial for main-
taining ecosystem resilience and integrity (Costanza et al., 1993). In the
absence of such shocks, the system will become highly connected and this
will provoke even larger perturbations that are more destructive to the
ecosystem because they reduce the ability of the system to survive similar
shocks in the future (Scholes and Walker, 1993).
Some empirical studies reveal that herbivores may enhance a system’s
ability to resist environmental perturbation. For example, in their work
on Florida mangroves, Simberloff et al. (1978) reported that the action of
isopod and other invertebrate root borers resulted in new growth of roots
at the point of attack. More extensive root systems in mangroves result in
greater stability and resistance to storms and therefore benefit the plant.
In Britain, Hulme et al. (1999) carried out a study to evaluate the effects
of sheep grazing on the productivity of upland Agrostis-Festuca grassland.
The experiment controlled sheep grazing at light, heavy and moderate
levels. Both low and high levels of grazing resulted in the spread of less
desirable species such as Nardus stricta and Molinia caerulea. Moderate
levels of grazing maintained preferred species such as Festuca rubra and
Agrostis capillaris and prevented the spread of Nardus stricta and Molinia
caerulea.
Non-equilibrium rangeland systems as practised by nomadic pastoral-
ists in parts of Africa have evolved opportunistic management regimes
that employ moderate levels of grazing intensity that do not eliminate key-
stone elements but instead maintain the resilience of these components.
Scholes and Walker (1993) have suggested that events such as fire and
herbivory may play an important role in maintaining the diversity and
resilience of such systems. Nutrient release following small fires supports
a flush of new growth without destroying all of the old growth. Rangeland
patches are affected but the forage resource remains intact. Small fires
prevent the accumulation of forest biomass, which fuels very large fires
that can decimate large areas of rangeland (Scholes and Walker, 1993) or
whole forests (Holling et al., 1994). Such events may affect the parameters
of the system and cause it to cross a threshold into an alternative state,
which may alter the system’s capacity to provide ecological services. For
example, the Yellowstone National Park in the United States employed

a ‘natural burn’ policy of management that culminated in catastrophic
forest fires.
As described above, human-induced perturbation on managed eco-
systems is a critical factor in maintaining biodiversity. However, the
application of best scientific practice by land managers may not in itself
be sufficient to achieve biodiversity conservation goals. This is because
markets may fail to account for the value of biodiversity to society. It is
essential therefore that land managers are aware of the limitations and
opportunities of the market. The next section explores the economic link-
ages between markets and biodiversity.
Biodiversity in managed landscapes: economic issues

As reviewed above, biological resources in many of the world’s low-
intensity managed habitats represent a significant contribution to economic
activity. However, many traditional low-intensity managed habitats are
threatened by development – a change in land use management due to the
prospect of increased private returns. According to the economic theory
of general equilibrium, the search for opportunities for increased private
returns can ensure that resources are allocated to the highest-value use
available, so that economic efficiency is achieved. This result depends on
a number of conditions. If these conditions are fully met, land use change
motivated by private profit need not be a cause for concern. However,
managed landscapes, in common with most environmental goods, have
characteristics that ensure that the necessary conditions will never be fully
met in practice. In general terms, this failure implies that any resulting
allocation of resources is likely to be economically inefficient, meaning
that it would be possible to reallocate resources in such a way as to make
at least one member of society ‘better off’.
Some mention of the distinction between economic efficiency and equity
is worthwhile at this point. Economists place great emphasis on economic
efficiency but this will not necessarily result in a fair outcome. For a society
to be sustainable, its welfare should not be declining over time (WCED,
1987; Pezzey, 1989). In theory there are potentially a number of efficient
time paths that are sustainable. However, efficiency does not necessarily
guarantee sustainability between, say, current and future generations,
in terms of the distribution of natural resources such as biodiversity
(Common and Perrings, 1992; Perman et al., 2003).
In the case of managed landscapes, the danger is that land use change
guided by market signals alone may lead not to beneficial development
but to loss of valuable and possibly irreplaceable resources. The neces-
sary conditions that must hold in order for market-led development to
be benign are typically discussed under two headings referring to their

absence in practice. These are (1) market failure and (2) policy failure.
Causes of biodiversity loss: market and policy failure

Market failure occurs when private decisions based on a set of prices, or
lack of them, do not generate an efficient allocation of resources (Hanley
et al., 1997). With respect to biodiversity the concern is that market prices
are not reliable indicators of social cost. Social cost refers to the opportu-
nities forgone by society in committing resources in some way (Coase,
1960) and social cost in this study is taken to mean the true value that
society as a whole places on natural resources. Private cost, on the other
hand, refers simply to the financial cost faced by the private individual
or firm undertaking the land use change, at current and expected market
prices.
This divergence between private and social cost occurs because managed
biodiverse landscapes generate benefits to society in addition to those that
are transacted in the market system: external benefits. An absence of such
external effects is one of the necessary conditions for market efficiency
referred to in the previous section. Typically, the reason these benefits
remain external to the market system is that they have the characteristics
Table 4.1 Characteristics of public and private goods
Pure Private Quasi-private Quasi-public Pure Public Goods

Goods Goods Goods
Rivalness in Regular payments Up to some Non-rivalness in

consumption; in form of taxes or capability constraint consumption; non-
excludability; changes are made (carrying capability), excludability; or
property rights, to finance supply non-rivalness in excessive costs of
market prices consumption excludability
use values non-use values

(market revealed) (non-market
revealed)
VALUATION DIFFICULTIES
Degree of ‘familiarity’ declining
Problems of ‘perception’ increasing
New information requirements increasing
Problems of reliability/validity increasing
Reliance of interdisciplinary research effort and findings increasing
Source: Turner (1993).

of public goods, in particular they are indivisible and perhaps also non-
excludable, making their exchange in markets unlikely (Table 4.1).
The public good nature of biodiversity creates difficulties for its valu-
ation. These will be discussed in some detail in the sections that follow.
Because managed landscapes provide high levels of unpriced public ben-
efits, in terms of wildlife and landscape quality, private agents will have no
incentive to take account of these benefits in decisions over land use.
Conventional economic theory seeks to cast government in the role of
an objective and well-informed ‘third force’ (in addition to individuals
and firms), with some ability to intervene to correct for market failures.
Government or policy failure occurs when policy decisions required to
correct for market failure are not implemented and fail to fully recognize,
or incorporate, the values associated with environmental resources. Policy
failure may also arise where government decisions themselves induce eco-
nomic inefficiencies. For example, agri-environment policies, through cre-
ating incentives for farmers to expand production, may result in a greater
privately optimal level of degradation than would be the case in the absence
of such policies. Poorly formulated policy instruments and incentives may
distort the allocation of resources unintentionally. Simpson et al. (1998)
suggest that high stocking rates are caused by incentives to graze moorland
to achieve profit maximization, encouraged by support from the Common
Agricultural Policy (CAP). They indicate that increases in the ewe flock
across the Northern Isles (in Orkney from 37 000 in 1983 to nearly 55 000
in 1992, and in Shetland from 116 000 in 1982 to 156 000 in 1993) was in
response to the EU’s sheep meat regime introduced in the early 1980s. The
scheme offered headage payments and a variable premium in fat lamb sales.
They suggest that the policy has been sufficient to increase stocking levels,
and hence heather utilization rates, across the Northern Isles. McNeely
(1993) suggests that in Botswana, national and European Union subsidies
have led to excessive uncontrolled grazing of rangelands and degradation
of grazing savanna, which have affected the long-term productivity of the
resource. Subsidies that aim to promote cash crops to secure export revenue
may result in land degradation, soil nutrient losses and a reduction in the
resilience of ecosystems (Grainger, 1990). Royalties in forestry can lead to
excessive rates of deforestation (Repetto, 1989; Barbier et al., 1991).
The catch-all term ‘market failure’ is defined so as to refer to all situa-
tions where the market signals perceived by private individuals fail to coin-
cide with social values (and fail to produce economic efficiency). However,
some necessary conditions for market efficiency, which may be violated
in practice, tend to be omitted from discussions of market failure, and are
worth briefly mentioning here. The discussion above relates mainly to what
might be called the ‘complete set of markets’ condition (Common, 1995).
Also important is the ‘complete information’ condition, requiring not

only that prices be widely known, but also that they reflect the full implica-
tions of any reallocation of resources. It is clear from the discussion above
that in the case of managed landscapes, such knowledge is available only
in partial and uncertain form, and is not reflected in actual market prices.
The effect of such uncertainty is discussed below.
The so-called ‘rationality condition’ may also be violated for environ-
mental goods such as managed landscapes and associated biodiversity.
The link between market efficiency and the (constrained) satisfaction of
the wants of individuals and of the society of which they are members relies
upon a number of assumptions about the nature of individuals’ prefer-
ences. Rationality of preferences includes the ability and willingness always
to make comparisons between goods. We will see below that stated prefer-
ences for environmental resources can include a refusal to do this, on the
grounds that a biological resource should be preserved ‘in its own right’. To
the extent that individuals do not in fact have ‘rational’ preferences, market
outcomes will tend to deviate from socially desired outcomes.
Valuing biodiversity
The main point that is frequently made by environmental economists
working on valuation with regard to market and policy failure is that
private resource users do not attribute sufficient weight to biodiversity.
Valuation, it is argued, aims to redress this imbalance and sets out to
determine what weight should be given to biodiversity in the interests of
society as a whole.
The literature indicates that a variety of methods have been employed to
estimate wildlife values in managed landscapes. Studies on wildlife value
have focused on their use and non-use values. These values are based on
an individual’s willingness to pay (WTP) or willingness to accept (WTA)
compensation. Gross willingness to pay might include the cost of travel,
purchase of equipment to participate in the recreation activity, actual fees
associated with the activity and consumer surplus. The concept of ‘total
economic value’ (TEV) has been used to describe the components of value
as shown in Figure 4.3. Use values associated with managed landscapes
refer to the actual and/or planned use of a service by an individual and
include recreational activities such as bird watching or hunting. Use values
also include the following: option value, that is, the value of the option
to guarantee use of the service by the individual in the future (Weisbrod,
1964); and quasi-option value, that is, the value of future information
protected by preserving the resource now, given the expectation of future
growth in knowledge relevant to the implications of development (Arrow
and Fisher, 1974; Perman et al., 2003).
Economic values in managed

landscapes
Use values Non-use values
1. Direct use 2. Indirect 3. Option 6. Existence

value use value value value
Livestock, crop Watershed Future uses as Cultural heritage

and forest protection per 1 and 2
Intrinsic worth
products
Nutrient cycling
Educational,
Microclimatic
recreational and
regulation
cultural uses
Carbon store
External support
etc
Source: Adapted from Barbier (1994).
Figure 4.3 Economic values in managed landscapes
More recently, empirical studies on wildlife values have placed emphasis

on non-use values. These refer to situations where an individual knows a
biological resource exists and will continue to exist, independently of any
actual or prospective use by the individual and where that individual would
feel a ‘loss’ if the resource were to disappear (Brown, 1990). Existence value
arises when the utility function of a consumer is enhanced by the knowl-
edge that a certain wildlife species exists. As indicated in Figure 4.3, non-
use values thus refer to situations in which individuals would like to see a
biological resource preserved ‘in its own right’. Non-use values include the
following: bequest value: the value of ensuring that the resource remains
intact for one’s future heirs (Krutilla, 1967); existence value: the value that
arises from ensuring the survival of a resource (Pearce and Turner, 1990;
Perman et al., 2003). Existence value is usually assumed to embody some
form of altruism, either for other human beings or for a concern for non-
human entities. For example, some of the literature distinguishes between
philanthropic motives based on the provision of services to other people,
and altruistic behaviour solely concerned with nature. The sum of all use
values and non-use values is referred to as total economic value (TEV).
In the literature, two classes of use value are sometimes defined – direct
use value and indirect use value. This is illustrated in Figure 4.3. Direct
use value is the same as that outlined above and includes, for example,
harvesting timber from a forest or the use of recreation services provided
by a national park. Indirect use value on the other hand refers to the life
support services provided by ecosystems. These include ecosystem func-
tions such as flood control, catchment protection, nutrient cycling and
carbon sequestration. The biological diversity of managed landscapes may
serve an important role in maintaining ecosystem functions and thus serve
to support the productive process.
Measuring biodiversity values

Early studies on wildlife values employed revealed preference methods,
such as the so-called travel cost method (TCM). This technique was first
proposed for use in recreation studies and was subsequently refined and
applied in empirical studies by Clawson and Knetsch (1966). The method
is based on the premise that it should be possible to infer values placed by
visitors on environmental outdoor recreation services from the costs that
they have incurred in order to experience these sites. Such costs include
costs associated with travelling to a recreation site and the imputed value
of people’s time. A statistical relationship between observed visits and the
cost of visiting a site is determined. This is then used to construct a demand
curve from which consumer surplus can be measured. The current value of
the resource and value of alternative policies affecting the resource can then
be evaluated using consumer surplus calculated from the demand curves.
The advantage of TCM is that the data collected involves actual consumer
behaviour. Its chief disadvantage is that it does not accurately value trips
for multiple purposes. TCM has been used extensively in the United
Kingdom and United States for valuing the non-market benefits associ-
ated with national parks and managed landscapes including public forests
(Bowes and Krutilla, 1989; Benson and Willis, 1991; Whiteman, 1991).
A second method of estimating wildlife value is a stated preference
method, the contingent valuation method (CVM). This involves the
construction of a hypothetical or simulated market for an environmen-
tal or wildlife resource. Contingent valuation techniques use surveys to
elicit individuals’ preferences for public goods by finding out what they
would be willing to pay (WTP) for them, or what they would be willing to
accept as compensation (so that they would not be worse off) for specified
changes in them. This approach circumvents the absence of markets for
public goods by presenting consumers with hypothetical markets in which
they have the opportunity to purchase the good. Willingness to pay is
determined either through a written questionnaire or using bidding games
implemented by personal interviews. Demand curves are then constructed,
and consumer surplus used as a measure of use and non-use value. The
CVM has the advantage of utilizing all the structural characteristics of

demand analysis. Its chief disadvantage is that respondent bias may exist,
pointing out the importance of the art of questionnaire design. Despite
its widespread use CVM is extremely controversial and the values derived
from the technique are treated with some scepticism by many economists
(van Rensburg et al., 2002; Mill et al., 2007). Some go as far as to suggest
that the technique should not be used as the basis for policy decisions
(Hausman, 1993).
The contingent valuation method began to be used widely from the
mid-1970s (Randall et al., 1974; Brookshire et al., 1976). Other detailed
accounts of the method can be found in Mitchell and Carson (1989),
Hanley and Spash (1993), Bateman and Willis (1995), van Rensburg et
al. (2002), and Mill et al. (2007). Relatively few contingent valuation
studies relate specifically to biodiversity (Diamond and Hausman, 1994;
Hanemann, 1994; Portney, 1994).
The value of species and habitats

Many empirical studies applied to wilderness areas indicate that the value
of recreational and other non-marketed direct values derived from areas
of high nature conservation value can be significant and may compare
favourably with competing commercial uses of the same resource. For
example, Hanley and Craig (1991) contrasted the trade-offs implicit in
permitting or prohibiting afforestation with respect to the flow country,
in northern Scotland (the largest body of blanket peat bog in the northern
hemisphere). The development would generate employment and produce
timber but displace extensive populations of internationally rare breeding
birds. They demonstrated that the total recreational value of the resource
exceeded the benefits derived from afforestation at discount rates of 6, 4
and 3 per cent. Similarly, Willis (1991) established that the total recrea-
tional value of the Forestry Commission estate in the United Kingdom
exceeded the value of timber sales.
Garrod and Willis (1997) carried out one of the few examples of contin-
gent ranking techniques applied specifically to biodiversity. They employed
a discrete choice contingent ranking approach to estimate the general pub-
lic’s WTP to increase the area of Forestry Commission forests managed
under three forest management standards designed to offer increasing
levels of biodiversity at the expense of commercial timber production.
This method enables relative preferences for different forest management
standards to be measured at the same time as WTP to promote biodiver-
sity. They suggest that the benefits of changing forest management to meet
these standards far outweigh the financial costs involved.
Some of the benefits associated with biodiversity can be deciphered from
expenditure on the preservation of endangered species. Several empirical

CVM studies have been used to determine values related to the conserva-
tion of individual and endangered species in protected areas (Stoll and
Johnson, 1984; Brown and Henry, 1993). Research on endangered or
threatened species includes the value of preserving the whooping crane
(Grus americana) population at the Arkansas National Wildlife Refuge
in Texas for viewers and non-viewers (Bowker and Stoll, 1988) at about
US$6 per person per year. Similarly, Boyle and Bishop (1987) estimated
the value of preserving the bald eagle at US$17.46 per person per year.
A study conducted by Brown et al. (1994) values the northern spotted
owl and its ancient old growth forest habitat using the contingent ranking
approach. In this study, respondents were offered five different policies.
Associated with each policy were the cost of the policy, the area preserved,
the estimated number of owl pairs preserved and their probability of sur-
vival. They estimated existence values for conserving the northern spotted
owl at about US$20 per person per year. Probabilistic theoretical models
have been used to determine the benefits of important wildlife species
such as the northern spotted owl in old naturally regenerated redwood
forests and have demonstrated the high marginal cost of preservation
(Montgomery et al., 1994). Estimates based on the probability of survival
and a reduction in timber stumpage supply give an estimated welfare cost
of US$21 billion to ensure an 82 per cent chance of the species surviving.
Increasing the chance of survival from say 90 per cent to 95 per cent was
estimated to cost an additional US$13 billion.
Indirect use values and ecosystem function

Much of the discussion in the second section of this chapter dealt with
the properties of ecosystems including their productivity, resilience and
stability. There is a significant literature on the value of ecosystem services
including indirect values (Ellis and Fisher, 1987; Barbier, 1994; Bell, 1997;
Daily, 1997; Barbier, 2000; Daily et al., 2000). Indirect values associated
with biodiversity can be measured using surrogate market approaches
using the production function approach. Information about a marketed
good (timber, crops or livestock sales) is used to infer the value of a related
non-marketed good (for example, forest, agricultural or wetland habitat).
The basic assumption underlying this approach is that, if, for instance,
biodiversity supports agricultural or forest production, then this ecologi-
cal service provides an additional environmental input into the agricul-
tural or forest enterprise.
For example, the stability of a managed ecosystem constitutes an indi-
rect use value and represents an important function to land managers.
As seen above, biodiversity may mitigate large inter-annual variation in
productivity (McNaughton, 1985; Walker, 1988). For instance, the eco-

nomic value of a change in diversity can be evaluated from the change in
livestock liveweight gain associated with a decline in forage biomass as a
result of a decline in grassland diversity.
A number of studies in the applied economics literature have used the
stochastic production function approach suggested by Just and Pope
(1978) to capture the value of crop diversity. These studies indicate that
genetic variability within and between crop species confers the potential
to resist stress, provide shelter from adverse conditions and increase the
resilience and sustainability of agro-ecosystems. Plot studies indicate that
intercropping can reduce the probability of absolute crop failure and that
crop diversification increases crop income stability (Walker et al., 1983).
Therefore, the greater the diversity between and/or within species and
functional groups, the greater is the tolerance to pests. This is because
pests easily spread through crops with the same genetic base (Sumner et
al., 1981; Altieri and Liebman, 1986).
Crop diversity may enhance farm productivity, stabilize farm income
and reduce the risk of outright crop failure (Long et al., 2000). The exist-
ence of a limited number of crops grown in an area makes these crops
more vulnerable to diseases and pests. By maintaining proper crop rota-
tions diversity can improve soil productivity and reduce the need for agro-
chemical applications. Land managers also recognize that soil and climatic
conditions can vary considerably. In such circumstances, growing differ-
ent crops and crop varieties can lead to more efficient use of resources.
Some crops can be grown on fertile land while others can utilize marginal
areas. Therefore, the greater the variability of soil and climatic conditions,
the greater the impact biodiversity will have on improved agricultural
production.
For example, Smale et al. (1998), report that crop diversity is positively
related to the mean of yields and negatively correlated with the variance of
yields in rain-fed districts of the Punjab in Pakistan. Di Falco and Perrings
(2003, 2005) found cereal diversity to be positively correlated with yields
and negatively correlated with revenue variability in two studies in south-
ern Italy. Di Falco and Chavas (2006) point out increased crop diversity
may also reduce the likelihood of complete crop failure. Diversity is impor-
tant also for commercial farmers, since they are dependent on diversity in
the breeding pool, regardless of whether it is provided on or off farm.
Other examples of indirect values associated with diversity include
mycorrhiza, which are important for the functioning of ecosystems and
can be considered as a complementary input to timber production. They
represent an indirect use value. Silvicultural practices that eliminate myc-
orrhiza from the system will involve the loss of timber revenue. Although
mycorrhiza are not consumed themselves, they are essential for the growth
of many timber species that are harvested and they are necessary to
support the production process that produces goods and services that are
consumed directly.
Similarly, the importance of bird species used as a biological control
agent can be captured from increased timber sales associated with insect
pest reduction. Takekawa and Garton (1984) used the substitution
method to determine the value of a bird species, the evening grosbeak
(Hesperiphona vespertina) in controlling spruce budworm populations
affecting stands of Douglas fir (Pseudotsuga menziesii) in Washington.
They substituted the costs of insecticide to produce the same mortality
that birds cause and established that it would cost at least $1820 per square
km per year over a 100-year rotation.
Biodiversity in managed landscapes: policy issues

Earlier sections indicated why markets may fail to protect biodiversity and
give some examples of how economic tools can provide a useful means by
which to measure the non-benefits and costs associated with biodiversity
and thereby go some way towards dealing with market failure. A further
solution is to develop economic incentives and instruments that correct for
market failure. In the section that follows we consider the importance of
policy with respect to biodiversity conservation.
The aims of a society may be formulated within the framework of
national environmental policy. Policy can be regarded as ‘the compendium
of statements, laws and other actions concerning government’s intentions
for a particular human activity under its jurisdiction’ (Miller, 1999).
Objectives concerning natural resources are not necessarily static.
History indicates that environmental policies have changed progressively
with time in response to changes in society. This has led to changes in the
public demands placed on environmental resources. Human populations
are concerned with using environmental resources as a means of survival
but also increasingly to meet recreation and conservation goals.
In managed landscapes there is also public concern about the impor-
tance of ecological functions – water quality, biodiversity, aesthetic values
and international and national organizations are under increasing public
pressure to take action to develop economic incentives to protect public
values on privately managed land (WCED, 1987).
Economic instruments
In what follows we outline two types of policy instruments of relevance to
land managers – economic incentives and command and control regula-
tions. We discuss economic incentives first.
McNeely (1988, p. 39) has defined incentives as ‘an inducement, which

is specifically intended to incite or motivate governments, local people
and international organisations to conserve biological diversity’. The idea
behind economic incentives is to increase the cost of non-compliance with
environmental standards yet allow the producer the flexibility to employ
the least-cost method of meeting these standards. By increasing the cost of
non-compliance the producer has a private incentive to meet the standards
set by the policy instrument. One of the advantages of incentive systems is
that they are seen by economists as a cost-effective alternative to inflexible
command and control environmental regulations (Hanley et al., 1997).
However, in practice subsidies are much more widely used because of the
resistance to other instruments by the agricultural sector (Hanley and
Spash, 1993).
Many incentives are based on the level of opportunity costs or financial
costs forgone by the producer. For example, the financial costs of conser-
vation as estimated by Willis and Benson (1988) in the United Kingdom
are offered to farmers as compensation for not developing their land. This
is based on profits forgone under a management agreement. The current
financial cost is the difference between the value of the output (less inputs) of
the land under intensive management minus the value of output (less inputs)
under a conservation regime. A complete financial evaluation of conserva-
tion also needs to include administrative costs, legal fees, labour and mate-
rial costs for the maintenance of habitats (Willis and Benson, 1988).
Once the specific costs to the producer are known, policy instruments
can be formulated that are targeted at the producer and that persuade
producers to achieve the desired environmental objectives. Typically,
agri-environmental policies employ market-based instruments such as
subsidies that create economic incentives that allow individual producers
to choose freely to adjust their activities thereby producing an environ-
mental improvement (Barbier et al., 1994). Taxes as opposed to subsidies
are generally preferred by economists because the latter inject income and
lead to expansion of the sector under consideration. Subsidies can attract
new entrants that may lead to greater aggregate levels of environmental
damage and to other market distortions (Hanley et al., 1997).
An example of such a broad appraisal is agri-environment policy used
to maintain ecologically important habitats such as the Environmentally
Sensitive Area (ESA) scheme in the United Kingdom. Specific areas of
land providing habitats for valuable species are identified as conserva-
tion areas under which agricultural management practices are regulated.
Typically the policy is aimed at the farmer or forester to meet the desired
environmental objectives where, for example, farmers might be expected
to employ ‘traditional’ agronomic practices. The producer is then expected
to change his or her management methods in accordance with certain

regulations specified under a ‘management agreement’. Such management
agreements usually involve an identification of, for example, the farming
practices necessary to achieve environmental objectives and then stipu-
late how they should be put into practice. In order to specify guidelines
for ‘good environmental practice’, policy-makers need to understand the
relationship between management practices and the species, population
or community concerned; for example, the specific relationships between
farm management methods and the species composition of grasslands.
An example of this includes the use of stocking restrictions to encour-
age heather moorland in the United Kingdom. The model developed by
Simpson et al. (1998) crucially relates heather productivity and survival
to varying intensities of the management variable (in this case, the stock-
ing rate). Reductions in stocking rate can then be used to target farmers
who are able to manage heather sustainably under, say, a management
agreement.
Typically, agri-environment policy under a management agreement
involves reductions in farm intensity in exchange for compensatory pay-
ments. In order to do this a precise estimate of the changes in management
intensity to meet environmental objectives is required. This then enables
the specific costs to the producer to be calculated based on opportunity
cost pricing procedures. In the example described above, Simpson et al.
(1998) suggest that in order to meet conservation guidelines for heather
conservation ewe stocking rates will have to be reduced to between 13 and
91 per cent on Orkney and between 5 and 89 per cent on Shetland. They
report that such a reduction would in most cases result in major finan-
cial losses to farmers who would need to be compensated if they were to
comply with their recommendations.
This process of European agricultural reform has influenced the objec-
tives of the CAP, which have undergone significant changes in recent
years. The aims of the CAP are now strongly oriented towards environ-
mental conservation rather than agricultural productivity. The devel-
opment of these initiatives has provoked many EU countries to adopt
environmental policies specifically aimed at encouraging producers to
adopt less intensive agronomic and silvicultural practices (Hanley, 1995).
The status of environmental objectives therefore is increasingly recognized
to be as important as other goals such as rural income stability, employ-
ment and support for agricultural commodities. As a consequence, the
monitoring and evaluation of environmental policy includes an increas-
ing environmental component. The appraisal of agri-environment policy
needs to include an assessment of physical economic targets but also needs
to meet environmental objectives.
The second type of instrument includes command and control regu-

lations. Situations may occur where economic activities need to be
restrained in areas that are especially rich in biodiversity to protect the
resource for present and future generations and yet it may not be pos-
sible to control market behaviour using incentives. Command and control
environmental regulations may be used in such circumstances. Regulatory
control involves the direct limitation or reduction of activities that degrade
an environmental resource in accordance with some legislated or agreed
standard (Barbier et al., 1994).
This is especially important where development initiatives that threaten
biological diversity involve uncertainty. In the case of risk, as opposed to
uncertainty, it is possible to completely list the range of possible outcomes,
and to assign an estimated probability to each outcome. Given this infor-
mation, and preferences over risk and return, rational decision-making
is possible. In circumstances of uncertainty, however, where the range of
possible outcomes is unknown, it is not possible to determine the expected
profitability of a project. Although in the case of species extinctions a
probability can be attached to the loss of species, the total consequence of
this in terms of the loss of environmental services and ecosystem support
and duration of these effects cannot be known with certainty. Decision-
making in the presence of uncertainty relies not on rational comparison of
all options, but on adoption of some decision rule that has appealing prop-
erties (Common, 1995). It has been argued that a precautionary approach
to the conservation of biological resources should be adopted.
The policy of taking action before uncertainty about possible envi-
ronmental damage is resolved has been referred to as the ‘precautionary
principle’. One justification for this is that the costs of damage to bio-
logical resources may exceed the costs of preventative action (Jackson
and Taylor, 1992). Also, irreversible damage may occur such as species
extinctions. The emphasis is thus on avoiding potentially damaging
situations in the face of uncertainty over future outcomes. It has been
proposed for decisions taken over the Convention on Biodiversity and
has been used in conjunction with the Montreal Protocol (Myers, 1992;
Haigh, 1993).
Ciriacy-Wantrup (1968) and Bishop (1978) have proposed ‘the safe
minimum standards’ approach, which involves setting quantitative, and
qualitative limits for, say, the preservation of species and their habitats. A
programme is developed to maintain such limits unless the costs of doing
so are ‘unacceptably high’. Hanley et al. (1991) indicate that sites of special
scientific interest (SSSI) in the United Kingdom provide an example of
this approach in practice. These sites may be lost if the costs of conserva-
tion are prohibitive in terms of the government’s conservation budget, but
they are still protected regardless of any cost–benefit analysis having been
undertaken.
Conclusions
Rapidly increasing human populations and associated economic develop-
ment around the world have imposed real pressures on natural habitat
and its biodiversity. This is a subject of major concern to policy-makers
and the public at large because it is recognized that biodiversity loss could
seriously diminish the options open to future generations. All too often
market and policy imperfections obscure the social costs of managed
lands, giving rise to inefficient land use and biodiversity loss.
Protected areas represent a high cost solution to biodiversity conserva-
tion in many areas. They impose considerable costs on producers, limit
future development options, reduce the supply of market produce and
they fail to engage land managers in conservation initiatives.
Joint production of commercial goods and biodiversity in managed
landscapes represents an important alternative to reserves. Indeed there is
evidence to suggest that biodiversity can coexist in landscapes of economic
importance and that it is important in supporting productive processes
in managed areas. However, highly intensive managed systems may pose
a threat to biodiversity in some areas and it is vital that managers and
policy-makers work together to develop strategies to avoid such losses.
Policy-makers should contribute to this process by developing instru-
ments that internalize biodiversity values into market behaviour. This
will help to avoid intervention failure and perverse incentives that lead
to biodiversity loss, ensuring that biodiversity values are protected and
provided efficiently.
Uncertainty over the benefits and costs of biodiversity and its role in the
functioning of ecosystems point towards the need for a diversified strategy
that includes protected areas as well as privately managed land used for
production. In the absence of a concerted effort by policy-makers and
land managers, the opportunity to develop initiatives that include private
lands in such a strategy to achieve biodiversity conservation goals will be
missed.
Note
1. Biodiversity thus represents the diversity of all life as being a characteristic property of
nature, rather than a resource. The term also has a broader meaning for the set of organ-
isms themselves. For example, a biodiverse tropical rainforest, therefore, refers to the
quality or range of diversity within it.
Bibliography
Ahlén, I. (1975), ‘Winter habitats of moose and deer in relation to land use in Scandinavia’,
Viltrevy, 9(3), 1–88.
Alés, Fernandez R., A. Martin, F. Ortega and E.E. Alés (1992), ‘Recent changes in landscape
structure and function in a Mediterranean region of SW Spain (1950–1984)’, Landscape
Ecology, 7(1), 3–18.
Alonso, J.A., J.C. Alonso and R. Muñoz-Pulido (1990), ‘Areas de invernada de la grulla
común Grus grus en España’, in J.A. Alonso and J.C. Alonso (eds), Distribución y
demografia de la grulla común (Grus grus) en España, Madrid: ICONA.
Altieri, M. and M. Liebman (eds) (1986), ‘Insect, weed, and plant disease management
in multiple cropping systems’, in C. Francis, Multiple Cropping Systems, New York:
Macmillan.
Arrow, K.J. and K.C. Fisher (1974), ‘Environmental preservation, uncertainty and irrevers-
ibility’, Quarterly Journal of Economics, 88, 312–19.
Baldock, D., G. Beaufoy, G. Bennet and J. Clark (eds) (1993), Nature Conservation and
New Directions in the EC Common Agricultural Policy, Arnhem: Institute for European
Environmental Policy, HPC.
Barbier, E.B. (1994), ‘Valuing environmental functions: tropical wetlands’, Land Economics,
70, 155–73.
Barbier, E.B. (2000), ‘Valuing the environment as input: review of applications to mangrove–
fishery linkages’, Ecological Economics, 35(1), 47–61.
Barbier, E.B., J.C. Burgess and A. Markyanda (1991), ‘The economics of tropical deforesta-
tion’, Ambio, 20(21), 55–8.
Barbier, E.B., J.C. Burgess and C. Folke (1994), Paradise Lost? The Ecological Economics of
Biodiversity, London: Earthscan.
Bateman, I. and K. Willis (1995), Valuing Environmental Preferences: Theory and Practice of
the Contingent Valuation Method, Oxford, UK: Oxford University Press.
Bedell, T.E. (1973), ‘Botanical composition of a subclover-grass pasture as affected by single
and dual grazing by cattle and sheep’, Agronomy Journal, 65, 502–4.
Bell, F.W. (1997), ‘The economic value of saltwater marsh supporting marine recreational
fishing in the Southeastern United States’, Ecological Economics, 21(3), 243–54.
Benson, J. and K. Willis (1991), ‘The demand for forests for recreation’, in Forestry
Expansion: A Study of Technical, Economic and Ecological Factors, Edinburgh: Forestry
Commission.
Bishop, R. (1978), ‘Endangered species and uncertainty: the economics of a safe minimum
standard’, American Journal of Agricultural Economics, 60(1), 10–18.
Bokdam, J. and M. Gleichman (2000), ‘Effects of grazing by free ranging cattle on vegetation
dynamics in a continental north-west European heathland’, Journal of Applied Ecology,
37(5), 415–31.
Borchert, M.I., F.W. Davis, J. Michaelsen and L.D. Oyler (1989), ‘Interactions of factors
affecting seedling recruitment of blue oak (Quercus douglasii) in California’, Ecology,
70(2), 389–404.
Bowes, M. and J. Krutilla (1989), Multiple Use Management: The Economics of Public
Forestlands, Washington, DC: Resources for the Future.
Bowker, J. and J.R. Stoll (1988), ‘Use of dichotomous choice nonmarket methods to value
the whooping crane resource’, American Journal of Agricultural Economics, 70(2), 372–81.
Boyle, K.J. and R.C. Bishop (1987), ‘Valuing wildlife in benefit–cost analyses: a case study
involving endangered species’, Water Resources Research, 23(5), 943–50.
Bradshaw, R. and F.J.G. Mitchell (1999), ‘The palaeoecological approach to reconstructing
former grazing–vegetation interactions’, Forest Ecology and Management, 120(1), 3–12.
Brookshire, D., B. Ives and W. Schulze (1976), ‘The valuation of aesthetic preferences’,
Journal of Environmental Economics and Management, 3(4), 325–46.
Brown, G.M. (1990), ‘Valuation of genetic resources’, in G.H. Orians, G.M. Brown, W.E.
Kunin and J.E. Swierbinski (eds), The Preservation and Valuation of Biological Resources,
Seattle: University of Washington Press.
Brown, G.M. and W. Henry (1993), ‘The economic value of elephants’, in E.B. Barbier (ed.),
Economics and Ecology: New Frontiers and Sustainable Development, London: Chapman
and Hall.
Brown, G.M., D. Layton and J. Lazo (1994), ‘Valuing habitat and endangered species’,
Discussion Paper No. 94-1 (January), Institute for Economic Research, University of
Washington.
Callaway, R.M. (1992), ‘Effect of shrubs on recruitment of Quercus douglasii and Quercus
lobata in California’, Ecology, 73(6), 2118–28.
Campbell, R.W. and R.J. Sloan (1976), ‘Influence of behavioural evolution on gypsy moth
pupal survival in sparse populations’, Environmental Entomology, 5(6), 1211–17.
Carrascal, L.M., L.M. Bautista and E. Lázaro (1993), ‘Geographical variation in the density
of the white stork Ciconia ciconia in Spain: influence of habitat structure and climate’,
Biological Conservation, 65(11), 83–7.
Casey, D. and D. Hein (1983), ‘Effects of heavy browsing on a bird community in deciduous
forest’, Journal of Wildlife Management, 47(3), 351–96.
Ciriacy-Wantrup, S. (1968), Resource Conservation: Economics and Policies, Berkeley:
University of California.
Clawson, M. and J. Knetsch (1966), Economics of Outdoor Recreation, Baltimore: Johns
Hopkins University Press.
Clout, M.N. and P.D. Gaze (1984), ‘Effects of plantation forestry on birds in New Zealand’,
Journal of Applied Ecology, 21(3), 795–815.
Coase, R. (1960), ‘The problem of social cost’, Journal of Law and Economics, 3(1), 1–44.
Common, M. (1995), Sustainability and Policy: Limits to Economics, Cambridge, UK:
Cambridge University Press.
Common, M. and C. Perrings (1992), ‘Towards an ecological economics of sustainability’,
Ecological Economics, 6(1), 7–34.
Conway, A., A. McLoughlin and W.E. Murphy (1972), ‘Development of a cattle and sheep
farm’, Animal Management Series No. 2, An Foras Taluntais, Dublin.
Costanza, R., W.M. Kemp and W.R. Boynton (1993), ‘Predictability, scale and biodiversity
in coastal and estuarine systems’, in C. Perrings, K.-G. Mäler, C. Folke, B.O. Jansson and
C.S. Holling (eds), Biodiversity Loss: Ecological and Economic Issues, Cambridge, UK:
Cousins, S.H. (1991), ‘Species diversity measurement: choosing the right index’, Tree, 6(6),
190–2.
Crozier, R.H. (1992), ‘Genetic diversity and the agony of choice’, Biological Conservation,
61(1), 11–15.
Cumming, D.H.M. (1993), ‘Multispecies systems: progress, prospects and challenges in sus-
taining range animal production and biodiversity in east and southern Africa’, Proceedings
of the 7th World Conference on Animal Production, Edmonton, Alberta.
Daily, G.C. (ed.) (1997), Nature’s Services: Societal Dependence on Natural Ecosystems,
Washington, DC: Island Press.
Daily, G.C., P.R. Ehrlich and N.M. Haddad (1993), ‘Double keystone bird in a keystone
species complex’, Proceedings of the Natural Academy of Sciences, 90(2), 592–4.
Daily, G.C., T. Soderqvist, S. Aniyar, K. Arrow, P. Dasgupta, P.R. Ehrlich, C. Folke,
A. Jansson, B.O. Jansson, N. Kautsky, S. Levin, J. Lubchenco, K.G. Maler, D. Simpson,
D. Starrett, D. Tilman and B. Walker (2000), ‘The value of nature and the nature of value’,
Science, 289(5478), 395–6.
Daily, G.C., P.R. Ehrlich and G.A. Sanchez-Azofeifa (2001), ‘Countryside biogeography:
use of human dominated habitats by the avifauna of southern Costa Rica’, Ecological
Applications, 11(1), 1–13.
Dambach, C.A. (1944), ‘A ten-year ecological study of adjoining grazed and ungrazed wood-
lands in northeastern Ohio’, Ecological Monographs, 14(3), 69–105.
De Miguel, J.M., M.A. Rodriguez and A. Gómez-Sal (1997), ‘Determination of animal
behavior–environment relationships by correspondence analysis’, Journal of Range
Management, 50(1), 85–93.
Debussche, M. and P. Isenmann (1994), ‘Bird-dispersed seed rain and seedling establishment
in patchy mediterranean vegetation’, Oikos, 69(33), 414–26.
DeGraaf, R.M., W.M. Healy and R.T. Brooks (1991), ‘Effects of thinning and deer brows-
ing on breeding birds in New England oak woodlands’, Forest Ecology and Management,
41(3–4), 179–191.
Di Falco, S. and J.-P. Chavas (2006), ‘Crop biodiversity, farm productivity and the man-
agement of environmental risk in rainfed agriculture’, European Review of Agricultural
Economics, 33(3), 289–314.
Di Falco, S. and C. Perrings (2003), ‘Crop genetic diversity, productivity and stability of
agroecosystems: a theoretical and empirical investigation’, Scottish Journal of Political
Economy, 50(2), 207–16.
Di Falco, S. and C. Perrings (2005), ‘Crop biodiversity, risk management and the implica-
tions of agricultural assistance’, Ecological Economics, 55(4), 459–66.
Diamond, P.A. and J.A. Hausman (1994), ‘Contingent valuation: is some number better
than no number?’, Journal of Economic Perspectives, 8(4), 19–44.
Díaz, M., E. González, R. Munoz-Pulido and M.A. Naveso (1996), ‘Habitat selection pat-
terns of common cranes Grus grus wintering in holm oak Quercus ilex dehesas of central
Spain: effects of human management’, Biological Conservation, 75(22), 119–23.
Dobzhansky, T. (1970), Genetics of the Evolutionary Process, New York: Columbia
University Press.
Edwards, K.J., G. Whittington and W. Ritchie (2005), ‘The possible role of humans in
the early stages of machair evolution: palaeoenvironmental investigations in the Outer
Hebrides, Scotland’, Journal of Archaeological Science, 32(3), 435–49.
Ellis, G.M. and A.C. Fisher (1987), ‘Valuing the environment as an input’, Journal of
Environmental Management, 85(2), 149–56.
Elton, C. (1958), The Ecology of Invasions of Animals and Plants, London: Chapman and
Hall.
Erdelen, M. (1984), ‘Bird communities and vegetation structure: I. Correlations and com-
parisons of simple and diversity indices’, Oecologia, 61(2), 277–84.
Espelta, J.S., M. Riba and J. Retana (1995), ‘Patterns of seedling recruitment in West-
Mediterranean Quercus ilex forests influenced by canopy development’, Journal of
Vegetation Science, 6(4), 465–72.
Faliñski, J.B. (1986), Vegetation Dynamics in Temperate Lowland Forests, Dordrecht: Junk
Publishers.
FAO (2004), http:www.faostat.fao.org/corp/statistics/en, accessed 13 August 2009.
Faraco, A.M., F. Fernandez and J.M. Moreno (1993), ‘Post-fire dynamics of pine wood-
lands and shrublands in the Sierra de Gredos’, in L. Traubaud and R. Prodon (eds), Fire in
Mediterranean Ecosystems, Ecosystems Research Report No. 5, pp. 101–13.
Fay, P.A. and D.C. Hartnett (1991), ‘Constraints on growth and allocation patterns of
Silphium integrifolium (Asteraceae) caused by a cynipid gall wasp’, Oecologia, 88(2),
243–50.
Ford, H.A. (1987), ‘Bird communities on habitat islands in England’, Bird Study, 34(3),
205–18.
Fox, L.R. and P.A. Morrow (1992), ‘Eucalypt responses to fertilization and reduced her-
bivory’, Oecologia, 89(2), 214–22.
French, D.D., G.R. Miller and R.P. Cummins (1997), ‘Recent development of high-altitude
Pinus sylvestris scrub in the Northern Cairngorm mountains, Scotland’, Biological
Conservation, 79(2–3), 133–44.
Fuller, R.J., D. Hill and G.M. Tucker (1991), ‘Feeding the birds down on the farm: perspec-
tives from Britain’, Ambio, 20(6), 232–7.
Garrod, G.D. and K.G. Willis (1997), ‘The non-use benefits of enhancing forest biodiversity:
a contingent ranking study’, Ecological Economics, 21(1), 45–61.
González Bernáldez, F. (1991), ‘Ecological consequences of the abandonment of traditional
land use systems in central Spain’, in J. Baudry and R.G.H. Bunce (eds), Land Abandonment
and its Role in Conservation, Options Mediterranéennes Ser. A15, pp. 23–9.
Gourlay, D. (1996), ‘Loch Lomond and Stewartry ESAs: a study of public perceptions of
policy’, unpublished PhD dissertation, School of Agriculture, University of Aberdeen.
Grainger, A. (1990), The Threatening Desert – Controlling Desertification, London: Earthscan.
Griffin, J.R. (1971), ‘Oak regeneration in the upper Carmel valley, California’, Ecology,
52(5), 862–8.
Haigh, N. (1993), The Precautionary Principle and British Environmental Policy, London:
Institute for European Environmental Policy.
Hanemann, W.M. (1994), ‘Valuing the environment through contingent valuation’, Journal
of Economic Perspectives, 8(4), 19–44.
Hanley, N. (1995), Rural Amenities and Rural Development: Empirical Evidence, Synthesis
Report to the Rural Development Programme, OECD, Paris.
Hanley, N. and S. Craig (1991), ‘Wilderness development decisions and the Krutilla–Fisher
model: the case of Scotland’s flow country’, Ecological Economics, 4(2), 145–64.
Hanley, N. and C. Spash (1993), Cost–Benefit Analysis and the Environment, Aldershot, UK
and Brookfield, VT, USA: Edward Elgar.
Hanley, N., A. Munro, D. Jamieson and D. Ghosh (1991), ‘Environmental economics and
sustainable development in nature conservation’, unpublished NCC Report.
Hanley, N., J.F. Shogren and B. White (1997), Environmental Economics: In Theory and
Practice, London: Macmillan.
Hausman, J.A. (ed.) (1993), Contingent Valuation: A Critical Assessment, Contributions to
Economic Analysis, Amsterdam: North Holland Publishers.
Hernandez, M.P.G. and F.J. Silva-Pando (1996), ‘Grazing effects of ungulates in a Galician
oak forest (Northwest Spain)’, Forest Ecology and Management, 88(1–2), 65–70.
Herrera, J. (1995), ‘Acorn predation and seedling production in a low-density population of
cork oak (Quercus-suber L.)’, Forest Ecology and Management, 76(1–3), 197–201.
Hester, A.J., F.J.G. Mitchell and K.J. Kirby (1996), ‘Effects of season and intensity of
sheep grazing on tree regeneration in a British upland woodland’, Forest Ecology and
Management, 88(1–2), 99–106.
Hoffman, M.T. (1988), ‘Rationale for Karoo grazing systems: criticisms and research impli-
cations’, South African Journal of Science, 84, 556–9.
Hoffman, M.T. and R.M. Cowling (1990), ‘Vegetation change in the semi-arid eastern Karoo
over the last 200 years: an expanding Karoo – fact or fiction?’, South African Journal of
Science, 86, 286–94.
Holling, C.S. (1973), ‘Resilience and stability of ecological systems’, Annual Review of
Ecological Systems, 4(1), 1–24.
Holling, C.S. (ed.) (1978), Adaptive Environmental Assessment and Management, Chichester,
UK: John Wiley and Sons.
Holling, C.S. (1986), ‘Resilience of ecosystem: local surprise and global change’, in E.C.
Clark and R.E. Munn (eds), Sustainable Development of the Biosphere, Cambridge, UK:
Holling, C.S. (1988), ‘Temperate forest insect outbreaks, tropical deforestation and migra-
tory birds’, Memoirs of the Entomological Society of Canada, No. 146, 21–32.
Holling, C.S. (1992), ‘Cross-scale morphology, geometry and dynamics of ecosystems’,
Ecological Monographs, 62(4), 47–52.
Holling, C.S., D.W. Schindler, B.W. Walker and J. Roughgarden (1994), ‘Biodiversity in
the functioning of ecosystems: an ecological primer and synthesis’, in C. Perrings, K.-G.
Mäler, C. Folke, B.O. Jansson and C.S. Holling (eds), Biodiversity Loss: Ecological and
Economic Issues, Cambridge, UK: Cambridge University Press.
Holmes, R.T., J.C. Schultz and P. Nothnagle (1979), ‘Bird predation on forest insects: an
exclosure experiment’, Science, 206(4417), 462–3.
Hulme, P.D., R.J. Pakeman, J.M. Torvell, J.M. Fisher and I.J. Gordon (1999), ‘The effects
of controlled sheep grazing on the dynamics of upland Agrostis-Festuca grassland’, Journal
of Applied Ecology, 36(66), 886–900.
Hutchinson, G.E. (1961), ‘The paradox of the plankton’, American Naturalist, 95(882),
137–47.
IUCN (1988), ‘General Assembly of the International Union for Conservation of Nature
and Natural Resources’, Costa Rica.
Jackson, T. and P.J. Taylor (1992), ‘The precautionary principle and the prevention of
marine pollution’, Chemistry and Ecology, 71(1–4), 123–34.
James, F.C. and N.O. Wamer (1982), ‘Relationships between temperate forest bird commu-
nities and vegetation structure’, Ecology, 63(1), 159–71.
Jane, G.T. (1994), ‘The impact of browsing animals on the stand dynamics of monotypic
mountain beech (Nothofagus solandri) forests in Canterbury, New Zealand’, Australia
Journal of Botany, 42(22), 113–24.
Jeltsch, F., S.J. Milton, W.R.J. Dean and N. van Rooyen (1997), ‘Analysing shrub encroach-
ment in the southern Kalahari: a grid-based modelling approach’, Journal of Applied
Ecology, 34(6), 1497–1509.
Joffre, R. and S. Rambal (1993), ‘How tree cover influences the water balance of
Mediterranean rangelands’, Ecology, 74(2), 570–82.
Joffre, R., J. Vacher, C. de los Llanos and G. Long (1988), ‘The dehesa: an agrosilvopastoral
system of the Mediterranean region with special reference to the Sierra Morena area of
Spain’, Agroforestry Systems, 6(1), 71–96.
Jorritsma, I.T.M., A.F.M. van Hees and G.M.J. Mohren (1999), ‘Forest development in
relation to ungulate grazing: a modelling approach’, Forest Ecology and Management,
120(1–3), 23–4.
Just, R.E. and R.D. Pope (1978), ‘Stochastic representation of production functions and
econometric implications’, Journal of Econometrics, 7(1), 67–86.
Kienast, F., J. Fritschi, M. Bissegger and W. Abderhalden (1999), ‘Modelling successional
patterns of high-elevation forests under changing herbivore pressure – responses at the
landscape level’, Forest Ecology and Management, 120(1–3), 35–46.
Kingery, J.L. and R.T. Graham (1991), ‘The effect of cattle grazing on ponderosa pine regen-
eration’, Forestry Chronicle, 67(3), 245–8.
Köhler-Rollefson, I. (1993), ‘Traditional pastoralists as guardians of biological diversity’,
Indigenous Knowledge and Development Monitor, 1(3), 14–16.
Kruess, A. and T. Tscharntke (1994), ‘Habitat fragmentation, species loss and biological
control’, Science, 264(5165), 1581–84.
Krutilla, J.V. (1967), ‘Conservation reconsidered’, American Economic Review, 57(4), 778–86.
Lindberg, K. (1991), Policies for Maximizing Nature Tourism’s Ecological and Economic
Benefits, Washington, DC: World Resources Institute.
Linder, P., B. Elfving and O. Zackrisson (1997), ‘Stand structure and successional trends in
virgin boreal forest reserves in Sweden’, Forest Ecology and Management, 98(1), 17–33.
Londo, G. (1990), ‘Conservation and management of semi-natural grasslands in
Northwestern Europe’, in U. Bohn and R. Neuhäusl (eds), Vegetation and Flora of
Temperate Zones, The Hague: Academic Publishing.
Long, J., E. Cromwell and K. Gold (2000), On-farm Management of Crop Diversity: An
Introductory Bibliography, London: Overseas Development Institute for ITDG.
Lovejoy, T.E. (1988), ‘Diverse considerations’, in E.O. Wilson and F.M. Peter (eds), Issues in
Risk Assessment, Washington, DC: National Academy Press.
Lynch, J.F. and D.F. Whigham (1984), ‘Effects of forest fragmentation on breeding bird
communiites in Maryland, USA’, Biological Conservation, 28(44), 287–324.
MacArthur, R.H. (1955), ‘Fluctuations of animal populations and a measure of community
stability’, Ecology, 36(3), 533–6.
MacArthur, R.H. and J.W. MacArthur (1961), ‘On bird species diversity’, Ecology, 42(3),
594–8.
MacArthur, R.H. and E.O. Wilson (1967), The Theory of Island Biogeography, Princeton:
Princeton University Press.
Main, A.F. (1982), ‘Rare species: precious or dross?’, in R.H. Groves and W.D.L. Ride (eds),
Species at Risk: Research in Australia, Canberra: Australian Academy of Science.
Maquis, R.J. and C.J. Whelan (1994), ‘Insectivorous birds increase growth of white oak
through consumption of leaf-chewing insects’, Ecology, 75(7), 2007–14.
Marañon, M. (1986), ‘Plant species richness and canopy effect in the savannah-like dehesa of
south-west Spain’, Ecologia Mediterranea, 12(1), 131–41.
Marrs, R.H. and D. Welch (1991), Moorland Wilderness: The Potential Effects of
Removing Domestic Livestock, Particularly Sheep, ITE Report to the Department of the
Environment.
Mate, I.D. (1992), ‘The theoretical development of machair in the Hebrides’, Scottish
Geographical Magazine, 108(1), 35–8.
Mattson, W.J. and N.D. Addy (1975), ‘Phytophagous insects as regulators of forest primary
production’, Science, 190(4214), 515–22.
May, R.M. (1973), Stability and Complexity in Model Ecosystems, Princeton: Princeton
University Press.
McNaughton, S.J. (1984), ‘Grazing lawns: animals in herds, plant form and co-evolution’,
American Naturalist, 124(6), 863–86.
McNaughton, S.J. (1985), ‘Ecology of a grazing ecosystem: the Serengeti’, Ecological
Monographs, 55(3), 259–94.
McNeeley, J.A. (1988), Economics and Biological Diversity, Gland, Switzerland: IUCN.
McNeeley, J.A. (1993), ‘Economic incentives for conserving biodiversity: lessons for Africa’,
Ambio, 22(2–3), 144–50.
Metrick, A. and M.L. Weitzman (1994), ‘Patterns of behavior in biodiversity preservation’,
Policy Research Working Paper No. 1358, Environment, Infrastructure, and Agriculture
Division, The World Bank, Washington, DC.
Mill, G.A., T.M. van Rensburg, S. Hynes and C. Dooley (2007), ‘Preferences and mul-
tiple use forest management in Ireland: a contingent valuation approach’, Ecological
Economics, 60(3), 642–53.
Miller, H.G. (1999), Forest Policy: The International and British Dimensions, Department of
Forestry, University of Aberdeen.
Miller, K.R. (1996), ‘Conserving biodiversity in managed landscapes’, in R.C. Szaro and
D.W. Johnston (eds), Biodiversity in Managed Landscapes: Theory and Practice, Oxford,
UK: Oxford University Press.
Miller, T.E. (1982), ‘Community diversity and interactions between the size and frequency of
disturbance’, American Naturalist, 120(4), 533–42.
Mitchell, F.J.G. and E. Cole (1998), ‘Reconstruction of long-term successional dynamics of
temperate woodland in Bialowieza forest, Poland’, Journal of Ecology, 86(6), 1042–59.
Mitchell, R.C. and R.T. Carson (1989), Using Surveys to Value Public Goods: The Contingent
Valuation Method, Washington, DC: Resources for the Future.
Montgomery, C.A., G.M. Brown and D.M. Adams (1994), ‘The marginal cost of species pres-
ervation: the northern spotted owl’, Journal of Environmental Economics and Management,
26(2), 111–28.
Moore, A.D. (1990), ‘The semi-Markov process: a useful tool in the analysis of vegetation
dynamics for management’, Journal of Environmental Management, 30(2), 111–30.
Morgan, R.K. (1991), ‘The role of protective understorey in the regeneration system of a
heavily browsed woodland’, Vegetatio, 92(2), 119–32.
Morris, R.F., W.F. Cheshire, C.A. Miller and D.G. Mott (1958), ‘The numerical response
of avian and mammalian predators during a gradation of the spruce budworm’, Ecology,
39(3), 487–94.
Myers, N. (1992), ‘Population/environment linkages: discontinuities ahead’, Ambio, 21(1),
116–18.
Naveh, Z. (1994), ‘The role of fire and its management in the conservation of Mediterranean
ecosystems and landscapes’, in J.M. Moreno and W.C. Oechel (eds), The Role of Fire in
Mediterranean-type Ecosystems, New York: Springer Verlag.
Naveh, Z. and R.H. Whittaker (1979), ‘Structural and floristic diversity of shrublands
and woodlands in northern Israel and other Mediterranean areas’, Vegetatio, 41(3),
171–90.
Nolan, T. and J. Connolly (1977), ‘Mixed stocking by sheep and steers – a review’, Herbage
Abstracts, 47(11), 367–79.
O’Neill, R.V., D.L. DeAngelis, J.B. Waide and T.F.H. Allen (1986), A Hierarchical Concept
of Ecosystems, Princeton: Princeton University Press.
Opdam, P., G. Rijsdijk and F. Hustings (1985), ‘Bird communities in small woods in an
agricultural landscape: effects of area and isolation’, Biological Conservation, 34(4),
333–51.
Pearce, D.W. and R.K. Turner (1990), Economics of Natural Resources and the Environment,
Hemel Hempstead, UK: Harvester Wheatsheaf.
Perman, R., Y. Ma, J. McGilvray and M. Common (2003), Natural Resource and
Environmental Economics, New York: Pearson Education, Longman.
Peterken, G.F. and C.R. Tubbs (1965), ‘Woodland regeneration in the New Forest,
Hampshire, since 1650’, Journal of Applied Ecology, 2(1), 159–70.
Pezzey, J. (1989), ‘Economic analysis of sustainable growth and sustainable development’,
Environment Department Working Paper No. 15, The World Bank, Washington, DC.
Pignatti, S. (1978), ‘Evolutionary trends in Mediterranean flora and vegetation’, Vegetatio,
37(3), 175–85.
Pimm, S.L. (1984), ‘The complexity and stability of ecosystems’, Nature, 307(26), 321–6.
Pimmental, D., U. Stachow, D.A. Takacs, H.W. Brubaker, A.R. Dumas, J.J. Meaney, J.A.
O’Neill, D.E. Onsi and D.B. Corzilius (1992), ‘Conserving biological diversity in agricul-
tural/forestry systems’, Bioscience, 42(5), 354–62.
Polasky, S., E. Nelson, E. Lonsdorf, P. Fackler and A. Starfield (2005), ‘Conserving species
in a working landscape: land use with biological and economic objectives’, Ecological
Applications, 15(4), 1387–1401.
Portney, P.R. (1994), ‘The contingent valuation debate: why economists should care’,
Journal of Economic Perspectives, 8(4), 3–18.
Power, M.E., D. Tilman, J.A. Estes, B.A. Menge, W.J. Bond, L.S. Mills, G. Daily, J.C.
Castilla, J. Lubchenko and R.T. Paine (1996), ‘Challenges in the quest for keystones’,
BioScience, 46(8), 609–20.
Randall, A., B. Ives and C. Eastman (1974), ‘Bidding games for the valuation of aesthetic
environmental improvements’, Journal of Environmental Economics and Management,
1(2), 132–49.
Ratcliffe, P.R. (1993), Biodiversity in Britain’s Forests, Edinburgh: The Forest Authority.
Raven, R.H., K. Beese, T. Eisner, N. Morin, T. Duncan, S.L.A. Hobbs, J. Hodges, W.F.
Rall, R.R. Colwell and J. Swings (1992), ‘Biotechnology and genetic resources: United
States–Commission of the European Communities Workshop’, 21–22 October, Airlie,
Virginia, USA; Brussels, Belgium: Commission of the European Communities.
Reid, W.V. (1996), ‘Beyond protected areas: changing perceptions of ecological management
objectives’, in R.C. Szaro and D.W. Johnston (eds), Biodiversity in Managed Landscapes:
Theory and Practice, Oxford, UK: Oxford University Press.
Reinhammar, L.-G. (1995), ‘Evidence for two distinct species of Pseudorchis (Orchidaceae)
in Scandinavia’, Nordic Journal of Botany, 15(5), 469–81.
Relva, M.A. and T.T. Veblen (1998), ‘Impacts of introduced large herbivores on Austrocedrus
chilensis forests in northern Patagonia, Argentina’, Forest Ecology and Management,
108(1–2), 27–40.
Repetto, R. (1989), ‘Economic incentives for sustainable production’, in G. Schramme and
J.J. Warford (eds), Environmental Management and Economic Development, Baltimore:
Johns Hopkins for the World Bank.
Roland, M.A. and P.D. Taylor (1997), ‘Insect parasitoid species respond to forest structure
at different spatial scales’, Nature, 386(6636), 710–13.
Rosenzweig, M.L. (2003), Win-Win Ecology: How the Earth’s Species Can Survive in the
Midst of Human Enterprise, Oxford, UK: Oxford University Press.
Schemske, D.W. and N. Brokaw (1981), ‘Treefalls and the distribution of understorey birds
in a tropical forest’, Ecology, 62(4), 938–45.
Schindler, D.W. (1990), ‘Experimental perturbations of whole lakes as tests of hypotheses
concerning ecosystem structure and function’, Proceedings of 1987 Crafoord Symposium,
Oikos, 57(1), 25–41.
Schindler, D.W., S.E.M. Kasian and R.H. Hesslein (1989), ‘Biological impoverishment
in lakes of the northeastern United States from acid rain’, Environmental Science and
Technology, 23(5), 573–80.
Schmida, A. and M.V. Wilson (1985), ‘Biological determinants of species diversity’, Journal
of Biogeography, 12(4), 1–20.
Scholes, R.J. and B.H. Walker (1993), An African Savanna: Synthesis of the Nylsvlei Study,
Silva, J. (1987), ‘Responses of savanna to stress and disturbance: species dynamics’, in B.H.
Walker (ed.), Determinants of Tropical Savannas, Oxford, UK: IRL Press.
Simberloff, D., B.J. Brown and S. Lowrie (1978), ‘Isopod and insect root borers may benefit
Florida mangroves’, Science, 201(4356), 630–32.
Simpson, I.A., A.H. Kirkpatrick, L. Scott, J.P. Gill, N. Hanley and A.J. MacDonald (1998),
‘Application of a grazing model to predict heather moorland utilization and implications
for nature conservation’, Journal of Environmental Management, 54(3), 215–31.
Smale, M., J. Hartell, P.W. Heisey and B. Senauer (1998), ‘The contribution of genetic
resources and diversity to wheat production in the Punjab of Pakistan’, American Journal
of Agricultural Economics, 80(3), 482–93.
Solow, A., S. Polasky and J. Broadus (1993), ‘On the measurement of biological diversity’,
Journal of Environmental Economics and Management, 24(1), 60–68.
Soulé, M.E. and B.A. Wilcox (eds) (1980), Conservation Biology, an Evolutionary Ecological
Perspective, Sunderland, MA: Sinauer Associates.
Staines, B.W. (1995), ‘The impact of red deer on the regeneration of native pinewoods’, in
Forestry Commission, Our Pinewood Heritage, Conference proceedings, Inverness, 1994,
Farnham, UK: Forestry Commission, RSPB, SNH.
Staines, B.W., R. Balharry and D. Welch (1995), ‘The impact of red deer and their manage-
ment on the natural heritage in the uplands’, in D.B.A. Thompson, A.J. Hester and M.B.
Usher (eds), Heaths and Moorlands: Cultural Landscapes, Edinburgh: Scottish Natural
Heritage.
Sternberg, M., M. Gutman, A. Perevolotsky, E.D. Ungar and J. Kigel (2000), ‘Vegetation
response to grazing management in a Mediterranean herbaceous community: a functional
group approach’, Journal of Applied Ecology, 37, 224–37.
Stoll, J.R. and L.A. Johnson (1984), ‘Concepts of value, non-market valuation and the case
of the whooping crane’, Transactions of the Forty Ninth American Wildlife and Natural
Resources Conference, 49, 382–93.
Stowe, T.J. (1987), ‘The management of sessile oakwoods for pied flycatchers’, RSPB
Conservation Review, 1, 78–83.
Sumner, D.R., B. Doupnik and M.G. Boosalis (1981), ‘Effects of reduced tillage and multiple
cropping on plant diseases’, Annual Review of Phytopathology, 19(3755), 167–87.
Sun, D., G.R. Dickinson and A.L. Bragg (1997), ‘Effect of cattle grazing and seedling size
on the establishment of Araucaria cunninghamii in a silvo-pastoral system in Northeast
Australia’, Journal of Environmental Management, 49(4), 435–44.
Sykes, J.M. and A.D. Horrill (1979), ‘Regeneration of native pinewoods’, Institute of
Terrestrial Ecology Annual Report 1978, pp. 103–6.
Takekawa, J.Y. and E.O. Garton (1984), ‘How much is an evening grosbeak worth?’, Journal
of Forestry, 82(7), 426–8.
Takekawa, J.Y., E.O. Garton and L.A. Langelier (1982), ‘Biological control of forest insect
outbreaks: the use of avian predators’, Transactions of the 47th North American Wildlife
Conference, pp. 393–409.
Telleria, J.L. and L.M. Carrascal (1994), ‘Weight–density relationships between and within
bird communities: implications of niche space and vegetation structure’, The American
Naturalist, 143, 1083–92.
Telleria, J.L. and T. Santos (1995), ‘Effects of forest fragmentation on a guild of wintering
passerines: the role of habitat selection’, Biological conservation, 71(1), 61–7.
Telleria, J.L., T. Santos, A. Sanchez and A. Galarza (1992), ‘Habitat structure predicts bird
diversity distribution in Iberian forests better than climate’, Bird Study, 39(1), 63–8.
Terborgh, J. (1985), ‘Habitat selection in Amazonian birds’, in M.L. Cody (ed.), Habitat
Selection in Birds, Orlando: Academic Press.
Tilman, D. (1996), ‘Biodiversity: population versus ecosystem stability’, Ecology, 77(3),
350–63.
Tilman, D. and J.A. Downing (1994), ‘Biodiversity and stability in grasslands’, Nature, 367,
363–6.
Tilman, D., J. Knops, D. Wedin, P. Reich, M. Ritchie and E. Siemann (1997), ‘The influ-
ence of functional diversity and composition on ecosystem processes’, Science, 277(5330),
1300–02.
Tinbergen, L. (1960), ‘The natural control of insects in pinewoods. I. Factors influencing the
intensity of predation by songbirds’, Archives Neerlandaises de Zoologie, 13(3), 265–43.
Tucker, G.M. and M.F. Heath (1994), Birds in Europe: Their Conservation Status, Cambridge,
UK: BirdLife International.
Turner, R.K. (ed.) (1993), Sustainable Environmental Economics and Management, London:
Belhaven Press.
UNEP (1995), Global Biodiversity Assessment, Cambridge, UK: Cambridge University
Press.
Valderrábano, J. and L. Torrano (2000), ‘The potential for using goats to control Genista
scorpius shrubs in European black pine stands’, Forest Ecology and Management, 126(33),
377–83.
Van Dorp, D. and P.F.M. Opdam (1987), ‘Effects of patch size, isolation and regional abun-
dance on forest bird communities’, Landscape Ecology, 1(1), 59–73.
Vane-Wright, R.I., C.J. Humphries and P.H. Williams (1991), ‘What to protect? Systematics
and the agony of choice’, Biological Conservation, 55(3), 235–54.
Van Hees, A.F.M., A.T. Kuiters and P.A. Slim (1996), ‘Growth and development of silver
birch, pedunculate oak and beech as affected by deer browsing’, Forest Ecology and
Management, 88(1–2), 55–63.
Van Rensburg, T.M., G.A. Mill, M. Common and J.C. Lovett (2002), ‘Preferences and mul-
tiple use forest management’, Ecological Economics, 43(2–3), 231–44.
Walker, B.H. (1988), ‘Autoecology, synecology, climate and livestock as agents of range-
lands dynamics’, Australian Range Journal, 10(2), 69–75.
Walker, B.H. (1992), ‘Biodiversity and ecological redundancy’, Conservation Biology, 6,
18–23.
Walker, B.H., D. Ludwig, C.S. Holling and R.M. Peterman (1981), ‘Stability of semi-arid
savanna grazing systems’, Ecology, 69(2), 473–498.
Walker T.S., R.P. Singh and N.S. Jodha (1983), Dimensions of Farm-level Diversification
in the Semi-arid Tropics of Rural South India, Economic Program Progress Report, 51
ICRISAT, Patancheru, India.
WCED (1987), Our Common Future, The World Commission on Environment and
Development, Oxford: Oxford University Press.
Weber, G.E., F. Jeltsch, N. van Rooyen and S.J. Milton (1998), ‘Simulated long-term veg-
etation response to grazing heterogeneity in semi-arid rangelands’, Journal of Applied
Ecology, 35(5), 687–99.
Weisbrod, B. (1964), ‘Collective consumption services of individual consumption goods’,
Quarterly Journal of Economics, 78(3), 471–7.
Weitzman, M.L. (1998), ‘The Noah’s ark problem’, Econometrica, 66(6), 1279–98.
Weslien, J. and M. Schroeder (1999), ‘Population levels of bark beetles and associated insects
in managed and unmanaged spruce stands’, Forest Ecology and Management, 115(2–3),
267–75.
Westoby, M., B.H. Walker and I. Noy-Meir (1989), ‘Opportunistic management for range-
lands not at equilibrium’, Journal of Range Management, 42(4), 266–74.
Whiteman, A. (1991), A Cost–Benefit Analysis of Forest Replanting in East England,
Edinburgh: Forestry Commission Development Division.
Whittaker, R.H. (1977), ‘Evolution of species diversity in land communities’, Evolutionary
Biology, 10(1), 1–67.
Willis, K.G. (1991), ‘The recreational value of the Forestry Commission estate in Great
Britain: a Clawson–Knetsch travel cost analysis’, Scottish Journal of Political Economy,
38(1), 58–75.
Willis, K.G. and J.F. Benson (1988), ‘Financial and social costs of management agreements
for wildlife conservation in Britain’, Journal of Environmental Management, 26(1), 43–63.
Willson, M.F. (1974), ‘Avian community organization and habitat structure’, Ecology, 55(5),
1017–29.
Wilson, E.O. (1985), ‘The biological diversity crisis’, BioScience, 35(11), 700–706.
Wilson, E.O. (ed.) (1988), Biodiversity, Washington, DC: National Academy Press.
Wilson, W.L. and A.D. Johns (1982), ‘Diversity and abundance of selected animal species
in undisturbed forest, selectively logged forest and plantations in east Kalimantan,
Indonesia’, Biological Conservation, 24(3), 205–18.
World Resources Institute (WRI) (2005), ‘Earth trends: the environmental information
portal’, http://earthtrends.wri.org/, accessed 13 August 2009.
Zohary, M. (1973), Geobotanical Foundations of the Middle East. Vol 1, Stuttgart: Gustav
Fisher Verlag.
5. How do institutions affect the
management of environmental resources?
Bhim Adhikari
Introduction
In recent years, institutions and institutional arrangements have become
central in the study of the success or failure of environmental resource
management. The enforcement of institutions such as contracts1 and prop-
erty rights plays a crucial role in managing natural resources, affecting the
equity and efficiency of resource management regimes. The centrality of
contracts and property rights in understanding the diversity of institutional
arrangements began in the research programme initiated by Coase (1937,
1960) and implemented by new institutional economics (NIE), and is now
widespread throughout the economics literature (Menard, 2000). NIE
focuses on explaining the determinants of institutions and their evolution
over time and evaluates their impact on economic performance, efficiency
and distribution (Nabil and Nugent, 1989). The theoretical framework of
NIE has been used in many disciplines, ranging from sociology, anthro-
pology and legal studies to applied fields such as policy analysis, planning
and organizational development. It is recognized that ‘institutions matter’
and that the associated incentive structure in a particular form of institu-
tion substantially influences economic performance (Bardhan, 1999).
NIE provides a coherent theory of how contracts and collective action
can be seen as the logical outcome of rational individuals’ utility maximi-
zation and how institutional changes alter the pattern of individual choice
and incentive directions. Lin and Nugent (1995) divide NIE into two
broad categories, one studying the demand for institutions and one study-
ing the supply of institutions. So the institutional analysis has adopted two
inter-related approaches: (1) the transaction costs and information costs
approach and (2) the collective action approach (Nabil and Nugent, 1989).
The former is concerned with the role of transaction costs in economic
organizations. The general hypothesis is that institutions are transaction-
cost-minimizing arrangements, which may change and evolve with changes
in the nature and sources of transaction costs and the means for minimiz-
ing them. The transaction costs approach is thus suitable for analysing
the functional role of common property institutions (that is, demand for
institutions). A second theme is the collective action approach, which
119
focuses on property rights. The existence of property rights facilitates

cooperation, which significantly reduces the costs of transaction. In this
way, along with technology and other traditional constraints, institutional
constraints enter into the decision process of individuals (Aredo, 1999).
The collective action theory is the appropriate framework for understand-
ing institutional change (that is, the supply of institutions). The law and
economics branch of NIE studies the behaviour of rational individuals in
a setting where the rule of law imposes prices on various non-market deci-
sions (Posner, 1987). A closely related area is the economics of property
rights and contracts (Alchian, 1965).
Institutions are an integral part of an economic system. They help
to guide human behaviour and act as a key to economic performance.
Institutions are the rules of the game, both in game theory settings and
in the arena where individuals exchange goods and services. Institutions
serve a number of important economic functions like facilitating market
and non-market transactions, coordinating the formation of expecta-
tions, encouraging cooperation and reducing transaction costs. Apart
from being behavioural constraints, institutions also serve as a kind of
knowledge in a world of imperfect information and imperfectly rational
individuals (Olsson, 1999). In the context of common pool resource (CPR)
management,2 institutions can be more specifically defined as a set of
accepted social norms and rules for making decisions about resource use:
these define who controls the resource, how conflicts are resolved and how
the resource is managed and exploited (Richards, 1997). Institutions are
often subdivided into formal and informal institutions. Formal institu-
tions include laws, contracts, political systems, organization and markets;
informal institutions are informal rules of conduct like norms, traditions,
ethics, value systems, religion and ideologies. The former include informal
cooperation and exchange, and moral or spiritual controls, often based on
traditional heads, organized user groups, village committees and district
councils. CPRs usually depend on a mix of both types of institutions. An
institutional arrangement3 is basically an arrangement between two differ-
ent economic units that govern and shape the way in which each economic
agent can negotiate and cooperate (Kherallah and Kirsten, 2001).
The role of property rights in managing local public goods4 is examined
in the NIE literature. Property rights are social institutions that define or
delimit the range of privileges granted to individuals or groups to specific
assets, such as parcels of land, water or forest (Libecap, 1989). According
to Coase (1960), externalities that arise from the use of public goods can
be internalized through bargaining and negotiation if property rights are
well established and transaction costs are zero. That is, voluntary nego-
tiation will lead to a fully efficient outcome providing that (1) rights are
How institutions affect the management of environmental resources 121
well defined, (2) transactions are costless and (3) there is no income effect
(Baland and Platteau, 1996). The Coase theorem was later criticized for
the following reasons. First, in a game involving more than two individu-
als (parties), the solution that leads to an efficient outcome through decen-
tralized bargaining depends on the initial assignment of rights. Second, the
negative effects of many resource and environmental problems occur in
the distant future. The Coase theorem is not clear about how the concern
of future generations is taken into account in the bargaining process.
Despite some crucial difficulties, there is no doubt that the Coase theorem
demonstrates the importance of property rights and transaction costs in
order to internalize the externality associated with public good manage-
ment. The basic motivations for contracting property rights are as follows.
First, individuals do not have to consider the full social costs of their
activities with respect to resource use in the absence of well-defined prop-
erty rights. Second, resources will be undervalued because reallocation of
the resources to higher-value uses becomes more costly and not feasible if
property rights are absent. I shall clarify this argument in the subsequent
discussion.
This chapter proceeds as follows. The second section focuses on the
current debate concerning property rights transformation, placing the
problem in a broader theoretical context. I then analyse the notion of
common property resources and show how open access resources differ
fundamentally from resources held under community ownership. The role
of transaction costs is then examined before considering the impact of
group heterogeneity.
Property rights transformation and resource management

Property rights are social institutions, including formal legal codes and
informal social norms, which define and enforce the range of privileges
granted to an individual or group of individuals with respect to specific
economic resources (see Barzel, 1997). The assignment and enforcement of
property rights is thus a legal mechanism that institutionalizes ownership
of resources to a particular agent (however, property rights may also be
informal institutions). According to the property rights school of think-
ing, the problem of over-exploitation and degradation of CPRs can be
resolved only by creating and enforcing private property rights (Demsetz,
1967; Cheung, 1970; Johnson, 1972; Smith, 1981). Private property is
considered to be the most efficient way to internalize the externalities gen-
erated from the over-exploitation of the commons. On this basis, restruc-
turing property rights remains one of the top priorities in land reform and
natural resource policy in many developing countries. A common practice
is to clarify poorly defined property rights over these resources, especially
exerting state ownership over communally managed resources to maximize

long-term economic rent.5 However, experience from different parts of the
world indicates that the efficiency of changing property rights cannot be
guaranteed simply by enacting new legal rules, since the very notion of
property rights is largely embedded in prevailing social customs (informal
institutions) that guide individual behaviour in respect of environmental
resource use. The nature of this transformation in property rights deter-
mines the parameters for the use of scarce resources and assigns incentive
structures and costs to economic actors. Since property rights institutions
range from formal arrangements, including constitutional provisions,
statutes and judicial rulings, to informal conventions and customs regard-
ing the allocation and use of property (Libecap, 1989), transforming
property rights requires complementary changes in social norms. Together
with new formal rules and other constraints, these changes in social norms
redefine economic opportunity and redraw the rules of the game that
govern the actions of economic actors, including business organizations
and individual entrepreneurs, in their pursuit of economic gain (Wang,
2001). Correspondingly, the new structure of property rights does not
necessarily ensure efficient utilization of environmental resources until
economic actors adjust their expectation and behaviour in response to
underlying changes in property rights.
Contemporary theoretical debate on property rights changes is broadly
dominated by two schools of thoughts, the ‘economic school’ and the ‘dis-
tribution school’ (Coase, 1960; Demsetz, 1967; North 1981; Eggertsson,
1990). The basic theme of economic reasoning in the domain of institu-
tional changes is that the propensity to achieve economic efficiency in
the allocation of resources is the main force propelling such changes.
Individuals will work to establish new rights or reallocate existing rights
only if the benefits from doing so exceed the costs (Libecap, 1989). Libecap
further argues that ‘capturing a portion of the aggregate gains from miti-
gating common pool losses is a primary motivating force for an individual
to bargain, to install or to modify property rights arrangements’ (Libecap,
1989 p. 19). An institution established to achieve such an objective remains
in place as long as it serves the purpose. Whenever the underlying eco-
nomic relations change, the existing institution cannot serve the purpose
efficiently and so provides the motivation to change the existing worn
out institution. The basic theoretical thrust of the economic school is to
view property rights evolution as a response to changes of relative prices,
either direct or indirect, via the opening of new markets, population
change, technological innovation and so forth (Wang, 2001). Changing
economic conditions create new opportunities that could not be captured
by the existing property rights structure, which is thus under pressure to
change (Demsetz, 1967). Demsetz provided a classic example of economic

reasoning driving property rights changes in relation to land following
the opening of the fur trade. Since the existing property rights arrange-
ments could not guarantee the maintenance of minimum stock required to
conserve fur-bearing animals, private property rights over land resource
emerged in response to this economic change to generate long-term com-
mercial gain (Wang, 2001).
The distribution school, on the other hand, emphasizes the role of
economic actors in changing the property rights structure of any par-
ticular resource system. This school argues that distributional issues can
complicate the process of evolution of new systems of property rights
since efficiency of improvement is blocked by distributional inequality.
Built upon a perceived weakness of the economic school, the distribution
school recognizes that distributional issues can complicate the process of
property rights evolution (Wang, 2001). Shifts in the political influence
of potential claimants can lead to the transformation of property rights.
The agents who benefit from such a change have a larger stake in new
systems of property rights and therefore have a greater incentive to act
in favour of such institutional change. However, the same change has a
corresponding disincentive effect on the other agents whose endowments
have been reduced. Economic inequality among competing resource users
may give rise to conflicts and can severely block any institutional response
to CPR problems. Libecap (1989) argues that distributional conflicts
inherent in any property rights arrangements, even those with important
efficiency implications, hinder the adoption of new institutions. In criticiz-
ing the economic school as being naive and single-minded in emphasizing
the ‘demand side of institutional change (that is, gain from the change),
the distribution school correctly points out factors such as distributional
conflicts on the “supply side” can block property rights change’ (Wang,
2001, p. 419).
Understanding institutions and institutional change is therefore a
prerequisite for optimal policy prescription. Since an institution creates
a social equilibrium, institutional change is thus a move from one equi-
librium to another. The success of institutional change in this regard can
be judged by whether this equilibrium is able to establish stable human
expectation with respect to the use of scarce resources. What is critical
in this regard is whether the very notion of different forms of institutions
for CPRs is understood while creating and enforcing new institutions for
sustainable resource management. From the perspective of institutional
approaches to resource management, Libecap (1989) suggested that
common pool losses are the primary motivation for contracting property
rights (institutional change). Libecap further hypothesized that, all things
being equal, the greater the losses of the common pool then the greater
the size of the anticipated aggregate benefits of institutional change
and the more likely new property rights will be sought and adopted.
Nationalization of environmental resources in many developing countries
seems to be influenced by this hypothesis as well as ‘the tragedy of the
commons’ metaphor. In this metaphor, resources managed under common
property rights and open accesses were frequently viewed as synonymous.
It was thought that a resource held under a CPR regime is inherently inef-
ficient since individuals do not get proper incentives to act in a socially effi-
cient way. As a consequence scholars have long questioned the incentive
for efficient use of CPRs under common property regimes (Gordon, 1954;
Scott, 1955; Hardin, 1968) and solutions have been proposed, such as state
control and management (Hardin, 1968) or privatization of the commons
(Demsetz, 1964). In explaining the tragedy of the commons, the economic
theory focuses less upon the weakness of the individuals and more upon
the imperfections in the social systems to which they respond. In fact,
economists are less convinced of the importance of human failing in deter-
mining social outcomes, simply because economists have believed in a
special form of social synergism since the time of Adam Smith. However,
more careful analysis of the foundation of common property regimes in
developing countries has shown that local institutional arrangements,
including customs and social conventions designed to induce cooperative
solutions, can overcome the collective action problems and help achieve
efficiency in the use of such resources (Gibbs and Bromley, 1989; Ostrom,
1990). In the following section, I present a critical analysis of the common
property and open access resource systems and explain how common
property institutions were misunderstood.
Property regimes: open access versus common property

The terms ‘open access resource’ and ‘common property resource’ are
often confused and sometimes mistakenly used interchangeably. While the
first term refers to resources over which no defined property rights exist,
the second specifies the resources managed under community ownership.
Hardin (1968) saw over-exploitation as an inevitable outcome of the use
of common resources, even when the individuals sharing the benefits of
such resources act in an economically rational way. Hardin was neither
alone nor novel in making the argument about open access resources.
An article on the open access problem can be traced back to the early
twentieth century. A little-known article by Jens Warming written in 1911
(cited in Stevenson, 1991) is perhaps the earliest more or less accurate
description of the open access problem. Two modern resource economists,
Gordon (1954) and Scott (1955) provided the general conventional theory
of common resources in their articles on fishery economics. Building on

Gordon and Scott, Smith (1968) provided a general theory of produc-
tion and consumption of common property resources using an algebraic
model. Smith applied his general model to fisheries, mining and timber
resources and showed how differences in production externalities,6 either
from production scarcity or from crowding by procedures, lead to different
outcomes. Most of these articles, however, advocated that a resource held
under a common property regime is not efficient since individual resource
users do not get proper incentives to act in a socially efficient way.
Hardin’s arguments have been formalized later on in the form of the
‘Prisoner’s Dilemma Game’ (Runge, 1981). The prisoner’s dilemma game
is conceptualized in a non-cooperative game theory in which all players
are assumed to have complete information about the game to be played.
Each player has a dominant strategy in the sense that the player is always
better off choosing a dominant strategy – to defect – no matter what the
other player does. The disturbing conclusion of prisoner’s dilemma is
that rational people cannot achieve collective outcomes. However, where
the situation is a recurrent one, for example, as in the case of a repeated
game, the logic changes (Axelrod, 1981). Free-riding in this circumstance
remains a possibility but not an imperative as described in the ‘Simple
Prisoner’s Dilemma Game’ (Runge, 1984; Sugden, 1984). Olson (1965)
also discusses the difficulty of getting individuals to act in such a way that
it increases their joint welfare. Olson challenges proponents of ‘group
theory’ who believe that individuals with common interests would volun-
tarily act to maximize the collective benefits. Olson argues that unless the
number of individuals is quite small, or unless there is coercion or some
other special device to make individuals act towards the overall common
interest, rational, self-interested individuals will not act to achieve their
own common interest (Olson, 1965). The tragedy of the commons, the
prisoner’s dilemma and the logic of collective action are closely related
concepts in the models that have defined the accepted way of viewing
many problems that individuals face when attempting to achieve collective
benefits (Ostrom, 1998).
The tragedy of the commons metaphor confused common property
regimes with open access regimes. It did not understand the very essence
of community wisdom to act together and institute checks and balances,
rules and sanctions, for sustainable management and utilization of envi-
ronmental resources. In other words, followers of this concept after Hardin
did not understand the fact that many resources used by rural communi-
ties are not open access but are managed under community ownership.
Scholars of the commons argued that Hardin confused common property
with open access, failing to distinguish between collective property rights
Table 5.1 Types of property rights regime relevant to common pool

resource
Regime
Open access (res nullius) Free for all; no defined group of users or owners
and benefit stream is available to anyone interested
in entering into harvesting the resource, resource
use rights are neither exclusive nor transferable
State property (res publica) Ownership and management control held by the
nation state, state has right to determine use/access
rules (but in real world, use rights and access rights
are often not enforceable without high cost)
Common property (res Use rights for the resource are controlled by an
communes) identifiable group and are not privately owned or
managed by government; rules exist concerning who
may use the resource, who is excluded from using
the resource, and how the resource should be used;
the co-owner has both rights and duties with respect
to use rate and maintenance of the property owned
Private property Owned by individual; individuals have right to
undertake socially acceptable uses; claim rests with
the individual or the corporation
Sources: Berkes and Taghi Farvar (1989); Bromley (1991).
and no property rights (Ciriacy-Wantrup and Bishop, 1975). One possible

reason behind this misunderstanding is a lack of clarity between prop-
erty and resource regime. While ‘property’ refers to rights and duties or
social relationships between individuals in respect to a resource, ‘regime’
refers to the type of ownership under which these resources are managed.
The economics literature describes four different types of property rights
regimes relevant to CPRs. Table 5.1 presents the four basic types of prop-
erty regimes, which will facilitate the subsequent discussion on open access
and common property resource regimes.
Before turning to the essential difference between common property and
open access, a brief statement about problems associated with common
goods deserves mentioning. According to an economic definition, a
common good is located between a ‘pure private good’ and ‘pure public
good’. The differences between these two goods reside in the concepts of
jointness and exclusion. A pure private good is a good with the property
of exclusivity, which means that the consumption of the good in question
by one individual prevents its consumption by another. The owner of such
a good can dispose of it as desired and can exclude other people from its
use. On the other hand, a public good can be jointly consumed with others
and is therefore non-exclusive. The rate of consumption of such goods
is independent of the number of consumers and how the good is utilized
(Oakerson, 1986). A common good has characteristics of both private
(subtractable) and public goods (difficult to exclude) since it contains a
certain degree of subtractability and excludability. This implies that the
consumption of a common good by one individual will reduce another
individual’s ability to consume the same good. The rate of consumption of
a common good, however, varies according to the number of users and the
type of use. It appears that it is difficult to exclude anyone from utilizing the
benefits generated by common goods, which characterizes the problem of
common property as the provision of these benefits becomes problematic.
However, it is not impossible for co-owners to jointly benefit from
common resources, as long as there are mechanisms to exclude non-
owners from their use. Common property resources are partly joint
and partly exclusive, which means that a type of property regime has to
apply to exclude certain sections of society from entering into resource
appropriation and exploitation. Communal or common property regimes
try to achieve this exclusion by making the resource accessible only to
an identifiable group or members, who devise certain mechanisms to
regulate the pattern of resource use. Many policy prescriptions towards
centralized management of CPRs in the developing world stem from a
fundamental misunderstanding of possible resource regimes. Due to this
confusion, common property carries the false and misplaced burden of
‘inevitable’ resource degradation that properly lies with situations of open
access (Bromley, 1991). Policy-makers without complete knowledge of
tenurial differences and systems of customary rights quite often advocate
the argument that the efficient utilization of CPRs is only possible under
state property regimes. Nationalization of Nepal’s forests is an example of
transformation of private/communal systems of property rights into state
property (de facto open access), which actually upset centuries of social
arrangements adopted by villagers to overcome resource degradation and
make common property regimes viable. In the following section I will take
up further discussions that underscore theoretical aspects of open access
and common property regime systems.
Open access
Open access conditions occur when there is no exclusively defined access
and use right to a particular resource system. Resources that fall into this
category are subject to use by any person who has the capability and desire
to harvest or extract the resource. This situation often results from the
absence, or the breakdown, of management and authority systems whose

very purpose was to introduce and enforce a set of norms of behaviour
among resource users with respect to that particular resource. Bromley
(1991) considers the open access situation as a resource regime in which
there are no property rights (res nullius). There is no defined group of
users or owners and the benefit from the resource is available to anyone.
All individuals have both privileges and no rights; no user has the right to
preclude use by any other party (Bromley, 1991). Resources held under an
open access situation are doomed to over-exploitation since each resource
user places immediate self-interest above social interest. In the absence
of informal/formal management institutions, there is a consensus that
CPRs are typically subject to Hardin’s tragedy of the commons. Since all
members of a resource-using group are assumed to behave in a socially
inefficient way, the carrying capacity of a resource system will eventually
exceed its rate of regeneration. If property and management arrangements
are not determined, and if investment is in the form of capital assets such
as an improved tree species or range revegetation, the institutional vacuum
of open access ensures that use rates will eventually deplete the asset
(Bromley, 1991).
Under open access, a right of inclusion is granted to anyone who wants
to use the resource and such property systems are likely to generate nega-
tive externalities (Baland and Platteau, 1996). Some CPRs are fugitive
(that is, move from one property to another, such as water) and can be
depleted, so are characterized by rivalry in exploitation (Stevenson, 1991).
The rivalry in consumption of a CPR indicates that extraction by one user
of the resources precludes another’s possession. For example, if one user
cuts a tree, another cannot use the same tree. However, for some ubiqui-
tous CPRs, such as the air, the relevance of rivalry might not be applica-
ble until they are consumed (or polluted) at a very high rate. Rivalry in
extraction indicates that a CPR is not a pure public good at all potential
use rates. As a community size grows, and therefore the number of rights
holders increases, the higher use rate will ultimately exceed the resource’s
regenerative capacity. Depletability of a CPR indicates that, along with
rivalry in consumption, resource supply might reduce to zero at some use
rates. This is true both of strictly exhaustible resources, such as oil and
minerals, and of renewable resources, such as fish and trees (Stevenson,
1991). Simple physical or economic exhaustion can reduce the former’s
supply to zero, and a sufficiently high use rate can extinguish the latter’s
capability to reproduce (Dasgupta and Heal, 1979).
The fugitive nature of some CPRs under open access means that they
must be ‘reduced to ownership by capture’ (Ciriacy-Wantrup, 1952: noted
by Stevenson, 1991). There are no formal property rights over the resource
in an in situ condition. This means that a physical unit of the resource in its
in situ or fugitive state cannot be associated with a particular owner unlike
a private property regime where an in situ resource can be said to belong
to a particular real or legal person. So in an open access condition anyone
who possesses the social and physical capital to exploit the resource and
the desire to enter into resource harvest can do so.
Another fundamental problem associated with an open access regime
is that the over-exploitation of CPRs under open access will then result in
symmetric or asymmetric negative externalities. A symmetric externality is
present in an open access regime where each entrant to resource use imparts
a negative externality to all other producers. The new entrant, in turn,
simultaneously has a negative externality imposed on them by the others.
The externality is reciprocal or symmetric. Common examples include fish-
eries, wildlife, open grazing land, ground water, unregulated woodland and
forests, and common oil and gas pools. On the other hand, an asymmetric
externality occurs when the production or consumption decisions of actors
enter the production or utility functions of others while the recipients of
the externality do not cause any reciprocal effects (Stevenson, 1991). A
typical example includes the classic case of a smoking factory dirtying a
nearby laundry’s clothes. Most of the literature on resources held under
open access has concentrated on the symmetric externality, however, the
concept can easily be extended into both types of externality.
There is a vast literature developing on the issue of resource exploitation
as characterized by an open access regime. In his analysis of deforestation
in Nepal, Wallace (1981) reached several important conclusions in respect
to exploitation of forest resources under an open access condition. First,
resource users over-consume the resource in two different ways: they over-
use the resource relative to other goods, and they over-use the resource
this year relative to next year. Both kinds of over-exploitation occur
because the cost of resource use to each individual is less than the cost to
society. For each user, the cost of a particular product from the commons
next year depends mostly on this year’s consumption by other users. The
benefit of the resource next year is assumed to be independent of this
year’s consumption. Unable to influence this year’s consumption by other
harvesters, each user will consume the resources until this year’s marginal
benefit equals this year’s marginal cost. Second, with two substitutable
resources, resource users may consume too much of one and too little of
another, even if total use is efficient. This unbalanced consumption mix
also results from the divergence of private and social costs of resource con-
sumption. Resource users in open access regimes tend to react to average
rather than marginal costs and the unbalanced consumption mix is the
result of different average and marginal costs (Wallace, 1981).
Third, resource users may use inefficient methods to harvest a resource

as in the case of unregulated common property resources. Competing
resource users over-use capital-intensive harvesting methods in their
attempts to out-harvest each other. In general, they do not consider the
negative impact of their activity on the welfare of another. Fourth, resource
users under open access regimes are not likely to invest in resource con-
servation, even if they know that investment improves the productivity of
the commons. No one has any incentive to invest unless there is an assur-
ance that other users will also invest in order to enhance the productivity
of the commons. This situation is similar to under-investment in public
goods such as clean air. Open access resource users invest in replenishing
the forest only until the marginal costs equal a fraction of the marginal
benefit. This under-investment results from a divergence between those
who invest in the improvements and those who reap the benefits. The
divergence results from a mismatch of the scale of some investments and
the amount of potential individual benefit, and from a lack of incentive to
invest in the resource for future benefits because of a competitive rush for
the resource exploitation in the present (Stevenson, 1991).
People who may have strong incentives to invest in protecting trees for
fodder or timber will have much less incentive to do so for public goods
like clean air and soil conservation because they fear others will ‘free-ride’
on their efforts or because they can free-ride themselves. Those who do
not invest because they see little direct benefit are still able to gain from
the investments by others (Varughese, 1999). This inefficiently low invest-
ment by resource users imposes a welfare loss on the group of community
members. Finally, users of CPRs in open access regimes under-invest in
information about the resource since they have no incentive to acquire
knowledge about planting methods, growth rates, or optimal cutting
techniques and so on. A person who has perfect information about a
CPR under an open access regime would not change his or her behaviour
regarding the resource use, because other users would capture most of the
benefits of any potential change. Thus, no one has any incentive to gather
the information necessary to increase the productivity of commons held
under open access regimes (Wallace, 1981), which seriously threaten the
long-term sustainability of natural resources.
Common property
Common property refers to resources for which there are communal
arrangements for the exclusion of non-owners and allocation among co-
owners. Common property exists when a defined group of resource users
holds property rights to natural resources and there is a restriction on the
number of people who can reap a benefit from the commons. As I noted
earlier, this is a resource regime, where exclusion is difficult and joint

use involves subtractability. On this front, they share the first attribute
with pure public goods and the second attribute with pure private goods.
Ciriacy-Wantrup and Bishop (1975) provide the first accurate description
of the concept of common property by specifying two fundamental char-
acteristics associated with common property regimes. First, the distribu-
tion of property rights among co-owners of common property is equal in
terms of their rights to use the resource. Second, potential resource users
who are not legitimate members of a resource-using group of co-equal
owners are excluded. This implies that, within the community, rights to
the resources are unlikely to be either exclusive or transferable; they are
often rights of equal access and use (Feeny et al., 1998). Common property
does not imply communal ownership, which has been described as a right
that can be exercised by all members of the community (Demsetz, 1967),
nor does it imply free access by all to the resource (North and Thomas,
1977). Bromley (1991) argues that a common property regime (res com-
munes) represents private property for the group of co-owners (since all
others are excluded from use and decision-making) and individuals have
rights (and duties) with respect to the resource in question.
Common property is said to be similar to private property in a sense
that there is exclusion of non-owners. The property-owning group varies
in nature, size and internal structure across a broad spectrum, but it
is a social unit with definite membership and boundaries, with certain
common interests, with at least some interaction among members, with
some common cultural norms and often their own endogenous authority
system (Bromley, 1991). The management group (the ‘owners’) has the
right to exclude non-members, and non-members have a duty to abide by
exclusion. Individual members of the management group (the ‘co-owners’)
have both rights and duties with respect to use rates and maintenance
of the property owned (Bromley, 1991). The rights of the group may be
legally recognized or in some cases they may be de facto rights. The fun-
damental difference between open access and common property is that in
open access situations, every potential user has a privilege with respect to
use of the resource since no one else has the legal ability to keep the person
out. Therefore, an open access situation is one of mutual privilege and no
rights (Bromley, 1991). In contrast, a common property regime indicates
a resource system in which there are rules and regulations defining who is
the legitimate user and who is not. Many misunderstandings found in the
literature may be traced to the assumption that common property regime
is the same as open access. Hardin’s prediction of the inevitability of over-
exploitation follows this assumption (Feeny et al., 1998).
For almost two decades after Hardin’s article, CPRs managed under
communal property and open access were frequently viewed as synony-

mous. It was thought that common property was inherently unstable and
pressures from free-riders were inevitable, leading natural resources to be
degraded in the ‘tragedy of the commons’. However, in many cases this
is not true. Evidence suggests that successful exclusion under communal
property regimes is the rule rather than the exception. More careful analy-
sis of the foundation of common property regimes, combined with closer
investigation of the management of collective goods in the developing
world, suggests that common property regimes are not only viable, but
in some circumstances are essential (Gibbs and Bromley, 1989). Even the
common grazing lands in Hardin’s classic ‘tragedy of the commons’ were
well looked after for many centuries, before they declined for reasons
unrelated to any inherent flaw in the commons system (Cox, 1985). The
tragedy tends to be related to the breakdown of existing commons systems
due to disruptions that have originated externally to the community
(Berkes, 1989). Hardin’s tragedy of the commons often results, not from
any inherent failure of common property, but from institutional failure to
control access to resources, and to make and enforce internal decisions for
collective use. Institutional failure could be due to internal reasons, such as
the inability of the users to manage themselves, or it could be due to exter-
nal reasons, for example an incursion of outsiders (Dove, 1993; Berkes
and Folke, 1998). Pressure on the resource because of human population
growth, technological change, or economic change, including new market
opportunities, may contribute to the breakdown of common property
mechanisms for exclusion (Feeny et al., 1998). The social and political
characteristics of the users of the resource and how they relate to the larger
political system affects the ability of local groups to organize and manage
communal property (Ostrom, 1987).
Stevenson (1991) noted seven different characteristics of common prop-
erty resources, which he regards as necessary to manage common property
successfully. The conditions are individually necessary because a resource
managed under common property must meet all seven of them and the con-
ditions are jointly sufficient for common property because all other resource
use regimes (in particular, various forms of open access and private prop-
erty) fail to meet at least one of the conditions (Stevenson, 1991). Based on
the analysis of Ciriacy-Wantrup (1971) and Ciriacy-Wantrup and Bishop
(1975) on the distinction between open access and common property
resources, Stevenson (1991) described the following characteristics as a
form of resource ownership under common property regimes:
1. The resource unit has bounds that are well defined by physical, bio-
logical and social parameters.
2. There is a well-delineated group of users, who are distinct from

persons excluded from resource use.
3. Multiple included users participate in resource extraction.
4. Explicit or implicit well-understood rules exist among users regarding
their rights and their duties to one another about resource extraction.
5. Users share joint, non-exclusive entitlement to the in situ or fugitive
resource prior to its capture or use.
6. Users compete for the resource, and thereby impose negative exter-
nalities on one another.
7. A well-delineated group of rights holders exists, which may or may
not coincide with the group of users.
The first point indicates that a resource held under common property
must be defined either by biological, physical or social conventions, or a
combination of these. At this point I again emphasize the clear distinc-
tion between the resource and common property regime. As was shown
above, common property refers to management institutions that underline
the relation between individuals, which differ from physical objects. In
contrast, the resource is the physical or intangible asset that a group can
own and manage as common property. Since the institution cannot exist
without the resource that it controls, demarcation of resources, nonethe-
less, must be included in the definition of the social institution of common
property (Stevenson, 1991).
The second point specifies that there are two groups in respect to the
resource: included users and excluded persons. The first group consists of
an identifiable and countable number of users. The second set of persons
do not have the right to use the resource (Stevenson, 1991). This is in con-
trast to open access where everyone is a potential user. Third, more than
two users utilize a common property resource. This clearly distinguishes
common property from private property, where a single person is consid-
ered to be the sole owner. Fourth, the existence of rules (formal/informal)
regarding resource appropriation and exploitation in order to guide the
groups of resource users is the main characteristic that helps distinguish
common property from an open access condition. This includes how rights
are transferred, what financial obligation a user has to the group, what
contribution he or she makes and how the rules themselves are changed.
The rules may be formal and explicit or they may be informal and implic-
itly accepted (Stevenson, 1991).
The fifth point provides an essential difference between common and
private property. It also highlights the relationship between common prop-
erty and a public good. Unlike common property, in private property the
in situ resource belongs to a particular owner. However, under a common
property regime, the user may have a secure expectation of getting certain
amounts of product from the commons, but not particular physical units.
The joint, non-exclusive entitlement condition means that participants in
common property arrangements have simultaneous, ex ante claims on any
particular unit of the resource and it can be argued that an essential step
in the use of common property resources (except the resources that have a
pure public good characteristic) is that they be ‘reduced’ to sole ownership
by capture (Stevenson, 1991). As indicated earlier, point five also provides
some basis to distinguish between common property and public goods.
First, some common property resources like national parks, reserves and
so on have public good characteristics that do not exhibit rivalry at low or
moderate levels of use. Reducing the resource to sole ownership through
capture does not apply in the way that it does to resources that exhibit
rivalry in extraction. Second, these resources exhibit joint, non-exclusive
entitlement, because all participants who use the resource have an ex
ante claim to benefits from the resource. For these reasons, reduction to
sole ownership through capture is not a necessary condition for common
property, but joint, non-exclusive entitlement is (Stevenson, 1991).
Point six indicates that, though multiple users compete for the resource
appropriation and exploitation in common property, they undertake
mutual capital investments in resource conservation. To better understand
this idea, reconsider the problem of common goods. As in an open access
condition, extraction by one user of the resource in a common property
regime may generate negative externalities for other users. However, the
difference lies in the extent to which externalities are generated. Point
seven recognizes that the resource users and resource owners do not
always coincide in a common property regime. Common property rights
holders may rent their resource use rights to the actual users subject to
the condition that the rights holders are a group of people who fulfil the
institutional criteria of common property regimes (Stevenson, 1991).
This is not meant to preclude the situation in which a government entity
coordinates or imposes rules regarding resource extraction on users and
rights holders. A common property resource, therefore, is a resource held
by an identifiable community of interdependent users in which these users
exclude outsiders while regulating use by members of the local community
(Feeny et al., 1998).
Transaction costs and natural resource management
Insights from the economics literature

Transaction costs have been a subject of discussion in the literature
on externalities over the past few decades. In his seminal article ‘The
Nature of the Firm’, Coase (1937) discussed why firms exist and sowed
the seed of the concept of transaction costs in the economics literature.
Coase underlines the important role of transaction costs in the organiza-
tion of firms and other contracts. Coase’s theory was later developed by
many scholars (Coase, 1960; Cheung, 1969, 1983; Alchian and Demsetz,
1972; Williamson, 1985). Coase (1960) observed that identifying relevant
parties, collecting information, undertaking negotiation, enforcing agree-
ments and so on could be sufficiently costly to prevent many transactions
from being achieved. When two or more parties agree to exchange or
transfer goods or services, the transaction can be seen as a form of con-
tract (Drennan, 2000). Arrow (1969) defines transaction costs as the costs
of running the economic system. Transaction costs are the costs of arrang-
ing, monitoring, or enforcing agreements; the cost associated with all the
exchanges that take place within an economy (Eggertsson, 1990; North,
1990). North (1997, p. 150) points out, ‘the study of transaction costs, in
addition to giving us insights into static economic analysis, also holds the
key to unlocking the doors to an improved understanding of economic
and societal performance through time’.
The importance of transaction costs can be studied through ‘transaction
cost economics (TCE)’, which is a new type of economics first mentioned
in Williamson (1991). According to him, TCE is focused on reducing
maladaptation costs through ex ante selection of governance structure
for the antecedent conditions. Its working hypothesis is that economic
organization is really an effort to ‘align transactions, which differ in their
attributes, with governance structures, which differ in their costs and com-
petencies, in a discriminating (mainly, transaction cost economizing) way’
(Williamson, 1991). TCE tries to explain how trading partners choose,
from the set of feasible institutional alternatives, the arrangements that
offer protection for their relationship-specific investments at the lowest
total cost (Shelanski and Klein, 1995). TCE maintains that in a complex
world, contracts are typically incomplete because agents are boundedly
rational, or because certain quantities or outcomes are unobservable. Due
to this incompleteness, parties who invest in relationship-specific assets
expose themselves to a risk: if circumstances change, their trading partners
may try to expropriate the rents accruing to the specific assets. A variety
of governance structures could be adopted in order to avoid this risk.
Nevertheless, the appropriate one depends on the particular character-
istics of the relationship. In this respect, TCE can be seen as the study of
alternative institutions of governance.
Transaction costs are incurred in different stages of production and
exchange. Dahlman (1979) separates transaction costs into three broad cat-
egories: (1) search and information costs, (2) bargaining and decision costs
and (3) policing and enforcement costs. He documented that all of these
costs represent resource losses due to lack of information. Transaction
costs are a real and unavoidable aspect of any economic system. It is not
even possible to eliminate transaction costs by prohibiting all transac-
tions because such a decree would have to be deliberated and enforced
and other institutions would emerge to replace banned markets. Libecap
(1989) points out that having lower transaction costs is a necessary rather
than a sufficient condition for adoption of an institutional arrangement.
The inevitability of transaction costs means that any notion of Pareto
optimality7 is incomplete until transaction costs are incorporated (Griffin,
1991). It is therefore appropriate to examine transaction costs when evalu-
ating the potential of new institutions as alternatives to the existing one.
Transactions differ in a variety of ways. Williamson (1985) identifies
asset specificity, uncertainty and frequency as three major attributes of
transactions that have direct cost implications. These three attributes of
the transactions are also relevant to many natural resource systems. Asset
specificity can be defined as investment in assets specifically relevant for
a particular transaction. This is generally regarded as the most impor-
tant of the three attributes because it creates market imperfections and
allows asset owners to earn rents (Drennan, 2000). It can take the form
of a physical asset related to location, an asset dedicated to a particular
consumer, or a tangible asset that can be easily duplicated. It turns out
to be the case that asset specificity creates information asymmetry, which
permits owners to earn economic rents. Having private information
makes it possible for people to behave opportunistically towards others
who contract for their services. However, asset specificity also reduces the
mobility of assets into alternative uses, with consequent market imperfec-
tions (Drennan, 2000). Most of the physical assets in the form of natural
resources (forests, wildlife, fish and so on) are very site specific. Based
upon this site specificity, species are considered endemic or endangered in
their natural habitat. Transactions in natural resource systems also differ
in the required specificity of human resources. It is useful to distinguish
between idiosyncratic site-specific knowledge, especially the indigenous
knowledge of a local community, and scientific knowledge required for
planning and implementation of resource management activities (Birner
and Wittmer, 2000).
The second attribute of transactions with costly implications for con-
tracting is the extent of uncertainty surrounding the contract. Transactions
surrounded by little or no uncertainty require minimal governance because
ex ante and ex post information about the transaction is available to all
concerned parties while uncertainty in defining and observing perform-
ance makes it difficult to place contracts on that performance (Drennan,
2000). However, natural resources are subjected to various types of threats

according to their causes (human-made vs. natural) and effect (irreversible
vs. reversible). The threat of irreversible loss of natural resources is con-
sidered to be a particular problem since it is associated with extinction of
endemic and endangered species. Whether the threats to natural resources
are caused by community members or outsiders also influences the appro-
priate governance mechanisms: using the instruments of social control
by communities or user groups is usually more effective in dealing with
threats caused by insiders than with those caused by outsiders (Meinzen-
Dick and Knox, 1999).
The last attribute of transaction, frequency, also has implications
for the extensiveness of governance investment. Frequent transactions
justify an investment in governance because the cost can be amortized
across many transactions and is therefore more likely to be recovered
while infrequent transactions are more likely to be controlled by ad hoc
arrangements put in place if and when the need arises (Drennan, 2000).
In the case of natural resource management, one can distinguish frequent
day-to-day management decisions and less frequent strategic decisions, for
example, on the establishment of protected areas. Most activities carried
out to implement management decisions are frequent, ranging from daily
to seasonal (Birner and Wittmer, 2000), which implies high transaction
costs of management.
Birner and Wittmer (2000) discussed ‘public relevance’ as an additional
attribute of transaction, which influences the necessity of public sector
involvement in commons management. Public sector engagement or inter-
vention is required to deal with various aspects of externalities that arise
while using common goods. From the outset, we have seen that negative
externalities occur due to the public nature of environmental goods that
affect the interest of society at large, especially concerning environmental
protection. So, public relevance is an equally important attribute of trans-
actions in the issue of intergenerational equity. For example, formulation
of biodiversity policy needs to pay adequate attention to how the inter-
ests of future generations are taken into account while making current
conservation/resource utilization decisions.
Fenoaltea (1984) considered ‘care- or effort-intensity’ as another key
attribute of transaction. Effort-intensive transactions are easier to monitor
and typically relate to production activities (for example, seedling produc-
tion, plantation, felling trees and so on) in natural resource management,
while care-intensive transactions are difficult to measure because they
leave ample room for shirking and even sabotage (Fenoaltea, 1984), which
are characteristic of protection activities. The monitoring problem of care-
intensive activities is aggravated if the outcome of a transaction is difficult
to measure. From the theoretical framework, it revealed that protecting

the relationship of the contracting parties from opportunistic behaviour
(as well as uncertainty, opportunism, asset specificity and frequency)
justifies the existence of an appropriate governance structure. Therefore,
governance structures offer various remedies for protection against the
uncertainty and opportunism of economic relations. As pointed out
by Pelletier-Fleury et al. (1997, p. 5), ‘exchange requiring investment in
specific assets, in a context of uncertainty and strong information asym-
metries, justifies the recourse to vertical integration as opposed to market
coordination, as this allows transaction costs to be contained’. Moreover,
particular types of transaction are thus handled under particular govern-
ance structures depending on the relative costs of production and transac-
tion (Pelletier-Fleury et al., 1997).
Transaction costs and collective action

At this point, it became clear that resource users enter into various kinds
of explicit and implicit agreements (contracts) in order to initiate collective
action or agree to exchange or transfer goods or services. The process of
contracting involves two parties agreeing to ex ante and ex post situations
and these contributions will be in the form of negotiation, monitoring and
enforcement, where substantial amounts of costs are incurred. Ex ante
costs involve the search costs of finding partners, setting up the agreement
and costs incurred negotiating with the potential partners. Ex post costs
are needed to ensure that exchange is carried out, or monitoring and if
necessary enforcing its performance. Hanna (1995) describes four different
resource management stages in which variable transaction costs are inevi-
table. These four different stages are: (1) the description of the resource
context, (2) regulatory design, (3) implementation and (4) enforcement of
agreed-upon rules. Description of the resource context includes a descrip-
tion of resource users, processors, markets and the analysis of associated
socioeconomic characteristics. Designing the regulation requires informa-
tion describing the resource context, which in turn depends on the quality
of contextual information provided. Implementation of a regulation is a
critical test of a regulation’s fit to its contexts. Monitoring and enforce-
ment of a regulation is the final area of transaction costs. Monitoring com-
pliance with regulations will be excessively costly if monitoring systems are
not designed to be consistent with resource dynamics or a user’s operation
(Hanna, 1995).
Despite the inevitability of transaction costs of resource management,
at this stage I would like to point out that empirical discussion regarding
transaction costs associated with different forms of governance structure
is very sparse. Some scholars argued that the costs of privatization of
communal resources, for example costs incurred for fencing, measure-

ment, title insurance, record-keeping and so on are greater than those of
collective management. However, when various hidden costs of resource
management are incorporated into the analysis, a somewhat modified
picture appears. A common property regime would not have the need
for extensive records on boundaries and sales, but instead require meet-
ings and discussions where the co-owners decided their strategies for the
coming period (Bromley, 1991). This may constitute a significant portion
of costs of resource management. In many field settings, efficient manage-
ment of common property resources is often challenged by various sources
of uncertainty that result in high levels of transactions costs. For individual
resource users, the transaction costs of resource management are related to
participation, opportunity cost of time involved in meetings, time required
to acquire information and to communicate and direct monetary expenses
for travel, communication and information. These costs are directly related
to the effectiveness and efficiency of a collective action; and at the com-
munity level these costs may well be borne by poor community members
(Adhikari et al., 2004; Adhikari and Lovett, 2006a; Meshack et al., 2006).
It may therefore be that benefits from collective action are exceeded by
transaction costs (Hanna, 1995). Thus, ignoring transaction costs in policy
design and evaluation leads to the risk of producing suboptimal policy rec-
ommendations. A starting point for analysing this lies in an examination
of what transactions occur, and what interactions are needed as the bare
minimum for effective policy operation (Falconer, 2000).
High transaction costs, whether perceived or actual, related to entry
into collective action may pose significant constraints on participation.
The private transaction costs incurred by individuals may be so high that
they might limit participation of poorer sections of society in some form
of collective action. The existence of transaction costs may also have
important distributional aspects. For example, sizeable fixed transaction
costs related to participation (that is, mandatory start-up costs to be con-
tributed by each participating community member at the initial stage of
collective action) may discourage poor community members entering into
community-based management regimes, as their income from the man-
agement of CPRs is relatively small. Room (1980) argues that economic
studies of participatory forest management have been biased towards
measuring benefits as opposed to costs, especially the likely major transac-
tion costs of management for local forest users. In most of the community-
based resource management systems with an initially degraded resource
base, the costs associated with management are reported to be higher than
the expected benefits. Nonetheless, in many economic models, physical
input and property rights are taken as variables and transaction costs of
resource management are seldom incorporated in the ‘price’ of resource

consumption, though they can be a significant component of resource use.
These costs vary with the attributes of the resource, the nature of use rights
and the socioeconomic circumstances of the local communities. However,
there has been very little attention paid to the socioeconomic significance
of such costs. Transaction costs are largely invisible and there has been
little attempt to quantify them (Falconer, 2000). While there has been a
recent growth in theoretical research in this area, systematic empirical
research is still lagging behind (Aggarwal, 2000).
Despite the importance of transaction costs in resource management,
there are very few empirical estimates of transaction costs. Moreover,
empirical estimates of transaction costs suffer from various uncertainties
that hinder their quantification. Benham and Benham (2000) proposed
four different arguments, which make empirical measurement of transac-
tion costs difficult. First, there is lack of clarity on definitions of transac-
tion costs in existing literature. These definitions offer powerful conceptual
insights, but they have not been translated into widely accepted operational
standards. Second, there is a problem in separating transaction costs from
production costs since quite often they are jointly determined. This leads
to formidable difficulties in estimating transaction costs separately. Third,
if the cost of transacting is very high, many forms of transaction may not
take place in any economic system of interest. Even if some specific form
of transaction does occur, it may not take place in the form of monetary
exchange. Therefore, of all potential transactions, very few may appear in
the market place. To understand why a particular transaction is likely to
be adopted by an individual requires knowledge of the opportunity costs
faced by this individual. To understand the choices made, we may need to
estimate the cost of transactions that did not actually occur (Benham and
Benham, 2000). Fourth, since individuals and groups in any given society
face different opportunity and thus transaction costs, many estimates may
be needed. Other things being equal, an individual’s political connections,
ethnicity, endowments and other characteristics will affect the opportunity
costs of a particular exchange (Benham and Benham, 2000). Though a few
measurements of transaction costs exist in natural resource management,
hardly any of these estimate their magnitude and variation. Transaction
costs in the public sector can be estimated on the basis of information to
be acquired from public agencies, which are typically part of the organiza-
tion’s budget. However, measuring the transaction costs of resource users
is often difficult (Birner and Wittmer, 2000) since most of these costs are
incurred indirectly, for example, time spent on meetings, carrying out
protection work and other daily activities. In this connection, an adequate
theory of natural resource use should incorporate the role of institutional
structures associated with different management regimes and their associ-

ated transaction costs (Kant, 2000).
Transaction costs, governance structures and natural resource management

In the previous sections, I have discussed the underlying theory of trans-
action costs and its relevance to the natural resource sector. I now turn
to appropriate governance structures for natural resource management
from the perspective of transaction cost economics. Menard (2000, p. 240)
points out that ‘variability in contractual arrangements results from the
necessity of setting up and monitoring transactions that have distinctive
characteristics, particularly with regard to the degree of uncertainty of the
environment in which transactions take place and to the degree of specifi-
city of assets that they mobilize’. Indeed, the different conditions demand
differing governance structures (enforcement procedures) in managing the
CPRs if the so-called tragedy of the commons is to be avoided. Drawing
heavily on Birner and Wittmer (2000), to which I return below, I shall offer
some explanation of how to go about choosing efficient governance struc-
tures for CPR management. I have made clear that asset specificity and
uncertainty provide the basis for selecting appropriate governance struc-
tures. In the view of Menard (2000, pp. 240–41), ‘while specificity of assets
plays a predominant role in the search for an efficient governance struc-
ture, uncertainty is the key factor in choosing the enforcement procedures
of contractual arrangements’. Though I will not go further on uncertainty
and asset specificity issues (see Williamson, 1985 for more detail), I will
review the literature that uses the analytical apparatus of transaction cost
economics, which helps to understand the type of appropriate governance
structure for natural resource management.
Birner and Wittmer (2000) divided transaction costs of natural resource
management into two different parts: transaction costs of decision-making
(TCD) and transaction costs to implement those decisions (TCI). The
decision costs (TCD) are incurred during the process of acquiring various
information prerequisite to making appropriate decisions and include
costs of coordinating activities, such as resources spent on meetings, set-
tling conflicts and costs arising due to delayed decisions. Transaction costs
of implementation (TCI) are influenced by the incentives of those carrying
out implementation activities to comply with the management decision
made, the presence of asymmetrical information and the measurability
of the outcome, the possibilities to use social control for monitoring and
the damage caused in the case of non-compliance (Birner and Wittmer,
2000). The economics literature suggests that the incentive for compliance
depends on direct benefits from compliance as compared with defection.
Moreover, the incentive for compliance is also influenced by the value that
resource users put on management decisions and the degree of members’

compliance with these obligations. The extent of group obligation depends
positively upon (1) the cost of producing the joint goods and (2) the degree
of dependency among members. The degree of members’ compliance with
these obligations depends positively upon the monitoring and sanction-
ing capabilities of the group. Since costs of monitoring and sanctioning
can be high, the degree of cooperative success will depend on the mecha-
nisms the group adopts to economize on such costs (Hechter, 1990 noted
by Molinas, 1998). Encouraging user participation in decision-making
processes, which possibly creates the legitimacy required for compliance,
can minimize monitoring costs. User participation in common property
regimes is the cornerstone to the sustainable management of the commons
in most developing countries. Because resource dependency is very high,
and the number of users is comparatively large, spatial extension and poor
infrastructure make monitoring costly. Conservation measures are also
care-intensive and resources are prone to irreversible damage due to a high
degree of dependency (Birner and Wittmer, 2000).
I now turn to the graphical representation of the nature and extent
of transaction costs as described in Birner and Wittmer (2000), who
presented a very subtle analysis of transaction costs of different govern-
ance structures. Transaction costs of public sector governance (TCp) and
co-management (TCcc) are presented in Figure 5.1a and b respectively.
Transaction costs of decisions and implementation change accordingly
(a) Public sector governance (b) Co-management (state/community)
Transaction costs Transaction costs

TCDp + TCIp = TC p
TCDcc + TC cc
I = TC
cc
TCDcc
TCIp
TCDp TCIcc
Care-intensity Care-intensity
Threat to resource Threat to resource
Measurement costs Measurement costs
Source: Birner and Wittmer (2000).
Figure 5.1 Decision and implementation costs of public sector governance

and co-management
in relation to the variable cost of resource management. Variable costs

described here represent the care-intensity of the implementation activi-
ties, the costs of measuring the outcomes and/or the intensity of the
threats to the natural resource in question. It is clear that the decision
costs of state governance (TCDp) are lower than those of co-management
(TCDcc) with lower variable costs (that is, little care-intensity, little threat
to resources and measurement of outcome is possible). This implies that
transaction costs are higher for a co-management strategy since a higher
degree of effort is required for joint decision-making and coordination.
The decision-making procedure becomes more complicated and costly
with increasing group size since time and effort required appear to be
rapidly increasing functions of the size of the group. This is, indeed, in
line with Olson’s hypothesis, which maintains that the smaller the group
size, the greater the likelihood of collective action. When the consent of
every party participating in collective action is required for agreement,
these costs may be very high indeed (Baden, 1998). Nonetheless, with
increasing variable cost, the decision costs under public sector governance
are assumed to increase more rapidly than those under co-management,
because the probability of making wrong decisions is assumed to be higher
due to a lack of idiosyncratic knowledge that further increases the decision
costs (Birner and Wittmer, 2000).
In contrast, transaction costs of implementation seem to be higher for
public sector governance than those of co-management. This is due to
the fact that community-based management systems have the potential
for solving the commons dilemma by internalizing transaction costs of
resource management within the community. The community has a kind
of built-in incentive of social capital8 and idiosyncratic knowledge that
can be used to make resource-specific decisions or overcome factors such
as social heterogeneity in the group (Adhikari and Lovett, 2006b). The
community can overcome the problem caused by asymmetrical informa-
tion through informal institutions and lower the opportunity costs of their
time which considerably reduces the extent of transaction costs. The com-
munity also has at its disposal the requisite social coercive mechanisms
to force compliance with expected harvest (Grima and Berkes, 1989).
Though Figure 5.1 does not specify how the transaction costs are distrib-
uted between state government and community, co-management seems
to be crucial to shift transaction costs from state agencies to local users
(Birner and Wittmer, 2000).
Transaction costs of public sector governance and co-management are
compared with those of hybrid private sector governance (TChp) in Figure
5.2. The state government’s involvement in common property resource
management seems to be essential for various types of conservation
TCp
TChp
TCcc
Transaction costs
c1 c2
Care-intensity
Threat to resource
Measurement costs
Figure 5.2 Comparative efficiency of different governance structures
initiatives like biodiversity conservation. The graph tries to deal with

‘public relevance’ attributes of transaction costs as described in the previ-
ous section. This specifically represents the resulting efficient governance
structure in relation to the variable costs. Hybrid private sector govern-
ance is assumed to have the lowest transaction costs for low values of
variable costs because (1) decision costs are low, (2) there is no need to
overcome collective action and coordination problems and (3) a private
enterprise can typically use stronger incentives than the state (Birner and
Wittmer, 2000). The figure demonstrates that contracted or regulated
private sector governance is comparatively efficient if c is smaller than
c1, pure state governance is comparatively efficient for c1 , c ,c2 and co-
management (public and collective action sector) is comparatively efficient
if c . c2 (Birner and Wittmer, 2000).
At this particular juncture, it is relevant to show how efficiency of gov-
ernance structure is influenced by the level of state capability and social
capital. This relationship is presented in Figure 5.3. As I noted earlier,
social capital is very productive since it makes possible the achievement of
certain outcomes that would not be attainable otherwise. Peer monitoring
can considerably reduce the cost of monitoring and this is one reason why
local informal institutions of resource management are able to perform
better than centralized government-mandated institutions (Aggarwal,
2000). The transaction costs of hybrid private governance and of state
governance increase more rapidly with increasing values of variable cost
if state capability is low. Experience from different parts of the world
TChp TCcc
Transaction costs TCp'
TC hp' TCp
TCcc'
c⬘ Care-intensity
Threat to resource
Measurement costs
Figure 5.3 Impact of state capability and social capital on governance

structure
suggests that sole state ownership over natural resources is less likely to
be efficient in protecting these resources. For example, biodiversity losses
may increase more rapidly due to difficulties faced by state governance in
preventing over-exploitation of biological resources due to high enforce-
ment and monitoring costs. As discussed earlier, increased social capital
reduces the transaction costs of collective action through coordinating
the resource users and implementation of instruments of social control.
Figure 5.3 shows that hybrid private governance is comparatively efficient
for c . c9 and co-management is the optimal choice for c . c9 (Birner and
Wittmer, 2000).
Table 5.2 provides a summary of the institutional choice and govern-
ance structure for natural resource management under different state
capability and social capital.
If both state capability and social capital are low, private sector manage-
ment with state regulation is superior to community-based management.
This significantly reduces the transaction costs of resource management.
Participatory management is especially suited to cases where there is a
high probability of strong community participation. User involvement
in decision-making processes enhances compliance with resource use
regulation. Moreover, community-based management is best suited where
equity issues need to be taken into account. Co-management may be the
optimal choice where governance structure places less demand on social
Table 5.2 Role of social capital and state capability in selecting

appropriate institutions for commons management
State Social Capital

Capability
Low High
Low (Hybrid: private sector Community-based management
management under contract or
regulation)
High Public sector management Hybrid: co-management
involving government agencies
and local communities
capital (state assistance in maintaining this governance structure) and

state capability (transparency may help to reduce corruption in the public
sector) (Birner and Wittmer, 2000). Birner and Wittmer (2000) summarize
the attributes of the most important governance structures for common
property resource management in Table 5.3.
Because transaction costs are a function of the chosen governance struc-
ture and physical characteristics of the resource, ex ante evaluation of pos-
sible resource regime and associated transaction costs is thus an apparent
challenge. Since measuring transaction costs is itself a very difficult task,
estimating the expected gain from selecting a particular governance struc-
ture is also troublesome. Vätn (2001, p. 671) points out, ‘when setting up a
regime one must also evaluate or make qualified guesses about the future
development of external costs, for example, costs due to technological
change, new products, etc, implying changes in matter cycles and habitat
interactions’. Theoretical insights presented in this section help to evaluate
and design appropriate forms of governance structure for natural resource
management in these circumstances.
Empirical studies
Having described transaction cost theory and the criteria in selecting
appropriate governance structures I now provide a brief overview of
empirical attempts that try to measure the transaction costs of collec-
tive action. The concern in this section is with three types of transaction
costs: (1) information costs, (2) bargaining costs and (3) enforcement
costs, as described in the previous section. Very few empirical studies
have attempted to measure the transaction costs of community-based
resource management. Davies and Richards (1999) extensively reviewed
the literature on economic analysis of participatory forest management to
Table 5.3 Important attributes of different property rights’ structure in natural resource management
Governance Structure Centralized/ Community Private Sector Regulated Regulated Co-

Concentrated Management Management Community Private Sector management
Public Agency Management Management (State/
Attributes Communities)
Instruments
Incentive intensity − + ++ + + +
Administrative control ++ − − + + −/+*
Participation, social control − ++ − + − +
147
Demand on frame conditions
State capability ++ − − + + +
Social capital − ++ − + − +
Accommodation of public
interests
Local − ++ − + − ++
Higher level/Sovereign ++ − − ++ ++ +
Note: ++ = strong, + = semi-strong, − = weak, * depending on the type of co-management arrangements.
Source: After Birner and Wittmer (2000); based on Williamson (1999).

understand the incentives for different stakeholders. They found that these
studies tend to be biased towards benefits in general as opposed to costs like
transaction costs and ex ante studies for project preparation as opposed to
ex post monitoring and impact analysis. Cheung (1987) suggests that the
challenge to economists is to specify and identify what these transaction
costs are and how they will vary under differing circumstances.
Despite this, few empirical studies on transaction costs exist. For
example, Crocker (1971) conducted an empirical analysis of the role of
transaction costs in natural resource transfer using the case of the impact
of air pollution on agricultural land use. He concluded that transaction
costs for affected farmland owners to bargain with polluters were very
high. Leffler and Rucker (1990) applied transaction costs analysis to the
structure of timber harvesting contracts and established empirical evidence
for the influence of specific types of transaction costs on contractual pro-
visions. Kumm and Drake (1998) estimated the private transaction costs
incurred in relation to participation in the Swedish agri-environmental
programme, using data from a survey of 90 randomly selected farmers.
Transaction costs were defined to include expenditure for assistance from
agriculture or conservation consultants, mapping, communication costs
related to participation and time inputs (individuals’ working hours). On
average, consultants’ costs accounted for approximately one-third of the
total costs and the individuals’ labour accounted for approximately two-
thirds. Transaction costs, as a share of actual compensation received, are
typically around 12 per cent and private transaction costs have risen over
recent years.
Drake et al. (1999) carried out a pan-European survey to determine
the cause of participation and non-participation of farmers in agri-
environmental programmes in eight EU member states. They outlined a
theoretical econometric participation function related to parameters like
the direct resource costs of conservation (in terms of reduced production
levels), the direct utility of the farmer derived from conservation activities,
and the transaction costs borne by farmers in relation to participation.
They found that transaction costs borne by farmers might pose constraints
on participation. Information-gathering, for example, on the economics
of organic conversion and how to change management practices, is often
a key component of the transaction costs incurred by farmers wishing to
participate in conservation schemes (Drake et al., 1999).
Aggarwal (2000) undertook a case study of group-owned wells in
southern India in an attempt to understand the possibilities and limita-
tions to cooperation in small groups by looking at the transaction costs
associated with these activities. He observed that start-up costs of well
construction in these villages included the costs associated with digging,
pump installation, construction of waterways and electricity connec-

tion. The major operating expenses were the costs of electricity used in
pumping, the cost of pump repair and maintenance and costs associated
with periodic removal of silt from the well. The community also invests
substantial amounts of money periodically for the expansion of the wells.
He observed that costs of negotiating are likely to be higher in the case of
expansion activities, particularly in groups where heterogeneity among
members in terms of their endowments and needs is high. Because of the
higher stakes involved in the case of expansion activities, a higher peer
pressure is required to enforce collective arrangements (Aggarwal, 2000).
Moreover, he also observed that instead of community-owned wells, most
of the villagers prefer to have their own private well. Their reluctance for
joint investment can be understood in terms of the high transaction costs
of negotiating and enforcing a complete contract that outlines the obliga-
tions of each member under the different possible contingencies that can
arise (Aggarwal, 2000).
Richards et al. (1999) undertook a participatory economic analysis of
community forestry in Nepal in an attempt to improve donor and project
understanding of the economic incentives faced by different stakeholders,
and in particular local forest users. The case study particularly seeks to
contribute to efforts to improve equity in the forest user groups and to
understand the role of recurrent annual transaction costs faced by com-
munity members. Transaction costs were simply measured in terms of the
opportunity costs of time spent in obligatory forest activities (planting,
protection, weeding and so on) and in various community meetings. They
found that transaction costs of resource management as a percentage of
total costs were significantly higher for the less forest-dependent commu-
nities than for more dependent forest user groups. The study concluded
that it is very important to include transaction costs in economic studies of
community-based resource management (Richards et al., 1999).
Empirical estimation of transaction costs is also documented in Kuperan
et al. (1998), who attempted to analyse a fisheries co-management system
in San Salvador Island in the Philippines. The transaction costs of fisheries
management were categorized into three major cost items: (1) informa-
tion costs, (2) collective fisheries decision-making costs and (3) collective
operational costs. They found that the difference in the total costs of
fisheries management between centralized government management and
co-management is significant; there is significant difference in the costs at
the different stages of management. For the first two stages, which are the
stages of initiating a new management regime and community education,
the costs are higher for the co-management approach compared with the
centralized approach. Nonetheless, transaction costs are lower in the latter
stage for a co-management approach when monitoring and enforcement

and conflict resolution become more important. This is because the costs
of monitoring and enforcement are likely to be lower as resource users
are more likely to comply with community-devised rules and regulations
as opposed to regulation imposed by a centralized government authority.
Since monitoring costs are the major transaction costs, and monitoring is
undertaken by the community, there is an opportunity for these costs to
decline over time as community acceptance of the rules and regulations for
managing the common property increases with a greater moral obligation
to obey those rules and regulations (Kuperan et al., 1998), that is, the costs
are internalized. They further argue that monitoring activities emerge as
the activity accounts for more than 50 per cent of the total costs of all the
activities involved in co-management. It takes up the bulk of the time as it
is a continuous day-to-day activity and it is a crucial activity for the main-
tenance of institutions (Kuperan et al., 1998).
The impact of group heterogeneity

I now turn to a slightly different issue, heterogeneity, which has direct
policy implications for the emergence of local management institutions.
One of the issues related to successful collective action is the contested
role of group heterogeneity, which is assumed to have something to do
with the way institutions evolve. Particularly important among these
issues is the question of socioeconomic, ethnic and political heterogeneity
and their effect on local-level collective action and resource management
(Keohane and Ostrom, 1995; Baland and Platteau, 1996, 1998; Schlager
and Blomquist, 1998; Uphoff, 1998; Bardhan and Dayton-Johnson, 2000;
Velded, 2000; Varughese and Ostrom, 2001). The assumption is that
socioeconomic differentiation and group heterogeneity make cooperative
arrangements more difficult. On the one hand, there is widespread reali-
zation that productivity-enhancing CPR governance is difficult when
appropriators are heterogeneous in regard to their socioeconomic endow-
ments. A large and influential component of the literature claims that
heterogeneity inhibits innovation of local management institutions since it
creates distrust and suppresses the level of mutual understanding among
community members. In such communities, the process of crafting rules
and regulations with respect to how a resource should be managed can
involve high levels of local dispute. Some economic and social science
literature emphasizes that homogeneity or heterogeneity among agents in
any society reflects the levels of trust, which influences the emergence of
local management institutions through its impact on costs of transactions
(Zak and Knack, 2001). Some other scholars, on the other hand, posit
that heterogeneity is not necessarily bad for collective action. Baland and
Platteau (1996, p. 301), point out that, too often, ‘heterogeneity is blamed
as a matter of principle without enough effort being devoted to spelling
out the precise conditions under which it undermines collective action’.
Indeed, the role of group heterogeneity in the evolution of local manage-
ment institutions is ambiguous. What deserves credit in this respect is to
clarify the various forms of heterogeneity and the way they bear upon
collective action dilemmas. In the subsequent discussion I will critically
examine group heterogeneity and the success or failure of collective action
drawing upon evidence from theoretical and empirical work from both the
economic and social science literatures.
The source of heterogeneities are diverse, and include differences in
asset holding, appropriation skills and access to technology, caste, gender,
ethnicity, opportunity cost, political influence and other local differences,
which might influence the equity of resource distribution and thus per-
formance of collective action. Kant (2000, p. 288), points out, ‘resource
users will often have somewhat different preferences regarding resource
management, or assign different priorities to the various objectives of
resource management, either because of differing personal interest in the
resource or differing degree of involvement in the social group’. The het-
erogeneity of individual interest with respect to how a resource is managed
affects individual incentives and thus economic use of the local commons.
As rightly pointed out by Seabright (1993), the degree of trust economic
agents, participating in some form of collective action, have in one another
serves a crucial role in common property regimes. He developed a model
of ‘habit forming’ cooperation. The model revealed the fact that players’
belief about each other’s trustworthiness is confirmed and contributes
to cooperative behaviour. Such a ‘habit-forming’ process is, however,
unlikely to work in a community which starts with a high level of het-
erogeneity with respect to resource management preference (Kant, 1998).
Moreover, economic inequality or socioeconomic heterogeneity among
the members of a resource-using group might be associated with different
degrees of access to, and control over, the local commons. Intra-group
heterogeneity in terms of private payoffs, therefore, can often lead to con-
flicts of interest and thus hinder the emergence of egalitarian institutions
and the performance of any cooperative arrangements. Ostrom (1990)
argues that none of the successful CPR situations involves participants
who vary greatly in regard to ownership of assets, skills, ethnicity, race, or
other variables that could strongly divide a group of individuals.
Socioeconomic heterogeneity, relative income effect and the political
framework in which policy decisions are taken determine the benefit-
sharing arrangements in commons management. Income inequality pro-
duces the gap between poor individuals’ ability to pay and willingness to
pay for collective action. The level of the poorer users is so low that they
might not be able to participate in community-based management systems,
because they cannot meet the costs involved. For example, groups that
depend heavily on daily wage labour sometimes find it difficult to contrib-
ute their share of resource management costs. As a result they are deemed
ineligible for the benefits. McKean (2000) observes that common property
regimes do not always serve to equalize income within the user group.
Communities may vary enormously in how equally or unequally they
distribute the products of the commons to eligible users. Decision-making
rights tend to be egalitarian in the formal sense; however, richer house-
holds may actually have additional social influence on decisions due to the
relative strength of their socioeconomic position. Greater homogeneity,
on the other hand, promotes both equity and efficiency by facilitating the
adoption of more coordination and cooperative arrangements at the local
level. In a similar vein, Guggenheim and Spears (1991) argue that in light
of likely scenarios of sociopolitical heterogeneity within spatially defined
community groupings, participation in community decision-making can
be skewed in favour of more powerful subgroups. Asymmetries among
participants facing CPR provision and appropriation problems, therefore,
can present substantial barriers to overcoming the disincentives associated
with collective action (Ostrom and Gardner, 1993).
The sources of heterogeneities are diverse and there are several possible
definitions of heterogeneity. A number of researchers seem to be much
focused on economic inequality as a major source of heterogeneity, that is,
inequality in wealth or income among the community members involved
in collective action. However, Bardhan and Dayton-Johnson (2000) also
noted other kinds of inequality within a resource-dependent community
such as ethnic and social heterogeneity, and environmental or state vari-
ables like low levels of trust or social cohesion. They noted a U-shaped
relationship between inequality and commons management: very high
and very low levels of inequality are associated with better commons
performance, while mid-range levels of inequality are associated with
poor outcomes. Social heterogeneity increases the cost of negotiation
and bargaining inherent in the process of crafting institutions. Economic
inequality, combined with other constraints, severely limits the possible
bargaining outcomes available to resource users. In this regard, inter-
related ‘commons outcomes’ that might be affected by inequality include
resource conservation, maintenance of infrastructure, the supply of local
institutions, monitoring and enforcement of regulations and conflict
resolution (Bardhan and Dayton-Johnson, 2000).
Economic opportunities available to resource users outside the local
commons are considered to be a complicated form of heterogeneity that
Bardhan (1999) considers ‘exit options’. He argues that the effect of exit
options on conservation is complicated. Nonetheless, this depends in part
on the nature of the relationship between wealth and the exit options. If a
resource harvester’s exit option is a concave function of wealth – meaning
that the value of the outside option rises very quickly with wealth at low
levels of wealth, but increases more slowly at higher wealth levels – then
increases in inequality, starting from relatively equal wealth distributions,
will reduce conservation and thus performance of cooperative arrange-
ments. In that case, the relatively poorer harvesters will not be interested
in optimal conservation; as her/his wealth declines, their gain from con-
servation falls off more rapidly than the gain from exercising her/his exit
option. On the other hand, if the exit option is a convex function of the
wealth level then increased inequality might either enhance or damage
the prospects for conservation: the effect is intermediate. If resource users
have relatively lucrative earnings outside the commons, this can affect
their individual incentives. Wealthier households might have reduced
incentives to participate in collective action, as their wealth endowments
afford them attractive outside earning opportunities. For those enjoying
such opportunities, the discounted value of their future income flows from
the CPR may fall below that of alternative incomes available. On the
other hand, users lacking such outside opportunities attach a higher value
to the future state of the resources, which may enhance the likelihood of
collective action.
In his simple numerical model, Kanbur (1992) investigated the role of
group heterogeneity in the success or failure of common property resource
management. He argues that cooperative arrangements are less likely
when agents are highly heterogeneous. Moreover, existing arrangements
are also likely to break down as a group becomes more heterogeneous.
Drawing upon the conclusions from various case studies, he vividly
presents the fact that greater equity (at least greater homogeneity) pro-
motes greater efficiency by facilitating adoption of cooperative arrange-
ments, in situations where externalities make non-cooperative outcomes
inefficient. Tang (1992) presents a synthesis based on careful analysis of
47 case studies of community irrigation systems and found that a low
variance of average family income among irrigators tends to be associated
with a high degree of rule conformance and good maintenance: 72 per
cent more of the cases with low-income variance are characterized by both
a high degree of rule conformance and good maintenance than the cases
with high income variance. This finding may throw new light on the fact
that successful schemes are to be found in relatively homogeneous commu-
nities, and it is precisely in these communities that one is unlikely to find
arrangements that favour one subgroup over another (Kanbur, 1992).
In their theoretical modelling, Baland and Platteau (1999) investigated

the impact of inequality on the ability of resource users to undertake
successful collective action with special reference to over-exploitation of
common property resources. They argued that in the voluntary provision
problem, inequality has an ambiguous impact on efficient outcomes, while
in regulated settings it tends to reduce the acceptability of available regula-
tory schemes, which makes collective action more difficult. Changes that
disequalize distribution of access rights through the definition or redefini-
tion of property rights will have two different effects. First, the agents who
benefit from such a change have a larger stake in the common property
resources and therefore have a greater incentive to take part in conserva-
tion efforts. The same change has a corresponding disincentive effect on
the other agents whose endowments have been reduced. Since the increas-
ing inequality redistributes incentives in different directions, it may have
ambiguous effects on the ability of users to initiate collective action. Under
such circumstances, a homogeneous group may better succeed in design-
ing and enforcing equitable conservation measures than a heterogeneous
one.
In contrast to what I have discussed earlier, heterogeneity and income
inequality in community-based property rights structures are also said to
be conducive to the successful outcome of collective action. Olson (1965)
hypothesized the possibility that groups where considerable heterogenei-
ties exist may be privileged if those with the most economic interests and
power were to initiate collective action to protect their own interests.
Wealthy users may be more concerned with resource conservation since
they can greatly internalize the benefits generated from the commons and
therefore have more incentive to contribute to the local commons. Olson
further argued that the greater the share in the benefits of a collective
action for any single member, the greater the propensity of this ‘large’
member to bear the costs involved in commons management. In a related
vein, Baland and Platteau (1999) revealed that increasing inequality could
stimulate the incentive of the big users to voluntarily contribute and simul-
taneously encourage the small users to free-ride on the former’s contribu-
tion. As a result, the net impact of inequality on collective action will hinge
upon the respective strengths of these two opposite effects. However, the
fact that rich users are more inclined to contribute does not imply that
increased inequality favours collective action. Again, inequality has an
ambiguous impact on collective action.
The Olson hypothesis is likely to make sense when management of a
CPR involves important ‘non-convexities’ in its production function.
Non-convexities indicate the large start-up costs, which are likely to be
incurred in setting up a commons management regime. These costs might
be labour costs of constructing irrigation infrastructure, fencing around

the pastures and maintenance costs like fire prevention measures in com-
munity forestry and cleaning of canals in community-managed irrigation
systems. Furthermore, start-up costs also involve organizational effort
to collectively mobilize community members towards collective action.
Bardhan and Dayton-Johnson (2000) considered that benefits from local-
level collective action are a non-convex function of the effort provided
to produce those benefits if there is a threshold level of aggregate effort
that must be supplied before any benefits are realized. As effort increases
beyond the threshold, however, benefits to the group begin to increase. It
is impossible for poorer members to contribute and bear a significant part
of these start-up costs in order to initiate community-based resource man-
agement. In this respect, wealthier and better-endowed users may be able
to mobilize the capital necessary to support and materialize the collective
action. Baland and Platteau (1997) also confirm the theoretical possibility
of this Olson effect in the presence of non-convexities.
Theoretical literature on heterogeneity is also supplemented by empiri-
cal studies undertaken in different parts of the world, especially in
developing countries. Drawing upon recent theoretical advances in the
analysis of cooperation, Molinas (1998) undertook an econometric analy-
sis of the determinants of successful collective action based on a survey
of 104 peasant cooperative institutions in Paraguay. This study shows
that controlling as much as possible for the specific characteristics of the
community and peasant committee, the relationship between community
inequality and cooperative performance is an inverted U. It was evident
that community members with bigger land holdings are expected to
benefit proportionally more from the committee’s activities than members
with smaller holdings. Since the provision of community infrastructure
will increase the average price of land in the community, users who own
more land naturally benefit more. In addition, the benefits accruing from
joint commercialization of the outputs and collective buying of inputs are
proportional to the scale of production, which in turn depends upon the
size of the land holding. The study concludes that local-level cooperation
is not monotonically related to either the degree of inequality of endow-
ments within the community or the local intervention; rather, it is of an
inverted U-shape form. Dayton-Johnson (2000) develops a simple model
of individual households’ incentives to provide collective maintenance
effort in a communally owned irrigation system in Mexico. He found that
economic inequality and social heterogeneity are consistently and sig-
nificantly associated with lower levels of infrastructure maintenance and
reduce the performance of collective action, while inequality in landhold-
ings has a negative, though complicated, effect on maintenance. Similar to
Molinas’ findings, his model shows that the effect of economic inequality
is complicated and not monotonic (Dayton-Johnson, 2000).
Varughese and Ostrom (2001) investigate the role of heterogeneity in
affecting the likelihood of collective action in a study of 18 forest user
groups in Nepal. Their study focuses on some important community
attributes that affect users’ incentives to cooperate in resource manage-
ment like the size of the community of resource users, locational differ-
ences among users with respect to forest areas, differences in forest users’
income and presence or absence of economic/social/ethnic disparities
and the availability of alternative forest resources. They, however, con-
clude that social and cultural heterogeneity does not have a determinant
impact on the likelihood or success of collective action. Successful groups
overcome stressful heterogeneities by crafting innovative institutional
arrangements well matched to their local circumstances within which user
groups decide how to organize themselves and which rules to adapt to
allocate rights and duties as well as costs and benefits. They further argue
that communities would collectively manage local-level natural resources
where there are very substantive benefits to be obtained through collective
action. Heterogeneity is not a variable with a uniform effect on the likeli-
hood of organizing self-governing enterprises, as communities put effort
into devising rules to cope with heterogeneities and they may be able to
invest more heavily in finding effective rules that are considered fair, effec-
tive and efficient to most users (Varughese and Ostrom, 2001).
A study of CPR management in Fulanui village in Mali (Velded, 2000)
demonstrated little direct relationship between the degree of heterogene-
ity and the success of collective action. It was found that homogeneity
among elite groups enhanced capacity for collective action. However,
when heterogeneity in economic interests between elite groups intensi-
fied and coincided with other dimensions of heterogeneity, such as het-
erogeneity in economic wealth, access to land and CPRs and agreement
over authority of the leadership, the collective action became difficult to
achieve. He further observed that heterogeneity of wealth and ethnicity
do not prevent ‘common interest’ among elite and subordinate groups in
collective action. Rather, Velded argues that the political elite of the com-
munity can assume leadership in local organization of CPR management
and provide an authority structure for monitoring and rule enforcement.
On the other hand, conflict was observed in a similar setting when eco-
nomic interests differed with regard to use of CPRs. Moreover, in terms of
resource use, access to local CPR grazing is more exclusively for the noble
elite and influential section of the community. This study suggests that as
long as there is reasonable homogeneity among elite groups, cooperative
outcomes can be relatively persistent even among heterogeneous social
groups (Velded, 2000). However, in line with Baland and Platteau (1995,
1996), Velded emphasized that there is a need to distinguish between
various forms of heterogeneity in analysis and explanation of collective
action, especially in relation to heterogeneity in endowments, entitlements
(wealth) and economic interests.
Kant (2000) developed an optimal control model for a dynamic
approach to forest regimes in developing economies integrating the natural
system and socioeconomic factors of the resource users. The model dem-
onstrates that the dynamics of the optimal forest regimes depend on the
change in natural factors, socioeconomic factors (user group heterogene-
ity) and on the interaction between natural and socioeconomic factors.
The model defines cultural, economic, ethical and other social differences
as a basic level of heterogeneity. Due to this basic heterogeneity, members
of the user groups may have diverse preferences for timber and non-timber
products and hence prefer a different mix of products, which is considered
as second-level heterogeneity. Preference for diversified forest products
often leads to different preferences for resource management regimes,
which can be further described as a third level of heterogeneity. Therefore,
heterogeneity with respect to resource regimes is treated as a function of
the product preference differences, which in turn is treated as a function of
cultural, economic and social heterogeneity. Though the model is crucial
to understanding optimal resource regimes, their linkages with socioeco-
nomic factors (heterogeneity), and the dynamics of these optimal regimes
and socioeconomic factors, it is explicit about how and to what extent
intra-community heterogeneity or wealth endowment affects equitable
benefit-sharing mechanisms among resource users.
While there has been a great deal of work on the management of local
environmental resources in recent years, there has been surprising little
work, either theoretical or empirical, on how inequality helps or hinders
cooperative management of the local commons (Bardhan, 1999). Not much
is understood on how socioeconomic status of resource users promotes or
discourages participation in CPR management and consequently the equity
of resource distribution. Nonetheless, evidence from South Asia suggests
that the socioeconomic status of resource users may place stringent limits
on the extent to which certain groups are able to participate and benefit
from the management of CPRs. The landless, agropastoralists, and other
politically and economically marginalized user groups may not be able to
take advantage of incentives for tree growing (Guggenheim and Spears,
1991). Moreover, participatory forest management in South Asia illustrates
the sharp equity problem elsewhere since social structure itself is for the
most part inherently unequal. These resource management initiatives were
supposed to help the poorest of the village population, lighten women’s
workload and relieve family budgets. The income of many such projects,
however, has been invested in other community development activities
like village schools, rural roads and small-scale irrigation and very often
productivity gained from such investments is captured by relatively well-
off farmers. Entitlement to products of the commons varies to a surprising
extent (McKean, 1992a). Hill and Shields (1998) observed that the com-
munity incentives in joint forest management (JFM) in India are not so
clear-cut, however, the main losses in JFM are fuel wood head loaders who
are often from the poorest subgroup within the village. Many studies have
shown that the poor and the disadvantaged do not necessarily benefit from
community-based forest management (Bhatia, 1999). While the aggre-
gate gains from reducing common pool problems or promoting economic
growth through the definition or redefinition of property rights are unlikely
to be controversial, the distribution of wealth and political power inherent
in the proposed rights structure will be a source of dispute (Libecap, 1989).
The recent literature on common property resource management,
however, indicates that sustaining environmental resources is not depend-
ent on a particular structure of property rights regime, but rather on a
well-specified property rights regime and a congruence of that regime with
its ecological and social context (Hanna and Munasinghe, 1995). Success
of the property rights regime depends upon the congruence of ecosystem
and governance boundaries, the specification and representation of inter-
ests, the matching of governance structure to ecosystem characteristics,
the containment of transaction costs and the establishment of monitoring,
enforcement and adoption processes at the appropriate scale (Eggertsson,
1990; Ostrom, 1990; Bromley, 1991 and Hanna et al., 1995). More impor-
tantly, the equity and distributional aspects of the CPR regime are con-
sidered to be one of the major determinants of long-term sustainability
of CPR institutions. Since the nature of property rights regimes and the
distribution of access to natural resources affect both the level of poverty
and the quantity and quality of the environmental resource base in the
long run (Dasgupta and Maler, 1991), property rights institutions will
have to be more egalitarian in order to avoid unilateral appropriations of
the commons. Empirical regularities that link inequality to better or worse
outcomes of community-based resource management would be a basis for
asset redistribution programmes including land reform and poverty alle-
viation that target communities based on the level of inequality (Bardhan
and Dayton-Johnson, 2000).
Conclusions
This chapter explores the role of institutions in managing local-level envi-
ronmental resources. The basic purpose was to explain the emergence of
local management institutions and their impact on successful management

of common property resources at the local level. The analytical approach
of this chapter builds on insights from ‘new institutionalism’ and theoreti-
cal and empirical literature from new institutional economics that under-
score the role of formal and informal institutions for a solution to the CPR
problem. Institutions are humanly devised constraints that shape human
interaction and that ultimately affect the performance of the economy
by their effects on the costs of exchange (North, 1990). Institutions serve
a number of important economic functions like facilitating market and
non-market transactions, coordinating the formation of expectation,
encouraging cooperation and reducing transaction costs. From the eco-
nomic perspective, it became clear that the problem of environmental deg-
radation is to get the institution right. The underlying causes of resource
degradation are found in those problems that systematically result in
institutional failures. Since the basic theme of economic reasoning in the
domain of institutional changes is to achieve equity and economic effi-
ciency in the allocation of resources, an institution established to achieve
such objectives remains in place as long as it serves the purpose. Whenever
the underlying economic relations change, new institutions will emerge
in response to underlying economic circumstances of the community.
However, North (1990) pointed out that not all institutions are efficient
and even inefficient institutions can persist for a long time. Institutions
do not always decrease transaction costs but might actually, when they
are inefficient, increase transaction costs (Olsson, 1999). Because transac-
tion costs are a function of the chosen governance structure and physical
characteristics of the resource, designing the governance structure depends
on the attributes of transaction, that is, asset specificity, uncertainty and
frequency of transaction associated with making various management
decisions. Because of the public nature of environmental goods, public
relevance also needs to be taken into account while selecting particular
types of management regime.
The chapter has analysed different resource management regimes and
addressed appropriate governance structures for environmental resource
management from institutional perspectives. I revisited and analysed open
access and common property resource systems and associated transaction
costs. It is shown that open access exploitation results from the absence of
well-defined property rights. Access to the resource system is unregulated
and is free and open to everyone interested in resource appropriation and
exploitation. Rent is completely dissipated at open access equilibrium.
In another words, zero average net revenue characterizes the equilibrium
of an open access regime. There is over-use resulting from resource users
ignoring the effects of their consumption on the costs faced by other users.
Similarly, there is over-use resulting from users ignoring the effect of

their consumption this year on the costs they will face next year. On the
supply side, CPRs held under open access regimes are like public goods.
Individuals cannot capture the benefits of their investments in these
resources, and as a result investment is inefficiently low, resources are mis-
allocated, and there is under-investment in information (Wallace, 1981).
Governance of natural resources can thus be conceptualized as a collective
endeavour of individuals organizing for the provision of, and appropria-
tion from, resources that have public good characteristics. Since individual
interests are unlikely to lead to sustainable management of CPRs in an
open access regime, the design of governance for resource management
has to include some elements of support from government to modify the
incentives for individual resource users (Varughese, 1999).
Analysis of CPR management under common property regimes indi-
cates that resource management under community ownership is not
operating in a vacuum. Instead, common property is a form of resource
management regime in which a well-delineated group of competing users
participates in extraction or use of a jointly held, fugitive resource accord-
ing to explicitly or implicitly understood rules about who may take how
much of the resource (Stevenson, 1991). The confusion in the conventional
literature over the tragedy of the commons arises from a failure to under-
stand the concept of property, and therefore to fail to understand common
property regimes (Bromley, 1991). The economics literature also discusses
the problem associated with common property, which results from some
type of adverse interaction among resource users and unrestricted access to
the resource system by all who care to use it. As discussed earlier, common
property resources share two important characteristics. First, exclusion of
resource users from these resources is difficult. Second, the use of resources
by one person subtracts from the welfare of other users. Natural products
like trees, water and wildlife are subtractable, and in most cases, exclu-
sion will be problematic and costly. If one individual uses more, then less
remains for another. These resources are therefore potentially subject to
depletion or degradation, that is, use that is pushed beyond the limits of
sustainable yields. So the problem raised by common property is usually
represented in the formal framework of the ‘prisoner’s dilemma’.
This chapter also attends to the transaction costs associated with
community-based resource management. An institution’s primary purpose
is to reduce transaction costs and thereby enhance economic performance.
Neo-classical economic analysis tends to be preoccupied with production
costs, largely ignoring the transaction costs associated with production
and economic exchange. Despite the importance of transaction costs in
functioning of an economic system, there are very few empirical estimates
of transaction costs and very few comparative estimates. The analysis

emphasized that transaction costs of community-based resource manage-
ment can be a significant part of resource management costs, which is gen-
erally ignored in economic analysis of participatory forest management.
The chapter also discussed the role of heterogeneity in the perform-
ance of collective action in general and distributional implications of such
regimes in particular. On the one hand, it turned out to be the case that
wealth might affect private benefits from the commons indirectly, through
social relations between one group and another, when users attempt to
deal with collective action dilemmas associated with joint use of common
resources. However, there is still disagreement among social scientists and
economists regarding the effect of heterogeneity on the capabilities of social
groups to undertake successful collective action. In fact, a general finding
from studies on the management of common property systems is that enti-
tlements to products of the commons are almost always based on private
holdings (Dasgupta, 1999). McKean (1992b) reaches a similar conclusion
that the distribution of benefits in collective action reflects inequalities in
private wealth. Drawing upon several theoretical and empirical studies in
Asia and Africa, Baland and Platteau (1996) provide several examples of
homogeneity/heterogeneity and claim that equitable access to resources is
possible even in heterogeneous village society in Japan (McKean, 1992a,
1992b) and an ethnically diverse village in India (Bardhan, 1993a, 1993b),
as well as most pastoral societies of Africa. In most of these villages,
scholars observed a kind of equitable access to locally managed CPRs
with local rules, norms and customary law. It seems that the impact of het-
erogeneity is ambiguous. Analysis of heterogeneity and distributive issues
of collective action have important policy implications for community-
based resource management initiatives, especially those aimed at shaping
resource use decisions by households and poverty reduction through
better management of local commons.
Notes
1. I borrow the idea of Libecap (1989) to conceptualize the process of defining or changing
property rights in terms of contracting. Contracting includes both private bargaining to
assign or adjust informal ownership arrangements and lobbying efforts among private
claimants, politicians and bureaucrats to define, administer and modify more formal
property institutions (Libecap, 1989).
2. Common pool resources refer to the natural resource that is available in the commons,
whereas common property regime refers to the institutional property rights arrangement
by which the resource is managed. The two are often confused as the same acronym,
CPR, is used for both.
3. Williamson (2000) notes that NIE operates at two different levels: macro and micro. The
macro level deals with the institutional environment, or rules of the game, which affect
the behaviour and performance of economic actors in which organizational forms and
transactions are embedded. Williamson discusses it as the set of fundamental political,
social and legal ground rules that establish the basis for production, exchange and dis-
tribution. The micro-level analysis, on the other hand, also known as the institutional
arrangements, deals with the institution of governance. These refer to the modes of man-
aging transactions including market, quasi-market and hierarchical modes of contract-
ing (Williamson, 2000).
4. Public goods are resources that can be accessed by anyone and the consumption of the
resource by one person does not affect the consumption by anyone else. Enjoyment of
a sunrise is a classic example. However, many natural resource public goods can be
readily over-exploited, which is why property rights need to be defined and consumption
regulated.
5. Economic rent is the difference between the actual cost of bringing something into pro-
duction and the amount that is paid for the product. Economic rents are low if there is a
lot of competition and tend to be higher when there are monopolies.
6. Production externalities are costs of production that are incurred by someone other than
the producer. For example, a logging company exploiting old growth forest will pay
stumpage costs on each tree extracted to the owner of the land, but does not compensate
society for the loss of a valuable habitat.
7. Pareto optimality is when goods are exchanged until no one can be made better off
without someone being made worse off. Under these conditions the market is efficient in
that it is responsive to supply and demand.
8. Social capital comprises features of social organization such as networks, norms and
trust that can improve the efficiency of a community by facilitating cooperation and
coordination of relations between actors and among actors (see Coleman, 1990; Putnam,
1993). Social capital is also productive since it makes possible the achievement of certain
outcomes that would not be attainable otherwise (Coleman, 1990; Molians, 1998) and it
also reduces the transaction costs of collective action.
References
Adhikari, B. and J.C. Lovett (2006a), ‘Transaction Costs and Community-based Natural
Resource Management in Nepal’, Journal of Environmental Management, 78(1), 5–15.
Adhikari, B. and J.C. Lovett (2006b), ‘Institutions and Collective Action: Does Heterogeneity
Matter in Community-based Resource Management?’, Journal of Development Studies,
42(3), 426–45.
Adhikari, B., S. Di Falco and J.C. Lovett (2004), ‘Household Characteristics and Forest
Dependency: Evidence from Common Property Forest Management in Nepal’, Ecological
Economics, 48(2), 245–57.
Aggarwal, R.M. (2000), ‘Possibilities and Limitations to Cooperation in Small Groups:
The Case of Group-owned Wells in Southern India’, World Development, 28(8),
1481–97.
Alchian, A.A. (1965), ‘Some Economics of Property’, II Politico, 30(4), 816–29.
Alchian, A.A. and H. Demsetz (1972), ‘Production, Information Costs, and Economic
Organization’, American Economic Review, 62(5), 777–95.
Aredo, D. (1999), ‘Institution and Development: A Note on Conceptual Framework’,
Economic Focus, 2(2).
Arrow, K.J. (1969), ‘The Organization of Economic Activity: Issues Pertinent to the Choice
of Market versus Non-market Allocation’, in The Analysis and Evaluation of Public
Expenditure: The PBB-System, Joint Economic Committee, 91st Congress, 1st Session,
Vol 1, Washington, DC.
Axelrod, R. (1981), ‘The Emergence of Co-operation among Egoists’, American Political
Science Review, 75(2), 306–18.
Baden, J.A. (1998), ‘A New Primer for the Management of Common-pool Resources and
Public Goods’, in J.A. Baden and D. Noonan (eds), Managing the Commons, Bloomington,
IN: Indiana University Press.
Baland, J. and J. Platteau (1995), ‘Does Heterogeneity Hinder Collective Action?’, Working
Paper, Centre for Research in Economic Development (CRED), University of Namur,

Belgium.
Baland, J. and J. Platteau (1996), Halting Degradation of Natural Resources: Is There a Role
for Rural Communities?, Oxford, UK: FAO and Clarendon Press.
Baland, J. and J.P. Platteau (1997), ‘Wealth Inequality and Efficiency in the Commons, Part
I: The Unregulated Case’, Oxford Economics Papers, 49(4), 451–82.
Baland, J. and J.P. Platteau (1998), ‘Division of the Commons: A Partial Assessment of
the New Institutional Economics of Land Rights’, American Journal of Agricultural
Economics, 80(3), 644–50.
Baland, J. and J.P. Platteau (1999), ‘The Ambiguous Impact of Inequality on Local Resource
Management’, World Development, 27(5), 773–88.
Bardhan, P. (1993a), ‘Symposium on Management of Local Commons’, Journal of Economic
Perspectives, 7(4), 87–92.
Bardhan, P. (1993b), ‘Analytics of the Institutions of Informal Cooperation in Rural
Development’, World Development, 21(4), 633–9.
Bardhan, P. (1999), ‘Distributive Conflicts, Collective Action, and Institutional Economics’,
Department of Economics, University of California at Berkeley, mimeo.
Bardhan, P. and J. Dayton-Johnson (2000), ‘Heterogeneity and Commons Management’,
Department of Economics, University of California, Berkeley, mimeo.
Barzel, Y. (1997), Economic Analysis of Property Rights, New York: Cambridge University
Press.
Benham, A. and L. Benham (2000), ‘Measuring the Transaction Costs’, in C. Menard (ed.),
Institutions, Contracts and Organizations: Perspectives from New Institutional Economics,
Cheltenham, UK and Northampton, MA, USA: Edward Elgar, pp. 367–75.
Berkes, F. (ed.) (1989), Common Property Resources – Ecology and Community Based
Sustainable Development, London: Belhaven Press.
Berkes, F. and C. Folke (1998), ‘Linking Social and Ecological Systems for Resilience and
Sustainability’, in F. Berkes and C. Folke (eds), Linking Social and Ecological Systems:
Management Practices and Social Mechanisms for Building Resilience, Cambridge, UK:
Berkes, F. and M. Taghi Farvar (1989), ‘Introduction and Overview’, Common Property
Resources: Ecology and Community-based Sustainable Development, London: Belhaven Press.
Bhatia, A. (1999), ‘Governance and Management of Community Forestry’, Management of
Mountain Commons in the Hindu Kush Himalayas, Newsletter No. 35, Kathmandu, Nepal:
International Centre for Integrated Mountain Development.
Birner, R. and H. Wittmer (2000), ‘Co-management of Natural Resources: A Transaction
Cost Economic Approach to Determine the Efficient Boundary of the State’, in Proceedings
of International Symposium of New Institutional Economics, 22–24 September 2000,
Tubingen, Germany.
Bromley, D.W. (1991), Environment and Economy: Property Rights and Public Policy,
Oxford, UK: Oxford University Press.
Cheung, S.N. (1969), ‘Transaction Costs, Risk Aversion and the Choice of Contractual
Arrangements’, Journal of Law and Economics, 12(1), 23–45.
Cheung, S.N. (1970), ‘The Structure of a Contract and the Theory of a Non-exclusive
Resource’, Journal of Law and Economics, 13(1), 49–70.
Cheung, S.N. (1983), ‘The Contractual Nature of the Firm’, Journal of Law and Economics,
26, 1–22.
Cheung, S.N. (1987), ‘Common Property Rights’, in J. Eatwell, M. Milgate and P. Newman
(eds), The New Palgrave: A Dictionary of Economics, volume 1, London: Macmillan.
Ciracy-Wantrup, S.V. (1971), ‘The Economics of Environmental Policy’, Land Economics,
47(1), 36–45.
Ciracy-Wantrup, S.V. and R.C. Bishop (1975), ‘Common Property as a Concept in Natural
Resource Policy’, Natural Resource Journal, 15(4), 713–27.
Coase, R. (1937), ‘The Nature of the Firm’, Economica, 4(16), 386–406.
Coase, R. (1960), ‘The Problem of Social Cost’, Journal of Law and Economics, 3(1), 1–44.
Coleman, J.S. (1990), Foundations of Social Theory, Cambridge, MA: Belknap Press.
Cox, S.J.B. (1985), ‘ No Tragedy on the Commons’, Environmental Ethics, 7(1), 49–61.
Crocker, T.D. (1971), ‘Externalities, Property Rights and Transactions Costs: An Empirical
Study’, Journal of Law and Economics, 14(2), 451–64.
Dahlman, C.J. (1979), ‘The Problem of Externality’, Journal of Law and Economics, 22(1),
141–62.
Dasgupta, P. (1999), ‘Economic Progress and the Idea of Social Capital’, in P. Dasgupta
and I. Serageldin (eds), Social Capital: A Multifaceted Perspective, Washington, DC: The
World Bank.
Dasgupta, P.S. and G.M. Heal (1979), Economic Theory and Exhaustible Resources,
Dasgupta, P. and K.G. Maler (1991), ‘The Environment and Emerging Development Issues’,
Proceedings of the World Bank Annual Conference on Development Issues.
Davies, J. and M. Richards (1999), ‘The Use of Economics to Assess Stakeholders’ Incentives
in Participatory Forest Management: A Review’, European Union Tropical Forestry
Paper No. 5, Overseas Development Institute, London.
Dayton-Johnson, J. (2000), ‘Determinants of Collective Action on the Local Commons: A
Model with Evidence from Mexico’, Journal of Development Economics, 62(1), 181–208.
Demsetz, H. (1964), ‘The Exchange and Enforcement of Property Rights’, Journal of Law
and Economics, 70(4), 414–30.
Demsetz, H. (1967), ‘Towards a Theory of Property Rights’, American Economic Review,
52(2), 347–79.
Dove, M.R. (1993), ‘A Revisionist View of Tropical Deforestation and Development’,
Environment Conservation, 20(1), 17–24.
Drake, L., P. Bergstrom and H. Svedsater (1999), ‘Farmers’ Attitude to and Uptake of
Countryside Stewardship Policies’, draft final report to the STEWPOL Project, Chapter 5,
FAIRI/CT95/0709, University of Uppsala, Sweden.
Drennan, L.G. (2000), ‘The Efficient Mix of Performance Controls: A Transaction Cost
Economics Perspective’, Proceedings of International Symposium of New Institutional
Economics, 22–24 September 2000, Tubingen, Germany.
Eggertsson, T. (1990), Economic Behaviour and Institutions, Cambridge, UK: Cambridge
University Press.
Falconer, K. (2000), ‘Farm-level Constraints on Agri-environmental Scheme Participation:
A Transactional Perspective’, Journal of Rural Studies, 16(3), 379–94.
Feeny, D., F. Berkes, B.J. McCay and J.M. Acheson (1998), ‘The Tragedy of the Commons:
Twenty-two Years Later’, in J. Baden and D. Noonan (eds), Managing the Commons,
Bloomington, IN: Indiana University Press.
Fenoaltea, S. (1984), ‘Slavery and Supervision in Comparative Perspectives: A Model’,
Journal of Economic History, 44(3), 635–68.
Gibbs, J.N. and D.W. Bromley (1989), ‘Institutional Arrangements for Management of
Rural Resources: Common Property Regimes’, in F. Berkes (ed.), Common Property
Resources: Ecology and Community-based Sustainable Development, London: Belhaven
Press.
Gordon, H.S. (1954), ‘The Economic Theory of a Common Pool Resource: The Fishery’,
Journal of Political Economy, 62(2), 124–42.
Griffin, R.C. (1991), ‘The Welfare Analytics of Transaction Costs, Externalities and
Institutional Choice’, American Journal of Agricultural Economics, August, 601–14.
Grima, A.P.L. and F. Berkes (1989), ‘Natural Resources: Access, Rights-to-use and
Management’, in F. Berkes (ed.), Common Property Resources – Ecology and Community
Based Sustainable Development, London: Belhaven Press.
Guggenheim, S. and J. Spears (1991), ‘Sociological and Environmental Dimension of Social
Forestry Projects’, in M.M. Carnea (ed.), Putting People First: Sociological Variables in
Rural Development, New York: Oxford University Press, pp. 304–39.
Hanna, S. (1995), ‘Efficiencies of User Participation in Natural Resource Management’,
in S. Hanna and M. Munasinghe (eds), Property Rights and the Environment: Social and
Ecological Issues, The Beijer International Institute of Ecological Economics and The
World Bank.
Hanna, S. and M. Munasinghe (1995), ‘An Introduction to Property Rights and the
Environment’, in S. Hanna and M. Munasinghe (eds), Property Rights and the Environment:
Social and Ecological Issues, The Beijer International Institute of Ecological Economics
and The World Bank.
Hanna, S., C. Folke and K. Maler (1995), ‘Property Rights and Environmental Resources’,
in S. Hanna and M. Munasinghe (eds), Property Rights and the Environment: Social and
Ecological Issues, The Beijer International Institute of Ecological Economics and The
World Bank.
Hardin, G. (1968), ‘The Tragedy of Commons’, Science, 162(3859), 1243–8.
Hill, I. and D. Shields (1998), ‘Incentives for Joint Forest Management in India: Analytical
Methods and Case Studies’, World Bank Technical Paper No. 394, Washington, DC: The
World Bank.
Johnson, O.E.G. (1972), ‘Economic Analysis, the Legal Framework and Land Tenure
Systems’, Journal of Law and Economics, 15(1), 259–76.
Kanbur, R. (1992), ‘Heterogeneity, Distribution and Cooperation in Common Property
Resource Management’, background paper for the 1992 World Development Report, Office
of the Vice President, Development Economics, The World Bank, Washington, DC.
Kant, S. (1998), ‘Community Management: An Optimal Resource Regime for Tropical
Forests’, Department of Economics, Faculty of Forestry, University of Toronto, mimeo.
Kant, S. (2000), ‘A Dynamic Approach to Forest Regimes in Developing Economies’,
Keohane, R.O. and E. Ostrom (eds) (1995), Local Commons and Global Interdependence,
Heterogeneity and Cooperation in Two Domains, London: Sage.
Kherallah, M. and J. Kirsten (2001), ‘The New Institutional Economics: Applications
for Agricultural Policy Research in Developing Countries’, International Food Policy
Research Institute (IFPRI), Washington, DC.
Kumm, K.I. and L. Drake (1998), ‘Transaction Costs to Farmers of Environmental
Compensation’, unpublished report, Department of Economics, SLU, University of
Uppsala, Sweden.
Kuperan, K., N. Mustapha, R. Abdullah, R.S. Pomeroy, E. Genio and A. Salamanca (1998),
‘Measuring Transaction Costs of Fisheries Co-management’, paper presented at the 7th
Biennial Conference of the International Association for the Study of Common Property,
Vancouver, www.indiana.edu/~iasip/Drafts/kuperan.pdf, accessed 18 August 2009.
Leffler, K. and R. Rucker (1990), ‘Transaction Costs and the Efficient Organization of
Production: A Study of Timber Harvesting Contracts’, Journal of Political Economy,
99(5), 1060–87.
Libecap, G. (1989), Contracting for Property Rights, New York: Cambridge University
Press.
Lin, J.Y. and J.B. Nugent (1995), ‘Institution and Economic Development’, in J. Behrman
and T.N. Srinivasan (eds), Handbook of Development Economics, Vol. III, Amsterdam:
Elsevier, pp. 2303–70
McKean, M. (1992a), ‘Success on the Commons: A Comparative Examination of Institutions
for Common Property Resource Management’, Journal of Theoretical Politics, 4(3),
247–81.
McKean, M. (1992b), ‘Management of Traditional Common Lands (Iriachi) in Japan’, in
National Research Council Proceedings of the Conference on Common Property Resource
Management, Washington, DC: National Academic Press, pp. 533–89.
McKean, M. (2000), ‘Common Property: What It Is, What Is It Good for, and What
Makes It Work’, ‘Forests, People and Governance: Some Initial Theoretical Lessons’,
in C.C. Gibson, M.A. McKean and E. Ostrom (eds), People and Forests, Communities,
Institutions, and Governance, Cambridge, MA: The MIT Press.
Meinzen-Dick, R. and A. Knox (1999), ‘Collective Action, Property Rights and Devolution
of Natural Resource Management: A Conceptual Framework’, paper presented at the
International Workshop on Collective Action, Property Rights and Devolution of Natural

Resource Management, Exchange of Knowledge and Implications for Policy, 21–25 June,
the Philippines, http://www.capri.cgiar.org/pdf/capriwp11.pdf, accessed 19 August 2009.
Menard, C. (2000), ‘Enforcement Procedures and Governance Structures: What
Relationship?’, in C. Menard (ed.), Institutions, Contracts and Organizations: Perspectives
from New Institutional Economics, Cheltenham, UK and Northampton, MA, USA
Edward Elgar.
Meshack, C.K., B. Adhikari, N. Doggart and J.C. Lovett (2006) ‘Transaction Costs of
Community Based Forest Management: Empirical Evidence from Tanzania’, African
Journal of Ecology, 44(4), 468–77.
Molinas, J.R. (1998), ‘The Impact of Inequality, Gender, External Assistance and Social
Capital on Local-level Collective Action’, World Development, 26(3), 413–31.
Nabil, M.K. and J.B. Nugent (1989), ‘The New Institutional Economics and its Applicability
to Development’, World Development, 17(9), 1333–47.
North, D.C. (1981), Structure and Change in Economic History, New York: Norton.
North, D.C. (1984), ‘Transaction Costs, Institutions and Economic History’, Journal of
Institutional and Theoretical Economics, 140, 7–17.
North, D.C. (1990), Institutions, Institutional Change and Economic Performance, New York:
North, D.C. (1997), ‘Transaction Costs through Time’, in C. Menard (ed.), Transaction Cost
Economics: Recent Developments, Cheltenham, UK and Lyme, NH, USA: Edward Elgar.
North, D.C. and R.P. Thomas (1977), ‘The First Economic Revolution’, Economic History
Review, 30(2), 229–41.
Oakerson, R.J. (1986), ‘A Model for the Analysis of Common Property Problems’, Common
Property Resource Management, proceedings of a conference prepared by the Panel on
CPRM Office for International Affairs, Washington, DC: National Research Council
National Academy Press, pp. 13–30.
Olson, M. (1965), The Logic of Collective Action: Public Goods and the Theory of Groups,
Cambridge, MA: Harvard University Press.
Olsson, O. (1999), ‘A Microeconomic Analysis of Institutions’, Working Paper in Economics
No. 25, Department of Economics, Gothenburg University.
Ostrom, E. (1987), ‘Institutional Arrangements for Resolving the Commons Dilemma: Some
Contending Approaches’, in B.J. McCay and J.M. Acheson (eds), The Question of the
Commons, Tucson: University of Arizona Press, pp. 250–65.
Ostrom, E. (1990), Governing the Commons, Cambridge, UK: Cambridge University Press.
Ostrom, E. (1998), ‘A Behavioural Approach to the Rational Choice Theory of Collective
Action’, American Political Science Review, 92(1), 1–22.
Ostrom, E. and R. Gardner (1993), ‘Coping with Asymmetries in the Commons: Self-
governing Irrigation Systems Can Work’, Journal of Economic Perspectives, 7(4), 93–112.
Pelletier-Fleury, N., V. Fargeon, J. Lanoe and M. Fardeau (1997), ‘Transaction Costs
Economics as a Conceptual Framework for the Analysis of Barriers to the Diffusion of
Telemedicine’, Health Policy, 42, 1–14.
Posner, R.A (1987), ‘The Law and Economics Movement’, American Economic Review,
77(2), 1–13.
Putnam, R. (1993), Making Democracy Work: Civic Traditions in Modern Italy, Princeton,
NJ: Princeton University Press.
Richards, M. (1997), ‘Common Property Resource Institution and Forest Management in
Latin America’, Development and Change, 28(1), 95–117.
Richards, M., K. Kanel, M. Maharjan and J. Davies (1999), Towards Participatory Economic
Analysis by Forest User Groups in Nepal, London: ODI.
Room, J. (1980), ‘Assessing the Benefits and Costs of Social Forestry Projects’, The Indian
Forester, 106(7), 445–55.
Runge, C.F. (1981), ‘Common Property Externalities: Isolation, Assurance and Resource
Depletion in a Traditional Grazing Context’, American Journal of Agricultural Economics,
63(4), 595–606.
Runge, C.F. (1984), ‘Institutions and the Free Rider: The Assurance Problem in Collective
Action’, Journal of Politics, 46, 154–89.
Schlager, E. and W. Blomquist (1998), ‘Resolving Common Pool Resource Dilemmas and
Heterogeneities Among Resource Users’, paper presented at ‘Crossing Boundaries’, the
Seventh Annual Conference of the International Association for the Study of Common
Property, 10–14 June, Vancouver, BC, Canada.
Scott, A. (1955), ‘The Fishery: The Objectives of Sole Ownership’, Journal of Political
Economy, 63(2), 116–24.
Seabright, P. (1993), ‘Managing Local Commons: Theoretical Issues in Incentive Design’,
The Journal of Economic Perspective, 7(4), 113–34.
Shelanski, H.A. and P.G. Klein (1995), ‘Empirical Research in Transaction Cost Economics:
A Review and Assessment’, The Journal of Law, Economics & Organization, 11(2),
335–61.
Smith, R.J. (1981), ‘Resolving the Tragedy of the Commons by Creating Private Property
Rights in Wildlife’, CATO Journal, 1(2), 439–68.
Smith, V.L. (1968), ‘Economics from Production of Natural Resources’, American Economic
Review, 58(3), 409–31.
Stevenson, G.G. (1991), Common Property Economics: A General Theory and Land Use
Application, Cambridge, UK: Cambridge University Press.
Sugden, R. (1984), ‘Reciprocity: The Supply of Public Goods through Voluntary
Contributions’, Economic Journal, 94(376), 772–87.
Tang, S.Y. (1992), Institutions and Collective Action: The Case of Irrigation Systems, San
Francisco: ICS Press.
Uphoff, N. (1998), ‘Community-based Natural Resource Management: Connecting
Micro and Macro Processes, and People with their Environment’, paper presented at
International Workshop in Community-based Natural Resource Management, 10–14
May, Washington, DC: World Bank.
Varughese, J. (1999), ‘Villagers, Bureaucrats, and Forests in Nepal: Designing Governance
for Complex Resource’, unpublished PhD dissertation, Indiana University, USA.
Varughese, J. and E. Ostrom (2001), ‘The Contested Role of Heterogeneity in Collective
Action: Some Evidence from Community Forestry in Nepal’, World Development, 29(5),
747–65.
Vätn, A. (2001), ‘Environmental Resources, Property Regimes, and Efficiency’, Environment
and Planning: Government and Policy, 19(5) 665–80.
Velded, T. (2000), ‘Village Politics: Heterogeneity, Leadership and Collective Action’,
Journal of Development Studies, 36(5), 105–34.
Wallace, M.B. (1981), ‘Solving Common-property Resource Problems: Deforestation in
Nepal’, unpublished PhD Dissertation, Harvard University, Cambridge, MA.
Wang, N. (2001), ‘The Coevolution of Institutions, Organizations, and Ideology: The
Longlake Experience of Property Rights Transformation’, Politics and Society, 29(3),
415–45.
Williamson, O.E. (1979), ‘Transaction-cost Economics: The Governance of Contractual
Relations’, Journal of Law and Economics, 22(2), 233–61.
Williamson, O.E. (1985), The Economic Institution of Capitalism, New York: The Free
Press.
Williamson, O.E. (1991), ‘Comparative Economic Organization: The Analysis of Discrete
Structural Alternatives’, Administrative Science Quarterly, 36(2), 269–96.
Williamson, O.E. (1999), ‘Public and Private Bureaucracies: A Transaction Cost Economics
Perspective’, Journal of Law, Economics, and Organizations, 15(1), 306–41.
Williamson, O.E. (2000), ‘The New Institutional Economics: Taking Stock, Looking Ahead’,
Journal of Economic Literature, 38(3), September 2000, 595–613.
Zak, P.J. and S. Knack (2001), ‘Trust and Growth’, The Economic Journal, 111(470),
295–321.
6. Analysing dominant policy perspectives –
the role of discourse analysis
David G. Ockwell and Yvonne Rydin
Introduction
The last decade has seen a ‘linguistic turn’ within policy analysis
(Edelman, 1988; Rydin, 1998, 1999; Hastings, 1999) as it becomes increas-
ingly accepted that language use and appeals to different discourses by
various actors in the policy-making sphere have a direct influence on
the nature of any policy. In this chapter we explore how to undertake a
discursive policy analysis. Rather than focus on the theoretical debates
on this approach, we address the practical problems and potential for
undertaking discourse analysis of environmental policy through a case
study of the policy governing anthropogenic fire in Cape York Peninsula,
Queensland, Australia. We begin by exploring the rationale for and ben-
efits of using discourse analysis. Then we emphasize the need to find an
appropriate ‘middle range’ theory for application in any specific context.
To illustrate our point, two alternative frameworks for undertaking such
an analysis are outlined. We then apply these frameworks in detail to our
case study and use them to understand why a particular policy perspective
has dominated fire policy in Cape York. This demonstrates the nature of
the insights that the two approaches facilitate and provides the opportu-
nity for exploring the methodological difficulties and practicalities of such
an analysis.
The arguments for a discursive approach to policy analysis

The term ‘discourse’ is both complex and contested. It has multiple roots
in the social sciences and humanities (Hastings, 1999, 2000). Dryzek (1997,
p. 8) defines a discourse as ‘a shared way of apprehending the world.
Embedded in language it enables subscribers to interpret bits of informa-
tion and put them together into coherent stories or accounts. Each dis-
course rests on assumptions, judgements and contentions that provide the
basic terms for analysis, debates, agreements and disagreements’.
The key point of attending to discourse within environmental policy
analysis is to respond to the assumption that policy language is a neutral
medium through which ideas and an objective world can be represented
and discussed (Darcy, 1999). This assumption overlooks the extent to
168
The role of discourse analysis 169
which policy is contingent on social constructions of reality and the way

the expression of policy issues will both be the result of power relations,
ideological contestations and political conflict, and actively shape such
relations, contestations and conflicts. Advocates of discourse analysis
claim that it is crucial to examine and explain how language is used in
such contexts in order to reveal aspects of social and political processes
that were previously obscured or misunderstood. Furthermore, discourse
analysis can serve to illuminate the way in which entrenched policy posi-
tions are to some extent sustained by the way in which policy problems are
linguistically framed (Scrase and Ockwell, 2009).
More specifically, three distinct benefits of policy discourse analysis can
be identified (Rydin, 2005). First, it enables one to understand different
policy actors’ perspectives and their self-presentation within the policy
process. These will be expressed through the language that policy actors
use and can help explain how different actors operate within policy con-
texts. An actor may use specific forms of language that are particularly
appropriate and effective in a given policy context; by contrast, the weak
situation of community representatives at formal hearings and inquiries
is often at least partly due to their lack of command of the appropri-
ate formal language. There are also strong links between the identity of
actors and their use of language; identity is constructed through linguistic
means. This has implications for how actors are categorized and treated
in policy contexts; what it is to be actor X in a certain policy situation is
discursively constructed. The argument here is that language is not just
a medium of interaction but is also constitutive of actors, their identities
and their values. Actors’ values, therefore, cannot be seen in terms of their
hard-wired preferences but rather as generated through debate, discus-
sion and enunciation of those values (DeLuca, 1999). Furthermore, this is
not an individual undertaking but is inherently social, occurring through
interactions between actors.
Second, the attention to language allows consideration of how actors’
power is at least in part discursive. Interaction between actors then becomes
not just a series of encounters in which interests are balanced against or
do battle with each other on the basis of their material resources. Rather
it points to how the language that actors actively or unconsciously use in
their communications with others is involved in persuasion and ration-
alization and influences the dynamics of policy debates. This is not just a
matter of individual actors’ capacity in using language. Linkages with pre-
vailing societal discourses will also be important in supporting a particular
actor’s case. The reliance on various forms of economic rationality are a
case in point, where the discursive reliance on a widely used and referenced
argument about the importance of economic growth carries weight quite
independently from the material power of economic actors and their skill
in persuasion (Rydin, 2003).
Another way of considering the inter-relationship between language
and actors’ interactions with one another is to see language as also con-
stitutive of the incentives facing actors – the costs and benefits (monetary
and otherwise) that they take into account in deciding on their behav-
iour. For example, researchers have suggested that reputation can be a
key factor shaping behaviour (Ostrom, 1990; Chong, 1991). Actors may
seek to promote and protect a good reputation and avoid behaviour that
is going to expose them to public shame and blame. But reputation is a
social variable constructed through social communicative interactions.
The detailed language of interactions will be central to assessments (by the
actor concerned and others) of whether a reputation is being damaged or
enhanced.
And third, as well as understanding policy actors and the dynamics
of policy processes more fully, the discursive dimension allows the pos-
sibility for devising new modes of communication to achieve normatively
better policy outcomes. There has been a lot of emphasis on creating
more inclusive and deliberative spaces for communication (Burgess et al.,
1998; Hillier, 1998; Healey, 1999; Mason, 1999). Much of the literature on
these policy innovations has arisen from an engagement with the notion
of policy as an inherently discursive process. There has, however, been a
tendency towards an overly procedural approach in devising new spaces
for communication. Graham Smith is one of the most insightful writers on
deliberation but in his discussion of institutional design and deliberation
(2003), he concentrates on procedures and decision rules, specifically how
to ensure equality of voice, defence of deliberation against strategic action
and sensitivity to the scope, scale and complexity of environmental issues.
But how do these procedures and decision rules change the communica-
tion within deliberation? This can only be understood through considering
the detailed language of that communication. The lesson of the linguistic
turn is that communication between actors is not just a matter of how that
communication is arranged. The language of the interaction also needs to
be considered.
Even more restrictively, deliberation is often equated with communica-
tion as if this not only characterizes deliberation but distinguishes it from
other types of policy intervention. This fails to see the communicative
dimension of all policy work (Majone, 1989) and, furthermore, the linguis-
tically mediated nature of all social activity. Language does play a pivotal
role in deliberation, but it is also implicated in many other institutional
arenas. This can provide insights into how language expresses values and
enables or disables agreement, including consensus.
The roots of this lack of attention to the details of language even within
those analysts of policy who see policy as essentially communicative,
derives from the Habermasian roots of most work on deliberative democ-
racy and its planning applications (Dryzek, 1990, 2000; Healey, 1999).
This identified a specific potential within communication between actors –
that is, the potential to create consensus. Where actors engage in commu-
nication with a performative stance towards mutual understanding, then
consensus between parties is immanent in the communication. The major
constraint on achieving such a consensus is pinpointed as the absence of
an ideal speech situation in which communicative rationality can hold
sway. This has led to the emphasis being placed on the circumstances
within which communication occurs rather than the nature of the commu-
nication in linguistic terms. This has been confirmed in Habermas’s most
recent work (1996). In our chapter we want instead to turn back towards
the language of policy and the detailed analysis of that language in specific
situations.
There is one particular problem within environmental policy that this
approach seems particularly well suited to address. This concerns the mis-
match between complex environmental problems and simple dominant
policy responses. This simplicity is at odds with the inherent variability
and complexity of the ecosystems whose healthy functioning such policy
aims to sustain. It ignores the fact that environmental problems are char-
acterized by a high degree of uncertainty. On one level, uncertainty exists
as a result of scientists’ incomplete understanding of ecosystem function-
ing. Additionally, however, environmental problems are invariably linked
to issues of resource distribution. As a result, there are always economic,
social and political implications of any environmental policy (see, for
example, Wheeler and McDonald, 1986; Rees, 1990; see also, for example,
Flournoy, 1993). As Dryzek (1997) highlights, when ecological systems
interact with economic, social and political systems through the policy
process, the level of uncertainty associated with environmental problems
is greatly magnified. Hajer and Wagenaar (2003) refer to this as policy-
making under conditions of ‘radical uncertainty’.
Increasing evidence suggests that such uncertainty is often reflected in
inappropriate environmental policies due to the application of standard-
ized management techniques that ignore the spatial, temporal, social,
economic and political complexity of environmental problems. Leach and
Mearns (1996) give the example of the woodfuel crisis in Africa, which is
widely perceived as a classic example of a supply gap where demand for
woodfuel exceeds supply. This has been met by a standardized response of
mass tree planting by governments, NGOs and inter-governmental organi-
zations. The basic assumptions that define the supply gap, however, ignore
more subtle issues such as the fact that most woodfuel comes from clearing
wood for agriculture or from lopping branches valued for fruit and shade.
From a broader perspective there is not one big problem of energy supply
but many smaller problems of command over trees and their products to
meet a wide range of basic needs. The range of policy solutions is therefore
equally diverse. Why is it then that one policy outcome can emerge and be
sustained in the face of conflicting evidence? Discourse analysis offers a
valuable approach to answering this question.
Theoretical perspectives on discourse analysis

Just as the term ‘discourse’ itself is a contested concept used in different
ways, so are there a variety of different perspectives on discourse analysis,
both methodologically and theoretically. One of the key distinctions that
runs through the literature is between those approaches that are derived
from a Foucauldian perspective and those derived from a Habermasian
one. The two methodologies that we explore in this chapter derive from
both sides of this distinction: Maarten Hajer, who developed his frame-
work from an engagement with Foucault’s work; and John Dryzek, whose
interest in discourse derives from his earlier work on Habermas and nor-
mative theories of deliberative democracy. These ‘middle-range’ theorists
are important because they have sought to develop and apply Foucault’s
and Habermas’s broader ideas to environmental issues and specifically
to environmental policy. They have done considerable work in qualify-
ing and operationalizing Foucault’s and Habermas’s theories. In doing
so, they have to some extent modified these original theories but retain
the essentials of the Foucauldian and Habermasian perspectives. We will
explore these two approaches next, before demonstrating how they can be
applied to a specific case study.
Hajer
In the development of the study of discourse Foucault’s work has been
pivotal. Through the study of the history of sexuality, madness and the
disciplinary basis of the academy, Foucault, referring to what he called
power/knowledge, developed the idea that knowledge, and hence dis-
course, is a reflection of power within society. As such, language is seen
as the operation of power (Foucault, 1980, 1984; Bevir, 1999; Rydin,
1999; Hastings, 2000; Watt and Jacobs, 2000). Hajer has sought to work
within a Foucauldian framework, in terms of engaging with the combined
concept of power/knowledge, while adjusting it to the problem of under-
standing environmental policy situations. Hajer’s work focuses on the
discursive nature of environmental policy-making (Hajer, 1995; Hajer and
Wagenaar, 2003). Discourse is seen as constituting both text and practice
with a strong emphasis on social constructivism running throughout

(Dryzek, 1995; Hajer, 1995; Bakker, 1999; Keeley and Scoones, 2000;
Richardson and Jensen, 2000). In this view discourses are produced both
through individual activities and institutional practices that reflect par-
ticular types of knowledge. Discourses are therefore actively produced
through human agencies that undertake certain practices and describe
the world in certain ways. Actors are not, however, seen as acting within
a vacuum. Discourses simultaneously have structuring capabilities. They
provide parameters within which people act and mould the way actors
influence the world around them (Hajer, 1995; Keeley and Scoones,
2000).
In Hajer’s view, politics is a struggle for discursive hegemony in which
actors struggle to achieve ‘discursive closure’ by securing support for their
definition of reality. There is a significant Foucauldian influence within
Hajer’s work in terms of the regulatory power of discourses as they act to
select appropriate and meaningful utterances and actions within a strug-
gle for hegemony in the policy-making process (Foucault, 1979, 1990;
Buttel, 1997; Rydin, 1998; Richardson and Jensen, 2000). The notion of
‘story-lines’ is brought in to describe the common adoption of narratives
through which elements from many different spheres are combined to
provide actors with symbolic references that imply a common understand-
ing (Hajer, 1995; Rydin, 1999). Essentially, the assumption is that actors
don’t draw on a comprehensive discursive system, instead this is evoked
through story-lines. By uttering a specific word or phrase, for example,
‘global warming’, a whole story-line is in effect reinvoked. Story-lines can,
in this way, therefore act to define policy problems.
The widespread adoption of a story-line results in the formation of
‘discourse coalitions’ where groups of actors are drawn to particular
story-lines as they represent common interests (Hajer, 1995; Bakker, 1999;
Rydin, 1999). These actors might not have ever met and might apply dif-
ferent meanings to a story-line, but in the struggle for discursive hegemony
that is assumed to play out within the policy-making process, story-lines
form the ‘discursive cement’ that keeps the discourse coalition together
by producing ‘discursive affinities’. Particularly strong discursive affin-
ity is referred to as ‘discursive contamination’. The Foucauldian basis of
Hajer’s approach views story-lines as playing an essential role in position-
ing actors. They add credence to the claims of certain groups and render
those of other groups less credible. Story-lines therefore act to create social
and moral order within a given domain by serving as devices through
which actors are positioned and ideas of blame, responsibility and urgency
are ascribed.
The social constructivist approach emphasizes the role of institutional
arrangements in structuring discourses, forming routine understandings

where complex research is often reduced to visual reproduction or catchy
one-liners that ignore uncertainty and entail significant loss of meaning.
Routine forms of discourse therefore express a continuous power relation-
ship that is particularly effective in that it avoids confrontation. The use of
the term ‘sustainable development’ in contemporary British public policy
arguably provides an example of this.
To shape policy, a new discourse must both dominate public discussion
and policy rhetoric and penetrate the routines of policy practice through
institutionalization within laws, regulations and routines (Hajer, 1993;
Nossiff, 1998; Healey, 1999). In terms of policy change then, promoting a
new story-line is a difficult task involving dismantling previous story-lines
and confronting the interests of those who were able to achieve promi-
nence for their claims and viewpoint originally (Rydin, 1999). Discourse
analysis from Hajer’s perspective is a method to shed light on the social
and cognitive basis of the way in which policy problems are constructed
(Hajer, 1995), with analysis focused on the socio-cognitive processes in
which discourse coalitions are established. He puts emphasis on the con-
stitutive role of discourse in political processes, but assigns a central role
to discoursing subjects, although in the context of a duality of structure.
Social action is seen as stemming from human agency, however, social
structures of various sorts both enable and constrain their agency. It is
therefore possible for agents to accomplish policy change through discur-
sive interaction within the context of these structures, but this inherently
requires deconstructing the discursive hegemony that existing dominant
political interests have achieved.
Dryzek
John Dryzek is well known as a normative political theorist. He has sought
to apply Habermas’s concept of communicative rationality and delibera-
tive democracy to the specific issue of the environment (Dryzek, 2000),
a development that has been welcomed as a major advance in political
theory. Here, though, the focus is on his work in analysing environmental
discourse. In this work he explicitly counters the Foucauldian approach
and instead adopts a more agency-centred model. Dryzek sees discourse
and power as interconnected in all kinds of ways, whereas Foucault would
deny such a distinction between power and discourse as discourse is the
operation of power (Dryzek, 1997). Furthermore, Dryzek sees constraints
on the power of discourse, as powerful actors may override developments
at the discursive level by ignoring them in terms of policy. Alternatively,
discourses may be absorbed to suit the interests of a firm or government
(ecological modernization springs to mind). Another constraint may arise
from the need for capitalist governments to fulfil a number of basic func-
tions irrespective of discourses that may have been captured by government
officials, especially continued economic growth. Within his rejection of the
Foucauldian discursive approach, Dryzek points to the very existence of
authors such as Foucault as evidence that individuals subject to discourses
are able step back and make comparative assessments and choices across
different discourses. Dryzek also rejects the hegemonic terms used by
Foucauldians to describe discourse. He asserts that variety is as likely as
hegemony, with the disintegration of the previously hegemonic discourse
of industrialism since the 1960s as evidence of just such variety.
Dryzek sets out his approach to discourse analysis with the aim of
advancing ‘analysis of environmental affairs by promoting critical com-
parative scrutiny of competing discourses of environmental concern’
(Dryzek, 1997, p. 20). He is thus interested in explaining how environmen-
tal discourses inform political programmes. A four-fold typology is put
forward where ways of thinking about environmentalism are character-
ized in terms of their departure from the discourse of industrialism (Elliott,
1999). These departures can be reformist or radical, prosaic or imaginative
and result in the identification of four main categories of environmental
thought. These four strands are categorized as follows. The discourse of
environmental ‘problem-solving’ is prosaic and reformist. It takes the
political-economic status quo as given but in need of adjustment to cope
with environmental problems, especially via public policy. The discourse
of ‘survivalism’, popularized by the Club of Rome (Meadows et al., 1972)
is also prosaic, but radical. It is radical because it seeks wholesale redis-
tribution of power within the industrial political economy and wholesale
reorientation away from perpetual economic growth to avoid exhausting
natural resources and the assimilative capacity of the environment. It is
prosaic because it sees solutions in terms of options set by industrialism,
especially greater control of existing systems by administrators, scientists
and other responsible elites. The discourse of ‘sustainability’ is imaginative
and reformist, beginning in the 1980s with imaginative attempts to dissolve
the conflicts between environmental and economic values that are charac-
teristic of the discourses of problem-solving and survivalism. Finally, the
discourse of ‘green radicalism’ is also imaginative, but radical. It rejects
the basic structure of industrial society and the way the environment is
conceptualized therein. Due to such radicalization and imagination, it
features deep intramural divisions.
In common with the roots of Habermas’s work in argumentation
theory, Dryzek also adopts a rhetorical method, marrying this with a more
basic social constructivist perspective (see Rydin, 2003, ch. 2). This offers
a fairly tightly controlled comparative analytical device for identifying the
elements through which discourses construct stories. First the basic enti-
ties whose existence is recognized or contrasted must be identified. Dryzek
refers to this as the ‘ontology’ of a discourse. Second, assumptions held
about natural relationships, such as Darwinian struggle or cooperation,
should be explored. This includes hierarchies of gender, expertise, political
power, intellect, race, and so on. Propositions about agency and motiva-
tion constitute the third area of investigation. Agents may, for example,
be seen as benign, public-spirited administrators or selfish bureaucrats.
They could include enlightened citizens, rational consumers, ignorant and
short-sighted populations, and so on. Finally, the key metaphors and rhe-
torical devices that a discourse invokes should be scrutinized. This might
include, for example, metaphors such as spaceships (Boulding, 1966) that
may act as rhetorical devices to convince listeners or readers. It may also
include other devices like an appeal to widely accepted institutions or
practices such as established rights. A discourse could also accentuate
negatives such as horror stories regarding government mistakes.
In order to demonstrate the nature of the insights that Hajer and
Dryzek’s approaches facilitate and to explore the methodological difficul-
ties and practicalities of such analysis, we now apply them to a case study
of environmental policy governing anthropogenic burning in Cape York
Peninsula (Cape York), Queensland, Australia.
Discourse analysis in practice – a case study1

Cape York is situated at the north-eastern tip of Queensland, Australia
(Figure 6.1). Covering an area roughly equivalent in size to England, Cape
York has a low population density of just 18 000 people mostly concen-
trated in a few mining towns and Aboriginal reserves as well as scattered
cattle stations. Northern Australia, including Cape York, is thought to
have a long history of anthropogenic burning stretching back at least
40 000 years (some estimates date it as far back as 70 000 years), coincid-
ing with the arrival of the first Aborigines (Stocking and Mott, 1981).
The idea of ‘fire-stick’ farming was popularized by Rhys Jones (1969)
to describe the practices of indigenous land users where low-intensity,
early dry-season burning across small areas was used to drive game into
hunting grounds and increase the productivity of resource-rich areas such
as monsoon forests.2
There has been considerable controversy over the impact of Aboriginal
use of fire on the ecology of Australia. Most prominent is the debate
around whether, in tropical northern Australia, Aboriginal burning
caused the recession of earlier rainforest in favour of savanna or whether
the recession of the rainforest was in fact the result of climate change
(Flannery, 1994; Rose, 1996; Bowman, 1998, 2000; Hill, 2003). Bowman
Injinco
Wattle Hills
Weipa Lockhart River
Great
Aurukun Barrier
Reef
Cape
Pormpuraaw York
Hope Vale
Peninsula
Kowanyam Cooktown
Cairns
Figure 6.1 Map of Cape York Peninsula
(1998, p. 2) characterizes the lack of scientific consensus surrounding

this debate as ‘an inherent circular argument concerning the cause and
effect of climate change, vegetation change, and burning through the late
Quaternary’. It is, however, widely accepted that the pattern of burning in
tropical northern Australia has changed in modern times, coinciding with
the displacement of Aborigines by European settlers. Late dry-season,
high-intensity burns now define anthropogenic burning with increased
fuel loads over larger areas. This has reduced fire-sensitive vegetation in
some areas. There has also been a lack of fire in other areas, which has
enhanced fire-sensitive ecosystems (Gill et al., 1990; Bowman, 1998, 2000;
Hill, 2003).
Anthropogenic burning in Cape York provides a typical example of
such changes in burning practices. Overall, an estimated 80 per cent of the
total area of Cape York currently burns each year (Cape York Peninsula
Sustainable Fire Management Programme, personal communication,
2004). Fire-assisted pastoralism forms the dominant land use in Cape

York. Pastoralists tend to burn land to promote the growth of green grass
for their cattle to feed on. There are also extensive areas of Cape York
(approximately 14.2 per cent of the total land area of Cape York in 2004
and increasing annually – Queensland Parks and Wildlife Service, personal
communication, 2005) set aside as national park and wildlife reserves under
the control of the Queensland Parks and Wildlife Service (QPWS) who also
use fire as a significant part of their land management approach. They
justify their use of fire through a number of reasons including hazard reduc-
tion where, it is argued, burning available ground fuel (dry leaf litter and
so on) avoids the spread of wildfires later in the dry season, habitat man-
agement, including weed and pest management, and maintaining habitat
diversity. There are also several Aboriginal reserves where indigenous
communities are free to pursue their own traditional burning practices.
Despite the lack of consensus within the scientific literature regarding
the environmental impacts of fire, environmental policy in Cape York
tends to be unquestioningly pro the use of fire as a land management tool.
This can be seen in both the policies of the Queensland Rural Fire Service
who are responsible for policing the use of fire in Cape York (RFS, 2001)
and the policies of those government departments with direct jurisdiction
over the management of areas of land there (see, for example, Grice and
Slatter, 1997; Gill et al., 1999; Marlow, 2000; QPWS, 2000; see also, for
example, DNR, 2001, 2004; EPA, 2002).3
The Permit to Light Fire system operated by the Queensland Rural Fire
Service (Queensland State Government, 1990) provides that landhold-
ers can, in theory, be prosecuted if they light a fire outside the terms of a
Permit. The problem, however, has always been in proving who lit a fire.
Once a fire is lit on Cape York it has the potential, especially at dry times
of year, to burn for weeks or even months across hundreds of thousands
of hectares of land, therefore affecting areas nowhere near where it was
first lit. The introduction of the Cape York Peninsula Sustainable Fire
Management Project has had some success in addressing the issue of
accountability. The project provides an online service4 where fires on Cape
York can be tracked by satellite and thus, when cloud cover does not inter-
fere, the origin of fires can sometimes be identified. The attitude amongst
most stakeholders, however, tends to be very much that if the fire is on
your land, it’s your problem whether you lit it or not. This is characterized
by a traditional saying often used by landowners on Cape York; ‘He who
owns the fuel owns the fire’ (Queensland State Government, 1990; RFS,
2001). The insinuation is that each landholder ought to engage in hazard
reduction burning to ensure that there is insufficient ground-fuel build-up
(dry leaf litter and so on) on their land to allow a fire to encroach.
As well as its role in coordinating the Permit to Light Fire system, the
Rural Fire Service also works with Cape York landholders in implement-
ing a series of ‘controlled’ burns at the beginning of each dry season. This
is done via a series of workshops held across Cape York prior to each dry
season where interested landholders can attend and request that the Rural
Fire Service carry out burning on their land. The Rural Fire Service then
flies a light aircraft along the boundaries of participating properties and
drops incendiary bombs that are intended to burn a series of firebreaks
between each property. The rationale for this practice is one of hazard
reduction. As outlined above, the rationale is that by burning the available
ground fuel on property perimeters, the spread of wildfires later in the dry
season might be avoided. Some stakeholders, however, are critical of this
practice as being too indiscriminate and not accounting for environmental
considerations in terms of whether the various affected ecosystems are
able to cope with regular, or indeed, any fire. Indeed, as Russell-Smith
et al. (2003) highlight, as well as there being no consensus over what fire-
oriented landscape management should aim for, neither is it clear whether
humans have the tools or resources to implement particular regimes over
a large spatial scale.
When applying for a Permit to Light Fire, a landholder is required to
disclose to the local Fire Warden any arrangement they have with other
government agencies that obliges them to protect some aspect of their
property from fire for environmental reasons or otherwise. Such arrange-
ments are rare but do include agreements between landholders and the
Queensland Department of Natural Resources, Mines and Energy to
undertake burning to kill back weeds such as rubber-vine (DNR, 2001),
or, in a few examples, to maintain the habitat of the endangered golden
shouldered parrot, Psephotus chrysopterygius, which relies on late dry-
season fires to create necessary nesting conditions (Crowley et al., 2003).
The only instance where such an agreement might limit burning is where a
public road runs through a property. There is only one main, unsurfaced
dirt road in Cape York so this is rarely an issue. Once a landholder has
informed their Fire Warden of any such agreement, it is then up to the
Warden to consider this when detailing the conditions of the Permit to
Light Fire. Wardens are all local stakeholders themselves chosen by the
Rural Fire Service on the basis of assumed local knowledge of their area
of responsibility (areas of responsibility range from just the Warden’s own
land to including theirs and a few of their neighbours’ properties). The
Rural Fire Service states that ‘the local volunteer Fire Warden should not
be responsible for policing environmental issues’ (RFS, 2001, 1-1) on the
basis of them not possessing sufficient knowledge to do so.
In addition to the Rural Fire Service, Queensland’s Environmental
Protection Agency also supports the pro-burning policy discourse. This

agency is responsible for achieving ‘ecologically sustainable development’
under the terms of the Environmental Protection Act 1994 (Queensland
State Government, 1994). The Queensland National Parks and Wildlife
Service is part of the Environmental Protection Agency with direct respon-
sibility for the management of National Parks. As outlined above, the
Environmental Protection Agency actively uses fire as a land management
tool on National Park land as part of their land management policy. This
is also in line with the rationale for the Department of Natural Resources,
Mines and Energy’s fire-relevant policies (Marlow, 2000; DNR, 2001).
The pro-burning policy stance in Cape York does not, however, have
unanimous support from all stakeholders. Whilst many stakeholders are
pro-burning, including Aboriginal communities, pastoralists and govern-
ment scientists whose rationales for burning were summarized above,
there are also two key stakeholder groups who are primarily anti-burning.
These are the residents of Wattle Hills (a self-sufficiency community pursu-
ing sustainable forestry practices on their 35 650 ha property), and several
independent scientists who cite a growing body of scientific and anecdotal
evidence that questions the environmental sustainability of the dominant
pro-burning policy paradigm in Cape York (see, for example, Ockwell and
Lovett, 2005). Cape York stakeholders can therefore be seen as polarized
between two opposing discourses of ‘pro-’ and ‘anti-’ burning.
In order to demonstrate the value of discourse analysis in understand-
ing the policy dominance of the pro-burning discourse in Cape York, we
analyse primary data in the form of the transcript of a seminar hosted
by the Cairns and Far North Environment Centre (CAFNEC) in 1992
entitled ‘Tropics Under Fire. Fire Management On Cape York Peninsula’
(CAFNEC, 1992). The seminar invited stakeholders to come together and
give short, 20-minute presentations outlining their views on the use of fire
in Cape York. Overall, ten presentations were made by the stakeholders
that were present. This seminar was held some time ago and other seminars
have been held since, however, this particular seminar has been chosen for
analysis for two reasons. First, analysis of later conferences and extensive
consultation with Cape York stakeholders has demonstrated that there
has been little change in attitude since the 1992 seminar was held. Second,
the 1992 seminar constituted the widest and most equally proportioned
representation of the various stakeholder groups. The transcript thus pro-
vides a useful summary of both the pro- and anti-burning discourses from
the perspectives of all the key stakeholders and interest groups.
Double close-reading of the transcript of the seminar by both authors
forms the basis of the analysis presented here. Both Hajer’s Foucauldian
and Dryzek’s Habermasian frameworks for undertaking discourse analysis
were then applied. As the analysis demonstrates, the discursive construc-

tion of the burning issue is more complex and nuanced than the two
broadly opposing pro- and anti-burning positions imply. We begin our
discourse analysis by applying Hajer’s Foucauldian-inspired framework
before moving on to examine what Dryzek’s Habermasian framework can
add to the analysis.
Applying Hajer’s analytic framework5

In keeping with Foucault’s own work, Hajer has applied his ideas in
detailed case studies. This allows more extended engagement with the
ideas he has developed than is possible here, in the context of a discourse
analysis of a specific text. Hence, the emphasis of the analysis will be on the
particular concept of story-lines. It must therefore be acknowledged that
some aspects of the Foucauldian approach are underplayed. The idea that
a discourse comprises a set of practices as well as a set of representations
is not given full weight by a text-based methodology. The emphasis on
practices within a Foucauldian approach should not, however, be taken
to exclude a concern with the text and the words it comprises. The use of
the Cape York seminar transcript provides an opportunity for considering
the story-lines embedded in these words and the potential for discourse
coalitions thereby created.
The main story-lines adopted by the key actors are set out in Table 6.1.
There is the clear distinction between two opposing discourses, constructed
as pro- and anti-burning. The strongly drawn distinction sets the context
or frame for the policy discussion and, as such, it can support the argu-
ment for two discourse coalitions coalescing around these two opposition
perspectives on the use of fire. In the ‘pro’ coalition are the Aborigines, the
government scientists and the pastoralists, while the residents of Wattle
Hills and the independent scientists make up the ‘anti’ coalition.
It might be argued that the anti-burning discourse is placed in the posi-
tion of challenging the established pro discourse, a task that is bound
to require additional discursive resources as the existing policy scenario
is well established. Those stakeholders who are anti-burning have to
promote a new story-line of ‘fire is undesirable’ and actively dismantle the
‘fire is desirable’ story-line. This suggestion of an embedded bias towards
the established ‘pro’ story-line is, however, somewhat undermined by the
framing of the seminar itself. The title of the seminar ‘Tropics Under Fire’
and the illustration on the front cover of the conference transcript of a bird
and other iconic animals (such as kangaroos) fleeing smoke and flames
lit by a giant, match-wielding human hand suggests that, at least within
CAFNEC (the environment centre that set up the seminar) there might be
a bias towards the ‘fire is undesirable’ story-line. Instead of relying on the
Table 6.1 Applying Hajer’s story-lines
Link to Dominant Story-lines

Oppositional
Discourses
Aborigines Pro-burning Fire as a cultural practice of traditional
owners and custodians of land, embodying
local indigenous knowledge
Linked to
History of repression, suppression and
removal of Aborigines leading to loss of
knowledge
Government Pro-burning Human intervention in natural systems
scientists needed to ensure conservation of habitats
and ecosystems
Elaborated as
Burning as promoting habitat diversity;
and
Burning as ‘fuel reduction’ preventing
larger-scale, ‘natural’ fires
Pastoralists Pro-burning Burning is environmentally beneficial
(through stabilizing ecosystems) but also
economically beneficial (through providing
fodder, safe mustering of cattle and
promoting tourism)
Wattle Hills Anti-burning Letting ‘nature take its course’ will
promote environmental quality and
economic benefits
Independent Anti-burning Scientific knowledge does not support
scientists burning; too much uncertainty about
impacts of burning
Linked to
Global stories of biodiversity
maintenance, rainforest protection and
endangered species
simple construction of the two opposition discourses, Hajer’s approach

highlights that any construction of a broad ‘anti’ or ‘pro’ discourse coali-
tion depends on how much the individual story-lines being used by actors
support or undermine each other. It is notable here, therefore, that the
three pro-burning story-lines all represent to some extent a different take
on the same underlying story about human engagement with nature.
There is a strong emphasis in the Aboriginal story-line on the historic
knowledge of how to use fire gained through millennia of active engage-

ment between Aborigines and their local environment. This is contextu-
alized with a moral claim that the Aborigines are the ‘custodians’ (p. 6)
of the land together with the anti-colonial discourse that emphasizes the
European repression of Aborigines. In this way, the Aboriginal story can
rely on strong affinities with the established cause of native title claims,
which has gained considerable political credibility since the early 1970s.
This reflects Hajer’s idea of how moral orders are established through
discourse in that blame and responsibility are attributed to European
descendants in order to justify prioritizing the Aboriginal representatives’
views on land management. In terms of social practice, it is significant
that an Aboriginal representative was, out of respect, asked to open the
seminar. But in terms of the discourse coalition analysis, the important
point is the existence of this clear narrative thread about the need for
active human–nature engagement; as such, this is available for connecting
with the other actors’ discourses.
Although couched in very different terminology, government scientists’
discourse follows the same narrative thread as that of the Aborigines.
Here, human engagement with ecosystems is promoted as necessary to
ensure conservation. Two subsidiary discourses are discernible: a story-
line of ‘habitat diversity’ (p. 36), preservation through burning and a
story-line of ‘fuel reduction’ (p. 37) where anthropogenic fire is endorsed
as a way of avoiding the catastrophic ecological consequences of naturally
occurring fires by reducing the volume of standing fuel. The basis for such
engagement is scientific knowledge as opposed to traditional knowledge
but the story is the same. By advocating ‘patch’ or ‘mosaic burning’ (p. 31),
which is thought to have been traditionally pursued by Aborigines (p. 27),
and claiming that this leads to ecologically desirable ‘stable’ vegetation
patterns, discursive affinity with the Aboriginal discourse is also exhibited.
The connection is also facilitated by constructing Aboriginal practices
in scientific terms, with reference to scientific papers on Aboriginal land
management and even the inclusion of one such paper in the transcripts
of the conference.
This narrative thread is reinforced again by the discourse of the pastor-
alists. Here, everyday economic practices are portrayed as supporting the
practice of burning in both environmental and economic terms. Again,
active engagement between humanity and nature (through burning) is seen
as beneficial in the long term. The pastoralist representatives maintain
the story-line of fire being environmentally beneficial at the same time as
emphasizing its economic desirability. By describing pastures as ‘botanical
communities’ (p. 29) they claim that fire can maintain ‘botanical stability’
(p. 32) and therefore preserve environmental ‘integrity’. This provides a
clear example of discursive affinity with the government scientists, which

is also evident in the pastoralists’ adherence to the story-line of burning
as desirable in terms of reducing wildfire risks. Discursive affinity with
Aborigines is also achieved by asserting that the economic benefits derived
by pastoralists from cattle fodder through post-fire new growth and
safe mustering of their herds are the same as those traditionally derived
by Aborigines. As well as the direct economic benefits from pastoral-
ism, burning is also endorsed as economically desirable for encouraging
tourism by clearing the ground for hiking and attracting charismatic
species such as wallabies and kangaroos to the fresh new after-growth.
So the story-lines analysis provides some basis for understanding the
strength of the pro-burning discourse coalition. It can also help understand
the weakness of the anti-burning coalition. The story-lines of this coalition
focus on opposing the dominant discourse instead of building links within
the coalition. The aim of the proponents of the anti-fire discourse tends
to focus chiefly on discrediting the government scientist story-line of fire
being ecologically desirable. The national parks agency, the QPWS, tends
to constitute the main focus for attack. The Wattle Hills representative
promotes a story-line of ‘fire as destructive to life’, which is diametrically
opposite to the government scientists’ story-line. The government ‘fire
management’ line is directly confronted and instead a non-interventionist,
‘let nature take its course’ line is promoted. Both independent scientists
advance a story-line of fire as environmentally damaging. Again, a direct
attempt is made to deconstruct the government scientists’ story-line of
fire as ecologically desirable by highlighting the degree of uncertainty sur-
rounding current scientific knowledge of the impacts of fire. Policies that
involve using fire are described as ‘stabbing in the dark’ (p. 43). The ‘fuel
reduction’ story-line is also attacked on the basis of negative impacts on
soil fertility as argued by the Wattle Hills representative. The argument
is that without fire, dead matter has a chance to decompose, resulting
in increased soil fertility, as opposed to being burnt and therefore losing
biomass through combustion.
Little discursive connection is evident between the ‘let nature take its
own course’ Wattle Hills discourse and the strong scientific rationality
of the independent scientists, with their attempts to link the anti-burning
story-line to other global environmental story-lines. While the Wattle Hills
residents highlight their own credentials as self-sufficiency pioneers, the
independent scientists have significant academic credentials and a strong
professional involvement with Cape York including, in one case, having
worked on television documentaries on bird life in the Cape. The rational,
scientific tone of their representations reflects their scientific background
with extensive intertextual reference made to scientific papers in support of
their arguments. Several popular global story-lines are invoked to support

avoiding the use of fire, including ‘biodiversity maintenance’, protection of
‘endangered species’, expansion of ‘valuable rainforests’ and maintenance
of ‘sustainable populations’ of wildlife. Little attempt is made to bridge
the ‘global’ of the scientists with the ‘local’ of the Wattle Hills residents
and yet, without this, the ‘anti’ discourse coalition remains discursively
weak.
Applying Dryzek’s analytic framework

The first part of the Dryzek framework concerns the identification of
actors’ discourses with his typology of four societal discourses on environ-
mental issues. Table 6.2 represents an attempt to classify the discourses
and sub-discourses present in the transcript in terms of Dryzek’s catego-
ries. This displays some difficulties. It was not possible to assign actors’
discourses to Dryzek’s categories in a simple and non-contestable way.
For example, the Aboriginal discourse can be considered to fall into
both the survivalism and green radicalism categories. There is a clear
radical strand to their discourse; the emphasis on the colonial history of
European settlement and repression marks it as such. While the status quo
of the established political economy is not challenged directly, it is implied
that Aboriginal rights should have priority over other concerns, includ-
ing any economic imperative. There is a tension, however, between much
of the tone of the Aboriginal presentations, which is distinctly prosaic,
and the sub-text, which emphasizes the distinctive spiritual claims of the
Aborigines. The latter is clearly imaginative not prosaic. The emphasis on
the prosaic in this particular context may be a strategic discursive decision
Table 6.2 Applying Dryzek’s discourses
Prosaic Imaginative
Reformist Problem-solving discourse Sustainability discourse
Government scientists Pastoralists
Independent scientists (less
critical discourse)
Pastoralists
Radical Survivalism discourse Green radicalism discourse
Aborigines (everyday Aborigines (underlying
management discourse) spiritual discourse)
Independent scientists (more Wattle Hills (self-sufficiency
critical discourse) discourse)
Wattle Hills (everyday
management discourse)
and/or may relate to the way that Aboriginal voices were largely repre-
sented by other semi-professionalized voices. In either case, it suggests
that the categories are not very helpful in furthering the analysis where the
Aboriginal discourse is concerned.
A similar point can be made about the discourses of the independent
scientists and the residents of Wattle Hills. The independent scientists
adopted a strong critique of contemporary industrialism. This very much
put them on the borders between a reformist and a radical discourse. It
certainly distanced them from the more firmly pro-status-quo stance of
the government scientists. And while, like the government scientists, they
remained prosaic rather than imaginative in their discourse, it did make it
difficult to simply assign the independent scientists to either the problem-
solving or survivalism category. The Wattle Hills story-line emphasized
the need to ‘let nature take its course’. Policies promoting fire were seen as
misleading and neglectful of life, reverence for life being of primary impor-
tance. This suggests a radical, imaginative departure from industrialism
that fits within Dryzek’s definition of ‘green radicalism’ (Dryzek, 1997). In
much of their more specific discussion of management practices, however,
the discourse of the Wattle Hills residents is distinctly prosaic, suggesting
a place in the survivalism category instead.
Indeed, only the government scientists (clearly a prosaic reformist
discourse of problem-solving) and the pastoralists (an imaginative but
reformist discourse of sustainability) fall clearly into one box of Dryzek’s
typology. This can be interpreted in two ways. Either the typology fails
the ‘ideal type’ test and is not a useful analytic tool. Or the lack of discur-
sive clarity of some of the actors’ discourses may be a significant factor in
explaining the pattern of discursive coalition formation and the success
or failure of individual actors to achieve discursive influence. Before con-
cluding on this, however, the second part of Dryzek’s framework deserves
consideration.
This concerns the detailed social constructivist/rhetorical analysis of
the discourses. Some, but by no means all of this analysis is summarized
in Table 6.3. What is immediately evident here is the wealth of detail that
such an analysis offers. Furthermore, such an analysis is able to incor-
porate more of the emotional impact of the different discourses. The
emphasis on story-lines, while couched in terms of narrative rather than
logic, still privileges an account of connections and makes little reference
to the language in which the narrative is delivered. Much of the impact
of language is not just to be found in the plot of the story being told but
in how that story is told. It is here that rhetorical analysis in particular
demonstrates its strengths.
Starting with the Aboriginal discourse, the key entities here are
Table 6.3 Applying Dryzek’s rhetorical analysis
Basic Entities Assumptions re Agency and Metaphors

Relationships Motivation and Rhetorical
Devices
Aborigines Aborigines Aborigines in Europeans in Very flat as
with moral tune with nature; pursuit of profit; largely spoken
responsibilities; Europeans Aborigines seek for, that is,
‘nature’; destructive of to preserve land mediated voice
Europeans nature
Government Responsible, Humans can Ethos of Fire as
scientists knowledgeable work with scientific ‘management
and realistic nature within a rationality or tool’; described
scientists, management/ scientifically as ‘patch
ecosystems scientific informed burning’, a
framework; managers ‘traditional’
scientific practice;
justification ‘rejuvenation’
of traditional through burning;
practices having an
‘evolutionary’
role; ‘holocaust’
references
Pastoralists Pastures as Emphasis on Profit can ‘Stability’,
‘botanical win-win promote ‘integrity’;
communities’; scenarios of environmental religious and
responsible and sustainable protection emotive imagery
irresponsible development
individuals
Wattle Hills Humans as Nature knows Non- Farmland
part of nature best interventionist; as ‘priceless
self-sufficiency resource’;
as route to emotional tone;
sustainability; use of Haiku
nature also has poetry; some
agency scientific rhetoric
Independent Humans and Scientific Scientific Scientific
scientists nature as knowledge rationality terminology
separate reveals legitimately plus emotional
relationships dominates; language
scientists as
responsible;
humanity could
destroy nature
constructed as Aborigines with an innate set of responsibilities and

rights in relation to the land, nature as a discrete entity, and Europeans,
constructed as bearing guilt and blame. While Aborigines are viewed as
being in tune with the ‘natural rhythms of life’ and intrinsically seeking
to preserve the land, Europeans are constructed as destructive towards
nature and operating in pursuit of economic profit. These constructions
support the story-line gleaned from a Hajerian analysis of the text. The
Dryzekian analysis adds little in this case largely because there is little
use of metaphors or other rhetorical devices. This in turn is because the
Aboriginal voices are largely reported or represented. Indeed, both the
Aboriginal representations at the seminar were made by white academics.
There is little active voice in this discourse and the resulting tone is rather
‘flat’. The only discursive colour can be found in the references to caring
for the land by referring to burning as ‘cleaning up’ land that would
otherwise be ‘neglected’ (p. 23). Love for and care of the land is equated
with burning it; a connection that is also made by the government scien-
tists (p. 34).
By contrast, the analysis of the government scientists’ discourse from
this perspective is very revealing. The main constructed entities are
scientists – constructed as responsible, knowledgeable but also realistic
– and ecosystems, which are a scientific category. This combination of
constructions gives considerable authority to the government scientists.
They constitute the main means of accessing knowledge about ecosystems.
‘Experience of over 50 years of research’ (p. 37) is referenced to support
government pro-burning policy. It is stated by one representative that the
government position on the use of fire is ‘so clearly established factually’
that anyone disputing it should ‘go and read the literature’ (p. 35).
But the government scientists also present themselves in a range of
moral terms: they are responsible: they recognize the limitations of sci-
entific knowledge in terms of uncertainty – ‘we are never going to have
perfect knowledge’ (pp. 12 and 14). They also refer to the importance
of local knowledge and of cultural heritage, therefore enabling the link
between their discourse and that of the Aborigines. This is taken further
in how the key relationships are constructed. Scientific knowledge is seen
as justifying the kind of traditional practices undertaken by Aborigines.
Intertextual reference is made to carbon dating evidence that suggests a
40 000-year history of the use of fire in Australia to justify an accusation
of ‘supreme arrogance’ on behalf of those opposed to burning as they are
‘denying the ancient order’ (p. 34). Science can work with tradition within
the context of an overall assumption of the possibility of positive human
engagement with nature that enables management. Furthermore, such
human management is inevitable and has always happened: ‘There is no
such thing as natural management’ (p. 13). This places modern scientific
management on a par with traditional Aboriginal management. The ethos
of scientific rationality that is relied on here, therefore, does not undermine
the potential for an alliance with the Aboriginal discourse.
As well as adding such detail to the workings of a discourse coalition
between the government scientists and the Aborigines, the rhetorical
analysis provides an insight into the emotional appeal of such a discourse.
The government scientists’ discourse is a rhetorically rich discourse. In
particular, there is very active rhetorical engagement with the key term of
‘fire’. As they say, colloquially, ‘Fire ain’t fire’ (p. 13). Instead the discourse
describes fire in terms of a ‘management tool’, a ‘traditional practice’, a
form of ‘rejuvenation’ (p. 12), and as having an ‘evolutionary’ role (p. 15).
Fire is not a negative thing when discussed in such language. It makes it
possible to combine positive reference to ‘fire’ in a discourse that also talks
of having ‘love’ for the land (p. 34).
And yet the negative effects of using fire are acknowledged. There is
reference to individual birds and animals killed by fire. This is, however,
set against a synecdochical account of how burning can save the habitat
of two key species – the malleefowl and ground parrots (pp. 12–13). The
extent of the intended burning is also firmly set in context by invoking the
extremely emotive term ‘the holocaust’. What the government scientists
are doing is making ‘choices’ (p. 14) or ‘playing God’ (p. 16) in order to
avoid ‘policies of inaction’ that would result in a ‘holocaust of extinction’
(p. 38). The fire of government scientists prevents the fire of complete
annihilation.
The third party to the dominant discourse coalition, the pastoralists, has
a similar discourse in the sense of mixing scientific and emotive rhetoric.
The pastoralists’ discourse again constructs the key entities in terms of, on
the one hand, nature described in scientific terms – botanical communities
with scientific names for species – and, on the other hand, humans as indi-
viduals who are able to act responsibly (although this is acknowledged as
an individual choice, not as inevitable). The relationship assumed between
nature and humans is seen in terms of the kind of win-win scenarios that
are typical of the broader sustainability discourse. The value that this dis-
course adds over the scientific/traditional management discourses of the
government scientists and the Aborigines is its suggestion that economic
profit can also be harnessed to the goal of conservation. As a result, the
combination of the three discourses is rhetorically very strong indeed.
Similar to the government scientists’ discourse, there is also a rich use
of emotive rhetoric within the pastoralists’ discourse, combining scientific
terminology with moral imagery. Pastures are portrayed as biotic commu-
nities with ‘stability’ and ‘integrity’, both moral and eco-scientific terms.
There is also use of quasi-religious rhetoric with reference to fire. A more

primal type of religiosity is invoked:
we are always going to have trouble with fire, as long as some of us feel a thrill,
a quickening of the pulse, as we light up the edge of a road, or feel a grim satis-
faction as we watch the flames leap up the hillside, because since mankind first
learned to use it, everyone loves a fire. (p. 42)
In these terms, the pastoralists also see fire as a strong cultural reference
point to support the management claims for their burning practices:
‘mankind has held fire in both a revered and feared position . . . Fire has
been given God qualities and worshipped’ (p. 29). The symbolism of fire is
related to scientific claims for fire as rejuvenating habitats with ‘old trees
sacrificed for new seedlings’ (p. 12).
Conversion, another religious theme, features too. The representa-
tive from the Cape York Peninsula Development Association (CYPDA)
employs a persuasive rhetorical technique, emphasized by Dryzek, in
stating his original affinity with the anti-burning discourse, which changed
over time as he realized that his views were factually misguided (pp. 39–42).
Fire is portrayed as an ‘emotional issue clouded by folklore’, which con-
trasts to the rational scientific reality of its being merely ‘a fast form of oxi-
dation’. The ‘community disharmony’ that exists over the use of fire is put
down to a simple ‘lack of understanding’ on behalf of those opposed to
burning. This establishes further discursive affinities with the government
scientists’ argument.
Turning to the Wattle Hills residents, this is the only discourse to con-
struct humans and nature as part of the same entity; effectively this places
humans as part of nature. This undermines the notion of human agency.
Instead, the appropriate role becomes one of non-intervention and self-
sufficiency. By contrast, nature is credited with substantial agency as well
as superior knowledge: ‘nature knows best’. This detailed construction of
the residents’ discourse puts them discursively at odds with the previous
three discourses; it therefore reinforces the structure of the discourse coali-
tions discussed above. In addition, the Wattle Hills discourse is the most
emotive of the five presented in the workshop. It uses Haiku poetry and
highly charged language. It refers to Australia as a ‘fire-shaped continent’
(p. 10) and anthropomorphizes the larger animals, referring also to the
‘tender growth’ where birds nest (p. 9). Reference is made to ‘QPWS arson-
ists’ alongside ‘pyromaniacs’ and ‘vandalism’ (pp. 9 and 17). The QPWS
policy, which implies ‘nature can no longer take its course’, is dismissed as
being ‘spawned’ by some ‘dark philosophy’ (p. 9). A strong line is taken in
establishing the destructiveness of fire through the use of emotive phrases
such as ‘bushfires kill and destroy’, they bring ‘death and destruction’
(p. 8) resulting in a ‘smoking, blackened landscape’ (p. 10). The impact of

‘The Almighty Match’ (CAFNEC, 1992) on wildlife, especially birds, pro-
vides the main focus for their arguments with reference made at one point
to one of the representatives having observed ‘young birds’ being ‘burnt
alive’ and ‘totally cooked’ (p. 44).
The following closing statement is typical of the non-conformist nature
of the representation: ‘Let us aim for perfection in all things but please,
now, stop diminishing our reservoirs of nature and spirit.’ The failure to
engage with the conventions of non-emotional presentation is explicitly
recognized by the representative himself who makes ‘no apology for intro-
ducing some emotion to the debate’. Such non-conformity, however, may
well be interpreted as making it difficult to gain credibility for the argu-
ment by positioning it outside the institutional conventions of the other
representatives. The Wattle Hills discourse does use a limited amount
of scientific rhetoric (p. 9) but has no rhetorical means of combining the
scientific and the emotive and this leaves the discourse as predominantly
emotional, a position that is bound to reduce its standing in any policy
debate.
It does, however, play a role in constituting the identity of Wattle
Hills. The residents of Wattle Hills pursue a very different environmental
management policy from most other landholders in the Cape and are well
known for their alternative lifestyle centred on self-sufficiency and natural
regeneration. This is reflected in their presentation’s overall departure
from the rational, scientific approach adopted by the rest of the repre-
sentatives at the seminar. Here, the discourse constitutes Wattle Hills as
a distinctive community. This, however, undermines their attempt to take
a central place within the policy debates. There is an attempt at an appeal
to the ‘economic advantage’ to Cape York landholders of a fire-free man-
agement regime, which, they claim, will ‘halt the decline in soil fertility’,
healthy farmland being a ‘priceless resource’ (p. 11). This could be inter-
preted as attempting discursive affinity by engaging with those for whom
economic gain is a priority. This is not, however, going to be a winning
trope in a debate framed around environmental protection and where
scientific rationality plays such a key role.
Finally, there is the discourse of the independent scientists. As might
be expected this has parallels with the government scientists’ discourse
in terms of the pattern of social construction and rhetoric. Humans and
nature are viewed as distinct with scientific knowledge legitimizing certain
practices, through revealing the key relationships affecting natural systems.
Scientists are again seen as responsible and scientific terminology is again
combined with emotional language. The difference here, however, is the
mainly negative ethos and loose use of apocalyptic rhetoric: ‘destroyed’,
‘disaster’ and ‘catastrophe’. There is also the use of tropes that imply
nature would be better off unmanaged. Burning is equated to a violation
(p. 17), suggesting a virgin state would be preferable. And wilderness, pre-
sented as unmanaged land, is compared favourably with managed land-
scapes (p. 20), in particular as wilderness is seen as the source of sublime
romantic encounters: ‘a place where we can stand with our senses steeped
in nature’ (p. 20).
The rhetoric used drives the independent scientists’ argument towards
the conclusion that humans will inherently destroy rather than conserve
nature and that management conflicts with the natural state of the land.
The government scientists, by contrast, managed the ethos to imply a more
positive message. In policy debates this may well carry greater weight than
a purely negative and oppositional discourse. While negative rhetoric
such as this is highly influential within environmentalism and can assist
in building coalitions among environmentalist groups, it is less effective
within governmental policy settings.
Concluding on discourse analysis of a dominant policy perspective

What we hope to have shown in this application of two discourse analysis
approaches to a specific environmental policy case study is the kind of
insights that can be achieved through an attention to language and dis-
course. For example, throughout the analysis, Hajer’s story-line concept
provided a powerful heuristic device enabling the conflicting claims of the
pro- and anti- burning discourses to be clearly illustrated. Two separate
discourse coalitions are discernible throughout, subscribing either to a
story-line of ‘fire is desirable’ or ‘fire is undesirable’. Precisely as Hajer
posits, the members of these coalitions subscribe to the same story-line
but tend to apply different meanings to it. For example, the Aboriginal
representatives subscribe to the ‘fire is desirable’ story-line on the grounds
of cultural tradition, whereas the government scientists subscribe on the
grounds of ecological desirability.
Discursive affinities are easily discernible within representations of
members of the same discourse coalition, such as the promotion of the
fuel reduction story-line by both government scientists and pastoralists.
Obvious attempts to discredit the storylines of opposite discourses are also
observed throughout the representations. From the perspective of Dryzek’s
work, his typology of societal discourses on the environment proves a
rough-and-ready tool for analysing the construction of the discourses.
By describing basic entities, assumptions and motivations within differ-
ent representatives’ story-lines, Dryzek’s approach is useful in enabling
differentiation between those subscribing to the same overall story-line. It
also provides an access point for the emotive use of language, which plays
a central role in the effectiveness of discourses. This is further enriched by

the detailed analysis that rhetoric affords. From the perspective of the two
approaches, the sustained dominance of the pro-burning discourse within
government policy can be explained by a combination of the failure of the
anti-burning discourse to achieve sufficient discursive affinity to effectively
challenge the dominant pro discourse and the rhetorical strength of the
various proponents of the pro discourse.
But it should be recognized that these are just two possible perspectives
on discourse analysis. For example, they ignore the issue of ‘pervasive
power’ accounts of dominant policy perspectives more commonly found
in mainstream Foucauldian approaches that do not make the adjustments
that Hajer considers necessary. This emphasizes the necessity of making
theoretical choices (with methodological implications) in undertaking
any discourse analysis. An emphasis on discourse is not sufficient; the
analyst needs to adopt a specific take on how discourse operates socially
and within policy contexts. Dryzek and Hajer offer two such possibilities.
They are able to expose the ontological and epistemological assump-
tions and constructions that underlie the dominant policy perspective on
anthropogenic burning in Cape York as well as challenging the stance of
objectivity that is often assumed by policy-makers in the face of conflict-
ing discourses. As environmental decision-making inherently takes place
under conditions of radical uncertainty, the significance of attending to
such exposure should not be underestimated.
We end with some reflection on the experience of undertaking discourse
analysis as a methodological approach. First of all, we would endorse
Lees’ point that any discourse analysis must be undertaken with rigour,
just as with any methodological approach (Lees, 2004). In our view this
involves having a clear theoretical framework that ties in closely to the
form of attention to discourse that is adopted. There is a need to specify
the concept and terms that will be used to analyse the discourse and even
to collect the data that constitutes evidence of the discourse. We have
found Hajer and Dryzek’s approaches helpful here in identifying such
‘middle-range’ concepts and terms. A detailed attention to the words and
language of texts has been invaluable in providing a rigorous form of dis-
course analysis that is transparent and justifiable. Collaborative work at
the level of close-reading has proved an essential element of maintaining
rigour while also introducing the creativity needed for achieving insight-
ful discourse analysis. We would see such analysis as inevitably a creative
process. It is also a reflective process, drawing on the everyday skills and
knowledge that any researcher has as a language user but in a way that
builds a link between theoretical understanding and linguistic practice
(Myerson and Rydin, 1996). These links between theory, careful empirical
work and a balance between creativity and reflection on the part of the
researcher are all elements that make for successful and effective discourse
analysis.
Finally, however, a note of caution must be raised in the context of the
pervasive power accounts of the influence of discourse discussed above.
These arguably suggest an implicit responsibility for us as analysts to be
aware of our own ontological and epistemological beliefs and understand-
ings. This implies a need to adopt a critical and fundamentally reflexive
approach where we explicitly consider how our own interests may wit-
tingly, or unwittingly, be reflected in the production of knowledge that
might be inherently perspective-bound (Hastings, 2000; Heller, 2001).
This is of particular importance given the degree of legitimacy afforded
to academic perspectives (Hastings, 2000). The question must always be
asked: would another author analysing the same text reach the same con-
clusions? Discourse analysis carried out in the spirit of such critical and
reflexive enquiry can then fulfil its potential as a heuristically powerful and
potentially emancipatory tool for policy analysis.
Notes
1. For further detailed explorations of this case study see Ockwell and Rydin (2006) and
Ockwell (2008).
2. This euro-anthropocentric interpretation of Aboriginal burning is contested by contem-
porary anthropologists and many Aboriginals – this is outlined further below.
3. See Ockwell and Rydin (2006) for a more in-depth discussion of the dominant pro-
burning policy paradigm in Cape York.
4. See www.firenorth.org.au/nafi2, accessed 19 August 2009.
5. Note: page references in parentheses from hereon refer to the ‘Tropics Under Fire’
seminar transcript under analysis.
References
Bakker, K. (1999), ‘Deconstructing discourses of drought’, Transactions of the Institute of
British Geographers, 24(3), 367–78.
Bevir, M. (1999), ‘Foucault, power, and institutions’, Political Studies, 47(2), 345–59.
Boulding, K.R. (1966), The Economics of the Coming Spaceship Earth, Baltimore, MD: Johns
Hopkins University Press.
Bowman, D.M.J.S. (1998), ‘Tansley Review No. 101 – The impact of Aboriginal landscape
burning on the Australian biota’, New Phytologist, 140, 385–410.
Bowman, D.M.J.S. (2000), Australian Rainforests: Islands of Green in a Land of Fire,
Cambridge, UK and New York: Cambridge University Press.
Burgess, J., C.M. Harrison and P. Filius (1998), ‘Environmental communication and the cul-
tural politics of environmental citizenship’, Environment and Planning A, 30(8), 1445–60.
Buttel, F.H. (1997), ‘The politics of environmental discourse: ecological modernization and
the policy process’, Social Forces, 75(3), 1138–40.
CAFNEC (1992), ‘Tropics under fire: fire management on Cape York Peninsula’, Public
Seminar, Cairns and Far North Environment Centre, Australia.
Chong, D. (1991), Collective Action and the Civil Rights Movement, Chicago: The University
of Chicago Press.
Crowley, G.M., S. Garnett and S. Shephard (2003), ‘Management guidelines for golden-
shouldered parrot conservation’, Brisbane: Queensland Environmental Protection

Agency.
Darcy, M. (1999), ‘The discourse of “community” and the reinvention of social housing
policy in Australia’, Urban Studies, 36(1), 13–26.
DeLuca, K. (1999), Image Politics: The New Rhetoric of Environmental Activism, London:
The Guilford Press.
DNR (2001), 2000/01 Technical Highlight, Queensland: Queensland Department of Natural
Resources, Mines and Energy.
DNR (2004), Vegetation Management Code for Ongoing Clearances. Cape York Peninsula
Bioregion, Queensland: Queensland Department of Natural Resources, Mines and Energy.
Dryzek, J.S. (1990), Discursive Democracy: Politics, Policy and Political Science, New York:
Dryzek, J.S. (1995), ‘Review of “Risk Society – Towards a New Modernity” – Beck, U.’,
Policy Sciences, 28(1), 231–42.
Dryzek, J.S. (1997), The Politics of the Earth. Environmental Discourses, Oxford, UK: Oxford
University Press.
Dryzek, J.S. (2000), Deliberative Democracy and Beyond: Liberals, Critics, Contestations,
Oxford, UK: Oxford University Press.
Edelman, M. (1988), Constructing the Political Spectacle, Chicago: University of Chicago
Press.
Elliott, L. (1999), ‘Review of “The politics of the earth: environmental discourses”’,
Australian Journal of Political Science, 34(1), 130–31.
EPA (2002), Fire Management, Brisbane, Queensland: Queensland Environmental Protection
Agency and Queensland National Parks and Wildlife.
Flannery, T. (1994), The Future Eaters, Sydney: New Holland Publishers.
Flournoy, A.C. (1993), ‘Beyond the spotted owl problem – learning from the old-growth
controversy’, Harvard Environmental Law Review, 17(2), 261–332.
Foucault, M. (1979), Discipline and Punish: the Birth of the Prison, New York: Vintage.
Foucault, M. (1980), Power/Knowledge. Selected Interviews and Other Writings 1972–1977,
Brighton: Harvester.
Foucault, M. (1984), The Care of the Self: The History of Sexuality Volume 3,
Harmondsworth: Penguin.
Foucault, M. (1990), The History of Sexuality Volume 1: An Introduction, London: Penguin.
Gill, A.M., J.R.L. Hoare and N.P. Cheny (1990), ‘Fires and their effects in the wet-dry
tropics of Australia’, in J.G. Goldammer (ed.), Fire in the Tropical Biota, Berlin: Springer-
Verlag, pp. 159–78.
Gill, A.M., J.C.Z. Woinarski and A. York (1999), Australia’s Biodiversity – Responses to
Fire. Plants, Birds and Invertebrates, Department of the Environment and Heritage,
Canberra, ACT.
Grice, T.C. and S.M. Slatter (eds) (1997), Fire in the Management of Northern Australian
Pastoral Lands, St Lucia, Queensland: The Tropical Grassland Society of Australia, St
Lucia.
Habermas, J. (1996), Between Facts and Norms: Contributions to a Discourse Theory of Law
and Democracy, Cambridge, UK: Polity Press.
Hajer, M.A. (1993), ‘Discourse coalitions and the institutionalization of practice’, in
F. Fischer and J. Forester (eds), The Argumentative Turn in Policy Analysis and Planning,
Durham, NC: Duke University Press, pp. 43–71.
Hajer, M.A. (1995), The Politics of Environmental Discourse. Ecological Modernization and
the Policy Process, Oxford, UK: Oxford University Press.
Hajer, M.A. and H. Wagenaar (2003), Deliberative Policy Analysis. Understanding
Governance in the Network Society, Cambridge, UK: Cambridge University Press.
Hastings, A. (1999), ‘Discourse and urban change: introduction to the special issue’, Urban
Studies, 36(1), 7–12.
Hastings, A. (2000), ‘Discourse analysis: what does it offer housing studies?’, Housing Theory
and Society, 17(3), 131–9.
Healey, P. (1999), ‘Sites, jobs and portfolios: economic development discourses in the plan-
ning system’, Urban Studies, 36(1), 27–42.
Heller, M. (2001), ‘Critique and sociolinguistic analysis of discourse’, Critique of
Anthropology, 21(2), 117–41.
Hill, R. (2003), ‘Frameworks to support indigenous managers: the key to fire futures’, in G.
Cary, D. Lindenmayer and S. Dovers (eds), Australia Burning: Fire Ecology, Policy and
Management Issues, Collingwood, Australia: CSIRO Publishing, pp. 175–86.
Hillier, J. (1998), ‘Beyond confused noise: ideas towards communicative procedural justice’,
Journal of Planning Education and Research, 18(1), 14–24.
Jones, R. (1969), ‘Fire-stick farming’, Australian Natural History, 16(7), 224–8.
Keeley, J. and I. Scoones (2000), ‘Knowledge, power and politics: the environmental policy-
making process in Ethiopia’, Journal of Modern African Studies, 38(1), 89–120.
Leach, M. and R. Mearns (1996), The Lie of the Land. Challenging Received Wisdom on the
African Environment, Oxford, UK: James Currey.
Lees, L. (2004), ‘Urban geography: discourse analysis and urban research’, Progress in
Human Geography, 28(1), 101–7.
Majone, G. (1989), Evidence, Argument and Persuasion in the Policy Process, New Haven,
NJ: Yale University Press.
Marlow, D. (2000), A Guide to Fire Management in Queensland (incorporating fire manage-
ment theory and departmental practice), Queensland: Queensland Department of Natural
Resources, Mines and Energy.
Mason, M. (1999), Environmental Democracy, London: Earthscan.
Meadows, D.H., D.L. Meadows, J. Randers and W.W. Behrens (1972), The Limits to Growth
(a report for the Club of Rome’s project on the predicament of mankind), New York:
Universal.
Myerson, G. and Y. Rydin (1996), The Language of Environment: A New Rhetoric, London:
UCL Press.
Nossiff, R. (1998), ‘Discourse, party, and policy: the case of abortion, 1965–1972’, Policy
Studies Journal, 26(2), 244–56.
Ockwell, D. (2008) ‘“Opening up” policy to reflexive appraisal: a role for Q Methodology? A
case study of fire management in Cape York, Australia’, Policy Sciences, 41(4), 263–92.
Ockwell, D. and J.C. Lovett (2005), ‘Fire assisted pastoralism vs. sustainable forestry – the
implications of missing markets for carbon in determining optimal land use in the wet-dry
tropics of Australia’, Journal of Environmental Management, 75(1), 1–9.
Ockwell, D. and Y. Rydin (2006), ‘Conflicting discourses of knowledge: understanding the
policy adoption of pro-burning knowledge claims in Cape York Peninsula, Australia’,
Environmental Politics, 15(3), 379–98.
Ostrom, E. (1990), Governing the Commons: The Evolution of Institutions for Collective
Action, New York: Cambridge University Press.
QPWS (2000), Fire Management System, Brisbane: Queensland Parks and Wildlife Service.
Queensland State Government (1990), Fire and Rescue Service Act 1990.
Queensland State Government (1994), Environmental Protection Act.
Rees, J. (1990), Natural Resources. Allocation, Economics and Policy, London: Routledge.
RFS (2001), Fire in the Landscape and Beyond 2000, Queensland: Rural Fire Service.
Richardson, T. and O.B. Jensen (2000), ‘Discourses of mobility and polycentric development:
a contested view of European spatial planning’, European Planning Studies, 8(4), 503–20.
Rose, D.B. (1996), Nourishing Terrains: Australian Aboriginal Views of Landscape and
Wilderness, Canberra: Australian Heritage Commission.
Russell-Smith, J., P.J. Whitehead, R.J. Williams and M. Flannigan (2003), ‘Fire and savanna
landscapes in northern Australia: regional lessons and global challenges – preface’,
International Journal of Wildland Fire, 12(3/4), v–ix.
Rydin, Y. (1998), ‘The enabling local state and urban development: resources, rhetoric and
planning in East London’, Urban Studies, 35(2), 175–91.
Rydin, Y. (1999), ‘Can we talk ourselves into sustainability? The role of discourse in the
environmental policy process’, Environmental Values, 8(4), 467–84.
Rydin, Y. (2003), Conflict, Consensus and Rationality in Environmental Planning: An

Institutional Discourse Approach, Oxford, UK: Oxford University Press.
Rydin, Y. (2005), ‘Geographical knowledge and policy: the positive contribution of discourse
studies’, Area, 37(1), 73–8.
Scrase, I. and D. Ockwell (2009), ‘Energy issues: framing and policy change’, in Ivan Scrase
and Gordon MacKerron (eds) Energy for the Future, A New Agenda, Basingstoke:
Palgrave Macmillan.
Smith, G. (2003), Deliberative Democracy and the Environment, London: Routledge.
Stocking, G.C. and J.J. Mott (1981), ‘Fire in the tropical forests and woodlands of northern
Australia’, in A.M. Gill, R.H. Groves and I.R. Noble (eds), Fire and the Australian Biota,
Canberra: Australian Academy of Science.
Watt, P. and K. Jacobs (2000), ‘Discourses of social exclusion – an analysis of “Bringing
Britain Together: A National Strategy for Neighbourhood Renewal”’, Housing Theory
and Society, 17(1), 14–26.
Wheeler, W.B. and M.J. McDonald (1986), TVA and the Tellico Dam 1936–1979. A
Bureaucratic Crisis in Post-industrial America, Knoxville: University of Tennessee Press.
7. Theoretical perspectives on international
environmental regime effectiveness: a
case study of the Mediterranean Action
Plan
Sofia Frantzi1
Introduction
Many modern environmental problems are not occasional random events
that suddenly arise, but are rather the result of long-term processes requir-
ing effective management through time instead of instant solutions. Their
causes and effects are complex issues, strongly interlinked with other
aspects of social, political and economic realities. When these problems
are of a transboundary, or global nature, then their management must
be attempted through regional (bilateral or multilateral) or international
agreements.
Traditionally the focus of academic research has been on issues associ-
ated with the challenge of achieving international cooperation, in other
words on regime formation, but recently there has been an increasing
interest in implementation issues, that is, regime effectiveness. This
chapter aims to discuss the concept of effectiveness of international envi-
ronmental agreements as debated within the academic literature. In the
first section the major theoretical perspectives on international relations
are presented as the context for understanding different explanations given
to international cooperation. Different approaches to defining and meas-
uring effectiveness of the agreements are then described in more detail. In
the second section there is specific reference to a particular example of an
environmental agreement. The Mediterranean Action Plan was chosen
for this purpose since it has not been studied extensively and in addition
its effectiveness is ambiguous according to different viewpoints. Finally in
the last section, a new definition of effectiveness is given, drawing insights
from the aforementioned literature, suggesting that for a regime to be
effective it has to use a holistic approach, to have a pragmatic vision and
to be of a dynamic nature. This perspective attempts to provide a new
approach to the future study of international environmental agreements.
198
A case study of the Mediterranean Action Plan 199
International environmental regimes

Modern environmental problems are often so extensive that they do
not respect national boundaries and cannot be managed by one country
acting alone. The need for international cooperation was at the forefront
of concern about the environment in the 1970s, and since the 1972 United
Nations Conference on the Human Environment, international environ-
mental institutions have proliferated and over 60 multilateral environ-
mental treaties have been signed. For example, new treaties have been
established for the protection of stratospheric ozone, the protection of
many regional seas from pollution, the control of European acid rain and
the conservation of biodiversity, amongst many others (Sands, 2003).
Extensive research has been devoted to the ‘high politics’ surrounding
the negotiations of these international agreements. However, little atten-
tion has been paid to the actual effectiveness of implementation after these
treaties come into force. The main question that has puzzled researchers
is: ‘Do regimes matter?’ Generally the sequence of events is that scien-
tists recognize an environmental problem, an international agreement is
negotiated, a regime is established and operates for some time, but does
the regime really make any difference? Some scholars argue that the envi-
ronmental impact of agreements might be negligible. Others answer that
it is the political benefits that are of significance and this diplomatic activ-
ity counterbalances any weakness in tackling the actual environmental
problem. It can be argued that it is the combination or trade-off of bene-
fits, in both environmental and political terms, that is the key to a regime’s
success. However, it can also be argued that regimes make a difference
irrespective of whether this difference is in the environmental or politi-
cal field. Below, the main academic and research viewpoints considering
effectiveness are described in more detail.
International relations and regime theory

The study of international environmental agreements has become an
increasingly important issue in the literature of international relations.
Historically the study of these agreements is based in realism, neorealism
and neoliberal institutionalism, evolving into what is now called regime
theory. In addition, international political economy approaches based on
historical materialism have often been used to study cooperation on envi-
ronmental problems. Some basic observations about these different theo-
retical perspectives are given below in order to demonstrate the background
to explanations used for international environmental cooperation.
Realism The realist approach has descended from traditional texts

such as Thucydides’ History of the Peloponnesian War, Machiavelli’s The
Prince and Hobbes’s Leviathan and has been mainly concerned with state
security (Haas, 1990, p. 35). Emphasizing the political sphere, the realist
approach analyses relationships among states only according to issues
of power and self-interest (Kütting, 2000a, p. 12). It assumes that states
are only guided by national interest and that their purpose must be to
maximize power, a process that ultimately leads to war as states compete
amongst themselves. According to realists the actors (states) act rationally
and prefer those options that best suit their interests, under the assump-
tion that they have full awareness of world events and thus can estimate
both costs and benefits of alternative solutions. Those solutions chosen
concern the acquisition of power (Haas, 1990, p. 35). Hence only when the
effectiveness of an international environmental agreement coincides with
the interests of the states, can the agreement be effective (Kütting 2000a,
p. 12). However, since international discussions about environmental
problems are often concerned with common threats to livelihoods and not
about power, there is a difference in focus between realist thought about
war and power on one hand, and concerns about environmental degrada-
tion on the other. Moreover, Haas (1990, p. 36) notes that there has been
substantial criticism about realism not being an appropriate model for the
analysis of environmental cooperation because of the importance it places
on matters of security, which are generally not salient features of environ-
mental agreements. However, if security could be extended to matters of
public health or security of borders then it could be included as a theme
when studying international environmental agreements.
Neorealism Neorealism is the most recent version of classical realism

in international relations and is also known as structural realism. With
Kenneth Waltz (1979) as its main representative (Keohane, 1986), this
approach describes and studies international relations according to the
system’s structure. Neorealists take methods from game theory and
microeconomics in order to explain how states behave under anarchy, and
how they negotiate among themselves, resulting in hypotheses about their
motives and the results of this negotiation (Haas, 1990, p. 37). However,
realism and neorealism share some basic principles such as the interna-
tional system still operating under anarchy and the states still being the
main actors within it. Neorealism, however, allows for some kind of
cooperation among states so as to reach a shared goal as, for example,
tackling a common environmental problem, since its centre of attention
has shifted from war (Kütting, 2000a, p. 13). This form of cooperation
can be explained in two different ways, first through hegemonic stabil-
ity theory and second through game-theoretic approaches (Haas, 1990;
Paterson, 1996).
Hegemonic stability: hegemonic stability theory suggests that coopera-

tion is most likely to occur when it is imposed by a dominant state or a
‘hegemon’ within a system (Haas, 1990, p. 40). The difference between
the states that just dominate and the hegemons is that the latter already
have their power and leading role legitimately approved by the other
states (Paterson, 1996, p. 94; Kütting, 2000a, p. 13). However, according
to Kütting (2000a, p. 14) this theory can only explain the existence of
cooperation among states but not the quality of that cooperation, because
the latter is out of its remit and therefore doesn’t have the appropriate
methods. For this reason it is not appropriate for studying the effectiveness
of international environmental regimes.
Cooperation under anarchy (rational choice and game theory): the
‘cooperation under anarchy’ tradition is another school within neorealism,
which suggests that even in the absence of a hegemon cooperation is still
possible. As Paterson (1996, p. 101) observes, scholars of this tradition,
influenced largely by game theory, believe that cooperation is indeed pos-
sible under conditions of anarchy without, however, suggesting generally
that this cooperation could change the primarily anarchic character of the
international political order. Rational choice and game theory study and
foresee the behaviour of the actors by calculating the best possible deci-
sion, under rational terms, for any actor under a particular state of affairs
(Kütting, 2000a, p. 14). This school looks at game-theoretic work focusing
primarily on repeated game situations such as the Prisoner’s Dilemma, the
Chicken Game and Stag Hunt. One of the best-known options in empiri-
cal research for measuring regime effectiveness by using rational choice
and game-theoretic approaches is the so-called Oslo-Potsdam solution, for
which further details are given later in this chapter.
A difference with the hegemonic stability school is that cooperation
under anarchy suggests that various factors can cause the maintenance
of the agreements by states after the decline of a hegemonic power that
was initially necessary for the creation of these agreements. Moreover,
the supporters of this school, in contrast to the realists, assume imperfect
information, variable interest and choices of the actors, and only limited
effort at seeking alternative solutions to the problem (Haas, 1990, p. 44).
However, according to some authors (Paterson, 1996; Kütting, 2000a)
rational choice, game-theoretic approaches and neorealist approaches in
general, do not offer a major contribution to the study of the effectiveness
of international environmental agreements for various reasons. First, they
focus on the behaviour of units (states) and do not really include the object
of cooperation (the environmental problem) in their analysis in the sense
of dealing with the environmental degradation per se (Kütting, 2000a,
p. 15). Second, their main assumption is that states can be treated as actors
with given interests on a particular matter, generated by their position in

the international system, whereas on environmental issues the interests
of states can vary according to their internal structure, for example, the
interests of states in the climate change debate (Paterson, 1996, p. 108).
Third, according to Young (2001, 2003), while specifically criticizing the
Oslo-Potsdam solution, these approaches encounter many analytical
and empirical problems that are largely to do with neglecting important
factors when accounting for the hypothetical situation in the absence of
the regime, and for the collective optimal solution.
Historical materialism and international political economy Another

approach often used for assessing international cooperation is an inter-
national political economy approach based largely on historical material-
ism (Paterson, 1996, ch. 8). Historical materialism is mainly concerned
with the distribution of economic resources and international equal-
ity, often expressed as the North–South divide. Historical materialists
explain cooperation in terms of the control of powerful capitalist states
(for example, North American and European countries) over weaker
ones (for example, developing or Third World countries). According to
them the world is broadly divided into three categories on the basis of
the division of labour internationally. These are the highly industrialized
Western countries, the industrializing countries and finally the develop-
ing countries (Haas, 1990, p. 47). Historical materialists identify a much
less democratic and equitable structure of international relations (both
economic and political) than the neorealists, by suggesting that in cases
where effective cooperation does take place it always repeats the principles
of capitalism, that is, reproducing the structures where the North takes
advantage of the South (Haas, 1990, p. 47). Some authors have found the
international political economy approach appropriate for understanding
the complex patterns of cooperation with regard to international environ-
mental agreements. For example, according to Paterson (1996, ch. 8) it
has been useful in assessing the difficult negotiations among countries over
global warming and the UN Framework Convention on Climate Change.
However, economic globalization gives rise to complex relations between
environment on one hand and global trade and investment on the other,
and so raises debates (Stevis and Assetto, 2001; Clapp and Dauvergne,
2005, ch. 5). According to Clapp (2006) there are three different views
within this debate. The first one can see positive effects for the environ-
ment from international growth and even in cases where some negative
side-effects appear, then environmental issues can find ways around them
without restricting economic relations. The second view is primarily
negative, suggesting that international economic growth can only harm
the environment, hence requiring environmental agreements to restrict

international economic relations. Finally, the last view is somewhere in
between, admitting the potential for both advantages and disadvantages,
arguing though that proper management of the global economy can gener-
ate benefits for both sides, environment and growth (Clapp, 2006). In this
sphere of ‘global governance’ some writers suggest that in order for this
link between trade and environment to work beneficially, the creation of a
World Environment Organization (Biermann, 2000, 2006) might balance
the negotiating power of the World Trade Organization. To conclude,
according to the new perspective on the relationship between international
political economy and the environment, the former could potentially offer
some explanation of international environmental cooperation that differs
significantly from historical materialism.
Neoliberal institutionalism Neoliberal institutionalism has dominated the

study of international environmental agreements (Paterson, 2000, p. 12)
and centres on the work of regime theorists such as Keohane, Young, Levy
and others. This theory evolved from the development of traditions as
old as those of Grotius and Kant (Paterson, 1996, p. 115; Kütting, 2000a,
p. 15). In spite of the establishment of the United Nations after the Second
World War, institutionalism faded mainly because it was considered to
have failed in preventing international violence during the inter-war period
(Paterson, 1996, p. 115). However, the strengthening of international
reliance and collaboration and the emergence of regional integration in
the 1950s and 1960s (in particular the European Community) led to its
recurrence in an advanced form and its subsequent significance in the
1990s (Paterson, 1996, p. 115; Kütting, 2000a, p. 15). Neoliberal institu-
tionalism, when studying the effectiveness of international environmental
agreements, is closely interlinked with regime theory. Regime theory and
a different approach within it, of great influence in the past decade, that of
Haas’s ‘epistemic communities’, will be discussed in detail below.
Regime theory: regime theory or neoliberal institutionalism evolved out
of general developments in the international relations sphere and specifi-
cally out of neorealism, thus producing a whole new range of views about
the role and importance of international institutions (Paterson, 1996,
p. 116). According to Krasner (1983, p. 358), who was one of the first and
more important authors on the subject, ‘once regimes are established they
assume a life of their own’. He suggests that there are many ways in which
international institutions affect outcomes by influencing state behaviour.
They can alter the capabilities of actors, including states’, they can alter
states’ interests, they can be a source of power that states can appeal to
and they may alter the calculations of states concerning the maximization
of their self-interest (Krasner, 1983, p. 361). So regime theory could

in many cases be seen as synonymous with institutionalism as already
described since both focus on the effect of the processes held to influence
states’ behaviour, and within which sovereign states are caught (Paterson,
1996, p. 117).
The best-known and cited definition of regimes was given by Krasner
(1983, p. 2) who stated that:
Regimes can be defined as sets of implicit or explicit principles, norms, rules

and decision-making procedures around which actors’ expectations converge
in a given area of international relations. Principles are beliefs of fact, cau-
sation and rectitude. Norms are standards of behaviour defined in terms of
rights and obligations. Rules are specific prescriptions or proscriptions for
action. Decision-making procedures are prevailing practices for making and
implementing collective choice.
This definition is closely related to Young’s and Keohane’s definitions of

institutions. Young (1989, p. 32) defines institutions as ‘social practices
consisting of easily recognised roles coupled with clusters of rules or con-
ventions governing relations between occupants of these roles’. Keohane
(1989, p. 3) gives another definition as ‘persistent and connected sets of
rules (formal and informal) that prescribe behavioural roles, constrain
activity, and shape expectations’. Moreover, Keohane et al. (1993, p. 5)
extend the definition of institutions by adding that ‘they may take the form
of bureaucratic organisations, regimes (rule-structures that do not neces-
sarily have organisations attached), or conventions (informal practices)’.
Later Levy et al. (1994, 1995, p. 274) in their work ‘The study of interna-
tional regimes’ (1995) define international regimes as ‘social institutions
consisting of agreed upon principles, norms, rules, procedures and pro-
grams that govern the interactions of actors in specific issue-areas’.
The above definitions, differing slightly one from another, all allow
for the study of international agreements regarding them as regimes and
explaining their attributes according to them. For the purposes of this
study Krasner’s definition will be the point of reference.
According to Krasner (1983, pp. 6–10) there are three orientations of
regime theory. The realist/structuralist view sees the states as actors in
the international system that want to maximize their power, thus they use
regimes only as means to establish rules expressing their interests. It does
not allow for regimes to have an independent impact on behaviour, so it
views the regime concept as useless. The modified realist/structuralist view
sees regimes as the outcome of negotiations and bargaining, often analysed
by rational choice and game theory, and includes other factors of inter-
national cooperation such as social or technological, hence moving away
from the pure politics of maximization of interest. This view suggests that
regimes may matter but only under fairly restrictive conditions, for instance
when independent decision-making leads to unwanted outcomes. Finally,
the Grotian view lays emphasis on social factors, and even though it sees
the states as still the main actors in the international sphere, it assumes
that these actors are necessarily bound by specific norms and rules. This
last orientation considers regimes as much more persistent and accepts
them as a fundamental part of all patterned human interaction, including
interaction in the international system (Krasner, 1983, pp. 6–10).
Nevertheless, the distinction between the above three orientations does
not really play an important role. Regimes cannot always be irrelevant,
and they cannot always be necessary. So the view that regimes may matter
under certain conditions, is the most appropriate. Their effectiveness is
of great importance, since only effective regimes may make a difference.
More details will be given below on the way that regime theory is applied
to the study of international environmental regimes when discussing how
regime theorists define and measure environmental regime effectiveness.
Epistemic communities: a popular tradition within environmental inter-
national regime theory is that of ‘epistemic communities’ (Haas, 1989).
This theory highlights the role of knowledge-based ‘epistemic communi-
ties’ consisting of specialists responsible for articulating policies and iden-
tifying the national interest. Initially, the term ‘epistemic community’ was
used in literature on the sociology of knowledge. It was later borrowed by
international relations specialists and adapted to describe a specific com-
munity of experts. This community ‘shares a belief in a common set of
cause-and-effect relationships as well as common values to which policies
governing these relationships will be applied’ (Haas, 1989, p. 384). The
community, even though originating from various disciplines, operates
within a common network where there is an exchange of ideas, concerns,
results and solutions, aiming at the same political objectives (Haas,
1990, p. 55). This approach focuses on the groups of people who initiate
cooperation rather than on which states are the leading actors who start
the process. However, supporters of this theory do not suggest it should
replace the older international relations theories, but rather complement
them. For instance, as will be described below, Haas (1990) in his study
about the Mediterranean Action Plan explains the cooperation by refer-
ring to ‘epistemic communities’, but he also offers other explanations from
the perspectives of realism/neorealism and historical materialism.
Defining and measuring regime effectiveness

Within regime theory there have been many efforts from researchers to
rigorously study international environmental regimes and try to identify
not only how these agreements were formed, but also if they were effec-
tive afterwards. There is a growing interest in the effectiveness aspect
of regimes, but it is a matter of debate because quite different defini-
tions are used, resulting in different ways of estimating effectiveness. As
Kütting (2000a, p. 30) observes ‘Within the effectiveness debate in regime
theory. . . on one level effectiveness is seen in terms of institutional work-
ings through good institutional structures. . . on another level effectiveness
is measured on the basis of environmental impact’.
Usually regime theorists look at effectiveness as institutional perform-
ance and not as environmental improvement. Even though some of them
recognize the need to look at the environmental impact, only a few actu-
ally try to measure it. For example, some of the Norwegian regime theo-
rists (Wettestad and Andresen, 1991; Underdal, 1992) have considered the
environmental problem but still focus on the institutional performance of
a regime. Also, Haas et al. (1993, p. 7) ask the question whether the quality
of the environment is better because of the regime but they do not indicate
how such change could be measured and how much of it could be assigned
to the regime itself, rather than to other external factors. Nevertheless,
change itself is not a sufficient measurement of effectiveness (Kütting,
2000b). However, recently there has been an attempt by Kütting (2000a,
2000b) to introduce the concept of environmental effectiveness when
studying environmental regimes by distinguishing the concept of effective-
ness as seen in institutional terms from that of accounting for improved
environmental quality, though still having a regime theory perspective.
Furthermore, the attempts to measure effectiveness have been mainly
qualitative. These qualitative methods vary in whether their view is
descriptive (trying to explain what did happen), predictive (trying to
estimate what will happen), normative (looking at what should ideally
happen) or explanatory (trying to explain the reasons why something hap-
pened) (Mitchell and Bernauer, 2002, p. 2). However, a small but increas-
ing number of researchers have approached the subject quantitatively,
recognizing the need for these methods to complement each other in
order to produce more reliable results. A brief discussion of some of these
methods is provided below.
Qualitative approaches In order to estimate whether international envi-

ronmental institutions are effective, Haas et al. (1993) refer to certain
conditions known as the three Cs. They measure the impact of interna-
tional institutions on three conditions essential for effective action in envi-
ronmental problems: high levels of governmental concern, a hospitable
contractual environment in which agreements can be made and kept, and
sufficient political and administrative capacity in national governments. In
each regime they examine three phases of activity; agenda-setting, interna-

tional policies and national policy responses, which are referring to each
of the three conditions respectively. Thus, a regime is deemed effective if
it increases governmental concern, enhances the contractual environment
and builds national capacity. They ask the question ‘Is the quality of the
environment or resource better because of the institution?’, but due to
lack of available data concerning changes to the state of the biophysical
environment that can be actually assigned to the institution, they decide to
focus on ‘observable political effects rather than directly on environmental
impact’ (Haas et al., 1993, p. 7).
Young (1999) looks at causal connections and behavioural mechanisms.
A regime is considered effective based on the extent it ameliorates the
problem that led to the regime’s creation in the first place. However, he
admits that this approach is practically difficult to analyse since complex
social and natural systems within which regimes operate do not allow for
the observed changes to be assigned to the regime itself. According to the
legal approach, the regime is effective to the extent it is followed by legal
compliance, and in the economic approach if it incorporates the legal
definition and adds a cost-efficiency criterion. In the normative approach,
effectiveness equals achievement of values such as fairness or justice,
stewardship and participation, whereas in the political approach a regime
is effective if it causes changes to the behaviour of actors, in the interests
of actors, or in the policies and performance of institutions in ways that
contribute to positive management of the targeted problem. Moreover,
Young differentiates the effects of environmental regimes in three dimen-
sions. First, he divides them into internal and external to the behavioural
complex, which is the group of actors, interests and interactions on a spe-
cific issue area. Second, he separates them into direct and indirect effects.
Finally, he divides them into good or bad according to the impact on the
problem, in other words if they ameliorate or worsen it (Young, 1999).
Another approach to the measurement of effectiveness focuses on
institutional factors and addresses a series of related questions based on
the identification of problem structure, institutions and institutional fit
and the analysis of legal and organizational issues that arise from this
approach (von Moltke, 2000). This research strategy begins with consider-
ation of a problem’s structure. It then proceeds to identify the institutions
that may be needed – and those that have been employed – to address the
issue in light of its problem structure. Von Moltke’s underlying hypothesis
is that it is more likely for a regime to be effective when it achieves a good
fit between problem structure and institutional characteristics, and that
it is the desirable fit between problem structure and institutions that is a
primary reason for its effectiveness. Moreover, he stresses the importance
of science assessment (the interpretation of the research for policy pur-

poses), and the need for transparency and participation. He goes on to
discuss the issue of dispute settlement mechanisms, without considering
them necessary for environmental regimes since they pursue effectiveness
and implementation in entirely different ways (von Moltke, 2000).
As mentioned earlier, one qualitative approach that is different from the
others in the sense of introducing the concept of environmental effective-
ness, is that of Kütting (2000a, 2000b). She suggests a distinction between
institutional and environmental effectiveness, since most regime theories
are interested in the structure of the institution and the behaviour of the
actors in it, judging its effectiveness by the occurrence of change in this
behaviour, which it is assumed would eventually lead to a positive envi-
ronmental result. However, the change in actors’ behaviour might not
actually result in environmental improvement, and even if it does, this
improvement might not be sufficient to solve the problem. In addition,
the assessment of the state of the environment before and after the regime
and how much of a change can be actually assigned to the regime itself
poses another methodological problem. For this reason, Kütting regards
the distinction between institutional and environmental effectiveness as
necessary, stressing, however, that a good definition should incorporate
both these dimensions since these are ‘two sides of the same coin’ (Kütting,
2000a, pp. 30–34). Her approach looks at four areas of environmental
effectiveness, which describe the relation of an environmental problem to
the particular regime established for its abatement, and the social struc-
tures within which they are found. These four determinants are economic
structures, time, science and regulatory structures.
Economic structures include not only the structures concerned directly
with the agreement but also refer to the economic organization of
the society. Environmental problems can occur through the economic
organization of the society but they can also be avoided through the same
structures. Time is crucial when damage may be irreversible and this is
frequently the case in environmental problems so the time plan of the
environmental regimes has to account for that pressure. Science is neces-
sary in policy-making in order to define the roots and the solutions to
the problems, but according to Kütting its importance should not only
be limited to being an input in the creation of the regime, but it should
also be regarded as a social activity consistent with other social processes,
emphasizing the constant interaction between science and policy. Finally,
regulatory structures are mainly concerned with institutional design and
effectiveness, referring not only to the structure of the agreement but also
to the other bureaucratic structures within which the regime operates, and
they are important because regime design matters (Kütting, 2000a, ch. 4).
Generally, when specific cases are studied in qualitative research, there

is a problem about generalizing the results and assuming they will apply
in all cases. Even though results may be reliable for a particular case,
they cannot always be extended to others. Moreover, no matter how well
a study of effectiveness is designed and carried out, its relative outcome
depends heavily on the initial definition of effectiveness, and the criteria
used to assess it.
Quantitative approaches A discussion of the main quantitative

approaches in the study of environmental regime effectiveness is given
below. Some of them are described briefly, whereas others are given in
more detail due to their complex statistical nature. One of the most well-
known options in empirical research for measuring regime effectiveness is
the so-called Oslo–Potsdam solution. This is an ‘umbrella term’ referring
to two closely interlinked approaches, that of Underdal (1992, 2002) and
that of Helm and Sprinz (2000).
Underdal (1992, 2002, pp. 5–6) focuses on the relationship between
the regime’s output – the institution established as a new set of rules and
regulations; its outcome – the change in the behaviour of states; and its
impact – the actual change in the state of the biophysical environment. He
suggests that regime effectiveness has two components: changes in human
behaviour and changes in the state of the biophysical environment itself.
Moreover, he asks some critical questions. First, what is the object to be
evaluated, because it makes a vast difference whether the evaluation con-
cerns only the regime, or whether it concerns the whole problem-solving
effort that might include various kinds of costs or positive side-effects
associated with the process of its establishment and maintenance. Second,
he discusses the standard against which this should be evaluated, stressing
however, that effectiveness is only a relative term and should be defined in
each regime independently. The issue he raises about standards is impor-
tant since environmental scientists and activists on one hand and regime
theorists on the other, could have diverse opinions about the nature of
standards against which they measure effectiveness. Third, he raises the
issue of methodology in order to measure the object of evaluation against
the standard. Methodologically, Miles and Underdal use counterfactual
analysis against certain behavioural and technical optima by comparing
the actual regime versus no-regime and the regime versus the collective
optimum (Miles et al., 2002, ch. 2). They use qualitative case studies (for
example, the Vienna Convention and Montreal Protocol, the International
Whaling Commission, inter alia) to assess effectiveness on a 0–4 scale for
behavioural change and on a 1–3 scale for environmental improvement.
They then normalize the scales to range from 0 to 1 in order to make
comparisons between them. A weakness of this approach is the difficulty of

estimating the counterfactual by assuming hypothetical conditions in the
absence of the regime. This is largely true since assessing the current state
of the environment is difficult in itself. Even more difficult, if not impos-
sible, is the idea of estimating how the state of the environment would be
today if the regime in question did not exist in the first place. Moreover, the
basis of this technique is still qualitative since environmental improvement
and behavioural change are still assessed through qualitative case studies.
Helm and Sprinz (2000) also use counterfactual analysis based largely
on the questions Underdal posed about the object of evaluation, the
standard against which it should be evaluated and the methodological
approach used. According to them regime effects are improvements in
the object of evaluation, measured by application of policy instruments
leading to changes such as emission reductions. A lower bound is deter-
mined by the no-regime counterfactual (NR), which is the degree of
policy-instrument application that would have occurred in the absence
of the regime. An upper bound is established by the collective optimum
(CO), the degree of application that would have been obtained by a perfect
regime. Accordingly, the regime potential is expressed in units of policy-
instrument use and is the difference between the no-regime counterfactual
and the collective optimum. The actual policies executed by countries (AP)
usually fall into this interval. Thus, the effectiveness of a regime can be
measured as the percentage of the regime potential that has been achieved,
where this score falls into the interval of 0–1 (Figure 7.1).
They estimate scores by using a combination of methods such as game
theory, optimization or experts’ judgements. However, their approach
has been criticized. Young (2001, pp. 110–14) points out that use of the
Nash equilibrium leaves no room for cooperation, since it assumes that all
actors try not to be taken advantage of, and it might also produce results
that are worse for everyone, compared with those that could be achieved
through other potential ways of cooperation. Moreover, he argues that the
interactive decision-making used to calculate the no-regime counterfactual
(NR) (AP) (CO)
Degree of use of instrument

Effectiveness score ES = (AP – NR)/(CO – NR)
where (NR) = no-regime counterfactual, (CO) = collective optimum, (AP) = actual performance
Source: Helm and Sprinz (2000).
Figure 7.1 General concept for measuring regime effectiveness

leaves out many important factors such as political, technological, demo-

graphic and social factors. He has similar concerns about the collective
optimum, pointing out that it neglects important side-effects of regimes
when accounting for regime consequences. Empirically, Young suggests
that the use of the counterfactual poses the same methodological prob-
lems discussed before since the use of expert judgements to estimate it are
insufficient especially when they do not account for social or technological
factors (Young, 2001, pp. 110–14). This critique led to a fruitful debate on
the issue and on potential ways to improve these approaches (Hovi et al.,
2003a, 2003b; Young, 2003).
Another approach to measuring effectiveness is given by Mitchell
(2004) who, in order to evaluate international environmental regimes, uses
regression analysis on panel data. He proposes a quantitative approach
by developing a model for a single regime’s effects. In this model he uses
time-series data for one country at a time for the 1985 Sulphur Protocol
of the European Convention on Long-range Transboundary Air Pollution
(LRTAP). He specifies the following model to estimate national sulphur
emissions for the LRTAP case (Mitchell, 2004, p. 127):
EMISS 5 a 1 b1*MEMBER 1 b2*INCOME 1 b3*POP

1 b4*COAL 1 b5*EFFIC 1 . . . 1 bN*OTHER 1 e
where EMISS is annual emissions of sulphur dioxide and MEMBER is

coded as 0 in years of non-membership of the country to the regime and as
1 in years of membership. Generic drivers of emissions of most pollutants
are also included such as per capita income (INCOME) and population
(POP). Emission-specific drivers are included, such as the country’s coal
power plants (COAL) and their average efficiency (EFFIC). The model
estimates difference in sulphur emissions and how these are explained
by the different variables. For instance b1 represents the expected differ-
ence in emissions that would arise from a country becoming a regime
member, holding all other variables constant. The coefficients of the other
independent variables b2 through bN correspond to the estimated increase
in emissions that would arise from a one-unit increase in that variable.
The t-statistic on the coefficients shows the statistical significance of the
independent variables, whereas the goodness-of-fit (R2) of the model equa-
tion as a whole provides an estimate of how completely the analyst has
modelled the dependent variable.
Mitchell (2004, p. 129) advances his method by developing another
model that allows comparison by combining data from different regimes.
He uses time-series data and data across regimes. As an example he devel-
ops a model to assess the simple claim that sanctions are necessary for a
regime to significantly influence behaviour. An extension of this model

could be used to evaluate how much a regime’s effectiveness depends on
contextual factors. For example, international conferences and reports
might raise the importance of an environmental issue for a few years,
and therefore lead to increased levels of implementation and compliance
(Brown Weiss and Jacobson, 1998).
An advantage of this technique, and also of other quantitative methods,
is that its conclusions can hold reasonably well across many cases even
though they cannot completely explain any specific regime (Mitchell,
2004, p. 122). However, it is important to avoid confusion between the
notions of statistical significance and policy significance of the independ-
ent variables (Mitchell, 2004, p. 128). For instance, a study might show
that an independent variable is statistically significant, which means that it
can definitely explain the variation in the dependent variable. Despite that,
the change in the variation might be so small as to be environmentally
meaningless.
Mitchell’s approach is a promising new angle to assess effectiveness
with the use of econometrics and by using actual scientific measurement of
the environmental problem (for example, emissions). However, it largely
depends on availability of similar data for other regimes. For instance,
when measuring marine pollution, it is almost impossible to keep a long
time-series record of pollutants released into the sea, which is necessary
for this type of analysis. Methodological problems would include which
pollutants to measure, at what locations (since pollution may be a local-
ized phenomenon), and how to connect these releases directly to the
regime’s regulations. Moreover, the high costs of marine monitoring deter
countries from keeping regular data. So this approach may prove innova-
tive and useful in certain cases, but its applicability in others remains in
question.
All the above quantitative techniques have many advantages, as they can
be based on actual measurements and their conclusions can be valid for
many cases. They counterbalance the problem of generalization of results
that qualitative techniques face. However, they might ignore aspects that
are difficult to measure numerically (for example, political benefits) and
might not completely explain particular cases. In that respect, quantitative
analysis should not replace qualitative approaches, but instead a combina-
tion of the two can enable an integrated study of regime effectiveness.
Other issues related to the study of regime effectiveness In addition

to the definition and measurement of regime effectiveness some other
issues related to the study of environmental regimes are worth men-
tioning, notably institutional economics, compliance and verification,
transparency, openness and participation, and environment and security.

These issues can directly or indirectly affect the effectiveness of regimes,
therefore they should be taken into account when studying a particular
regime.
Institutional economics: within the framework of effectiveness of inter-
national environmental regimes, and since they belong to the broader cate-
gory of institutions, an issue that is certainly worth looking at is economic
efficiency or cost-effectiveness. This is the extent to which the production
of the best economic outcome is produced by means of the least-cost
combination of inputs. As North (1990) observes, transaction costs are
the measure of economic efficiency of institutions. He stresses the message
from Coase’s theorem that when it is costly to transact, then institutions
matter (Coase, 1960). North’s theory of institutions combines human
behaviour with the costs of transacting. The key to the costs of transacting
is the costliness of information. This is because transaction costs include
the price of what is being exchanged, and the costs of protecting rights and
policing and enforcing agreements. He also argues that it needs resources
not only to protect property rights and to enforce agreements but also to
define these rights and agreement rules beforehand. Environmental regimes
must perform certain functions such as limiting use, coordinating users
and responding to changing environmental conditions, which include the
transaction costs of coordination, information-gathering, monitoring and
enforcement. It is easily possible to create a regime so costly to implement
that it overcomes the benefits to be gained from its existence. Therefore,
when examining the effectiveness of international environmental regimes,
researchers should also take into account economic efficiency and transac-
tion costs. No matter how effective a regime is in the amelioration of the
problem it was designed for, it could not perform in the long term if it costs
the countries too much.
Compliance and verification: when studying international environ-
mental agreements and their effectiveness, Ausubel and Victor (1992)
introduce the importance of verification of compliance. They suggest that
verifiable international environmental agreements have more chances to
have successful negotiation procedures and thereafter are more likely to be
implemented properly by the participants. They define verification as ‘the
process determining whether or not a party is in compliance’ (p. 4) and note
that it has not been regarded as a significant aspect of most international
environmental issues to date. In order to fulfil this criterion the creation of
large costly new international or national organizational infrastructures is
necessary, which in most cases has not been done, so most of the formal
information under the regimes is collected, if indeed it is, by national
organizations already existing before the regime was established. In many
cases other actors such as NGOs are involved in this process. However,
verification is still mainly dependent on national reports, which might
be unreliable or even false especially when national interests are at stake.
Hence, practically, it could be the case that compliance is not achieved
even if reporting indicates to the contrary. Furthermore, it is crucial to
properly set the standards against which compliance will be measured so
as to be meaningful (Ausubel and Victor, 1992). Only recently have studies
paid attention to the issue of compliance in international climate regimes
(Barrett and Stavins, 2003; Victor, 2003) noting that successful imple-
mentation means high levels of participation and compliance. Barrett
and Stavins (2003), commenting on the Kyoto Protocol, find that it does
not induce significant participation and compliance and propose different
approaches to improve it by offering positive or negative incentives. In the
Montreal Protocol, for instance, a threat of restrictions on trade of CFCs
or products containing CFCs between the countries participating in the
agreement and those not participating proved successful in motivating
more countries to participate. However, it is commonly acknowledged,
especially in the case of the Kyoto Protocol, that compliance alone (even
if fully achieved) cannot always mitigate the environmental problem,
since some of the heavier polluters might not choose to participate in the
agreement at all.
Transparency, openness and participation: one of the issues requir-
ing attention from international environmental regime practitioners and
scholars is transparency and openness. According to Ausubel and Victor
(1992) transparency refers to the clear presentation of the regime’s activi-
ties and information collected, whereas openness means access of actors
to the negotiating process and information, irrespective of whether these
actors come from within the government or not. They also note that suc-
cessful environmental regimes should provide for these conditions, since
in their case studies of arms control regimes the latter proved unsuccess-
ful partly due to concealment and restricted participation (Ausubel and
Victor, 1992). Moreover, von Moltke (2000) also stresses the importance
of transparency and participation in environmental affairs in general,
though remarkably few formal rules have been adopted in international
environmental agreements to address these needs. A first step in this
direction was the adoption in 1998 of the Aarhus Convention on Access
to Information, Public Participation in Decision-making and Access to
Justice in Environmental Matters (von Moltke, 2000).
Domestic politics: another important factor that affects international
environmental cooperation is change in the patterns of domestic politics.
According to Weale (1992, p. 200), domestic public policy can naturally be
affected by actors and procedures in the international sphere. However,
with the internationalization of political life, domestic actors and pro-

cedures may similarly affect and shape foreign policy-making. Active
pressure groups may play a crucial role by shedding light on important
issues and attracting media attention. This extra power can prove very
useful with regard to international environmental agreements since it can
be used to push governments into participating and complying with them.
Moreover, as Carter (2001, p. 239) observes, domestic political pressure
can originate from environmental groups, from the media, public opinion
or political parties (especially the Greens). This pressure can persuade a
government to change its position in the negotiations surrounding an inter-
national environmental regime, often resulting in that country becoming
a ‘lead state’ with a key role in persuading or forcing other states to join
efforts to form the regime. Carter gives as examples the swing of the West
German government in the 1980s from veto to lead state on acid rain as a
response to the Green Party becoming an electoral force and the decision
of the Australian Labor Party to reject the Antarctic Minerals Treaty as
a result of its pro-green position at the 1987 election, which aimed to win
the support of environmentally conscious voters (Carter, 2001, p. 239).
Finally, Haas et al. (1993, p. 17) argue that ‘lead states’ are subject to more
intense domestic political pressure than other countries, something that
led to US leadership on marine oil pollution in the 1970s and on ozone in
the 1980s. This pressure, together with the frequently greater damage to
the country from the environmental problem, and the advanced policies
for that problem, increase governmental concern and capacity, resulting in
promotion of institutional solutions to the problems by the ‘lead states’.
Environment and security: traditionally, in political science, security has
been considered as protection of a sovereign state from other sovereign
states that might threaten it by means of military power (Morgenthau,
1978). Nowadays there is an increasing concern that environmental
problems can threaten security by leading to violent conflict. According
to Swatuk (2006) in the 1990s two different debates arose within the aca-
demic community. One is concerned with the redefinition of security in
order to include environmental concerns, whereas the other focuses on
the ways and extent that environmental issues may threaten security in the
first place. The two sides have failed to reach a consensus. The first tries
to interlink environmental change with the causes of conflict, identifying
ways in which this might happen (Homer-Dixon, 1999). The other group,
by contrast, suggests that whilst the high degree of global interdependence
might result in environmental problems producing complex situations, it is
rather unlikely to lead to violent conflict (Deudney, 1990).
Some argue that many environmental problems may present signifi-
cant threats to human health and welfare, which in turn would affect the
well-being of nations themselves and therefore these problems should be

taken into account when considering issues of national and regional secu-
rity (Kullenberg, 2002). This necessity for combining security matters with
environmental issues is most appropriate in many cases of environmental
problems, and especially so when addressing maritime affairs (Kullenberg,
2002). Moreover, Carter (2001) observes that according to realists the
environment could be considered a security issue in cases where global
commons problems might cause conflicts among countries. Such cases
where military conflict arose straight from disputes over environmental
problems are rare. However, a significant emerging issue is the rising
number of environmental refugees who, while trying to escape from natural
disasters such as drought, famine, degraded land and deforestation, seek a
more secure future by crossing national borders (Carter, 2001, p. 227).
As Paterson (2000, pp. 18–23) puts it, again according to a realist view,
there are two senses in which environmental change can threaten security.
It may lead to interstate war especially over shared renewable resources,
traditionally water, although this is an unlikely possibility. It is likely
to cause internal instability of states, especially when combined with or
caused by population growth. In that case environmental change may lead
to a complete collapse of the social structure by unplanned urbanization,
spreading disease and ecological marginalization of poor people. Haas
(1990, p. 36) also recognizes the threat that environmental degradation
poses to international security. He admits that the realist view of security
has received criticism concerning whether it is appropriate for environ-
mental issues and he finds it ambiguous. If the idea of security includes
public health, security of borders, social and economic stability, then
cooperative solutions would be more easily achieved for environmental
problems. Haas argues that countries might still underestimate the envi-
ronmental issues when matters of national power are involved, describing,
for instance, the political tension that persisted in the negotiations of the
Mediterranean Action Plan, resulting from a Greek–Turkish diplomatic
incident at sea. No matter which side of the debate one takes, environmen-
tal degradation does seem to be associated with the security of nations,
even if only in the sense of internal stability and social integrity, hence
environmental regimes should also be assessed as an aspect of security.
The Mediterranean Action Plan and the Barcelona Convention

Having reviewed the literature concerning effectiveness of international
environmental regimes in general, in this part of the chapter the discus-
sion will focus on a particular environmental regime, the Mediterranean
Action Plan. As mentioned earlier, it was chosen as a case because it has
not been studied extensively (but see Haas, 1989, 1990; Skjaerseth, 1996,
2002; Kütting, 2000a, 2000b), and also its effectiveness is ambiguous

according to different viewpoints.
The Mediterranean Action Plan (MAP) was created in 1975, under the
auspices of the United Nations Environment Programme (UNEP), only
three years after the Stockholm Ministerial Conference set up the latter
programme. MAP was adopted as a Regional Seas Programme under
UNEP’s aegis. The UNEP Regional Seas Programme is a promising
attempt to develop treaties and soft rules and standards at the regional
level taking into consideration the different characteristics – both needs
and capabilities – of the different regions (Sands, 2003). MAP was the first
plan adopted and has worked since then as a model for designing the other
plans.
The Barcelona Convention was signed in 1976 and forms the legal part
of MAP, in force since 1978 and amended in 1995. It includes six pro-
tocols, namely, the Dumping Protocol, the Prevention and Emergency
Protocol, the LBS (Land-Based Sources) Protocol, the SPA (Specially
Protected Areas) and Biodiversity Protocol, the Offshore Protocol and the
Hazardous Wastes Protocol. The Barcelona Convention is complemented
by a research component (MED POL), policy-planning programmes
(Blue Plan and Priority Actions Programme) and financial/institutional
arrangements.
The Mediterranean Action Plan (UNEP/MAP) involves 21 countries
bordering the Mediterranean Sea, as well as the European Union, which
are Contracting Parties to the Barcelona Convention and its protocols.
MAP’s main objectives (UNEP, 1995b: Annex IX) are:
● to ensure sustainable management of natural marine and land

resources and to integrate the environment in social and economic
development, and land use policies;
● to protect the marine environment and coastal zones through pre-
vention of pollution, and by reduction and, as far as possible, elimi-
nation of pollutant inputs, whether chronic or accidental;
● to protect nature, and protect and enhance sites and landscapes of
ecological or cultural value;
● to strengthen solidarity among Mediterranean coastal states in man-
aging their common heritage and resources for the benefit of present
and future generations; and
● to contribute to improvement of the quality of life.
Origins, negotiations and formation of the Mediterranean Action Plan

Haas (1990, ch. 3) gives a detailed overview of the history and negotiations
up to the adoption of MAP, which is summarized below. Early worries
about Mediterranean Sea pollution arose between the late 1960s and 1974
when some Mediterranean officials expressed for the first time a need
for action and governments sought ways to obtain information on the
extent of marine pollution by identifying sources and types of pollutants
and on possible ways to deal with the situation. Since adequate informa-
tion was not yet available the attention focused on oil pollution resulting
from maritime traffic and accidental spills, as this was the most visible
form. Afterwards, however, several scientific meetings and conferences
revealed a variety of pollutants and their sources, with the most impor-
tant being the land-based, so in 1974 a first draft of a treaty was prepared
by the Food and Agriculture Organization. However, later the same
year, Mediterranean governments approached another United Nations
organization, the United Nations Environment Programme (UNEP) to
guide and support this regional effort, which in turn, with the help of 40
Mediterranean marine experts, developed a comprehensive plan. Finally,
in 1975 the Mediterranean Action Plan was adopted including seven
monitoring and research projects, for an entire set of pollutant types and
sources, and several pilot demonstration projects (Haas, 1990, ch. 3).
Thereafter MAP gradually widened its scope through creation of proto-
cols covering land-based sources of pollution, marine dumping, tanker oil
pollution, as well as pollution transported by rivers and in the atmosphere
and by extending the lists to include more pollutants. The environmental
assessment component of MAP also evolved as the research and monitor-
ing projects increased from seven to 12 and some interim standards were
developed (Haas, 1990, ch. 4).
However, following the 1992 UN Conference on Environment and
Development ‘Earth Summit’ in Rio and the requirements of the Rio
Declaration on Environment and Development (Agenda 21), MAP
attempted to translate the results of the summit onto the regional
Mediterranean level, and adapted Agenda 21 to the Mediterranean
context by setting up Agenda MED 21. This led to adoption of the Action
Plan for the Protection of the Marine Environment and Sustainable
Development of the Coastal Areas of the Mediterranean (MAP II) on 10
June 1995 (UNEP, 1995b). MAP II reflected both increasing concern for
the pressures exerted on the Mediterranean environment and commitment
of Mediterranean states to the ideal of sustainable development.
International environmental cooperation and the creation of the

Mediterranean Action Plan
Regional cooperation was necessary to create a treaty aimed at protection
of the Mediterranean against pollution. Environmental cooperation, as
with any other international relations procedure, requires different actors
or states to coordinate decisions and actions with the other actors involved.
Reaching an international political agreement is difficult and there are dif-
ferent explanations about the conditions under which cooperation in the
Mediterranean Basin was achieved through the framework of MAP and
the Barcelona Convention. Haas (1990) summarizes these different inter-
pretations into the main categories described below, explaining the causes
of cooperation, its effects and its forms in each one of the views.
Realism and neorealism are concerned mainly with the relationship
between state power and order in security affairs and the political economy
of advanced industrialized societies. Realists and neorealists would relate
cooperation to the distribution of power between the Mediterranean
states. Under this perspective the regional hegemonic leadership of France
would play a key role in developing cooperation under conditions of
international anarchy. This hegemony would dictate that the scope of
the agreements would mainly cover pollutants of interest to France but
also extend to other issues of national French interest. The strength of
cooperation – how weak or binding it is – would be dependent on French
power and might also depend on information available. Under a realistic
view the duration of the cooperation – how persistent it is – would also
vary with the two previous factors and the effects of the cooperation would
be to strengthen the influence of France in the region and achieve common
benefits for all the Parties. However, this explanation did not prove
adequate when, after the decline of the regional French hegemony, MAP
continued to exist and to receive increased support both from the hegemon
and also from weaker states, showing that it is difficult to predict potential
change in the motives of the states (Haas, 1990, ch. 6).
Historical materialism, as discussed earlier, is basically concerned with
distribution of economic resources and international equality, very often
expressed as the North–South divide. Historical materialists explain
cooperation in terms of the control of powerful capitalist states (that is,
European countries in the case of the Mediterranean region) over weaker
less-developed ones (that is, North African and/or Middle East coun-
tries in the same case). According to them the imperialism of European
states would lead to cooperation under conditions of capitalism. The
scope of the cooperation would not be clear but it would strengthen
areas where European states have interests. Both strength and duration
of the cooperation would vary with European dominance and effects of
cooperation would be imposition of unwanted forms of development
on less-developed countries, excluding alternatives, and the provision
of relatively more benefits to European states, thus increasing com-
mercial dependence of the less-developed countries on them. So, in the
context of MAP, under a historical materialist interpretation, northern
Mediterranean countries would try to impose capitalist policies on the

southern Mediterranean developing countries. However, the negotiations
proved to be a compromise where both sides’ interests were equally rep-
resented, indicating that historical materialism was not able to provide a
satisfactory explanation of cooperation (Haas, 1990, ch. 7).
A third set of explanations introduced by Haas (1989, 1990) involves the
‘epistemic communities’ theory. This theory highlights the role of special-
ist knowledge-based ‘epistemic communities’ in formulating government
policy and altering national interests and finally leading to international
cooperation. The ‘epistemic community’ approach gives a more flexible
character to the cooperation, having a broader scope than the other expla-
nations. According to this approach the acquisition of new information
and the negotiations between the states would lead to cooperation under
conditions of scientific uncertainty. The scope of the cooperation would
be broad and specifically outlined by the ‘epistemic community’ and the
strength and duration of the cooperation would vary with the extent of
the involvement of the ‘epistemic community’ and coalitions within the
states. This cooperation would lead to adoption of convergent pollution
control policies, and would eventually inspire Mediterranean governments
to design and implement new models of comprehensive environmental
policy. Indeed, the countries where scientific experts were strong had
deeper involvement in MAP and became its strongest proponents, and
vice versa. The ‘epistemic communities’ explanation complements the
previous two theories, since it accounts for variability in the preferences of
the states through time, an aspect missing from other explanations (Haas,
1989, 1990, ch. 8). However, even though this theory has been useful in
explaining the negotiations and creation of MAP, it is open to question
whether the current operation of the regime is based on ‘epistemic com-
munities’. Moreover, the generalizability of the theory to explain other
regimes is not yet proven.
Structure of the Mediterranean Action Plan and its components

According to Raftopoulos (1993) regional action plans usually consist of
five components: the assessment component, the management component,
the legal component, the institutional component and the financial com-
ponent. The basic characteristics of each MAP component are described
below.
The legal component of MAP MAP seeks to achieve all its objectives
through its legal component, the Barcelona Convention and related proto-
cols. The Convention for the Protection of the Mediterranean Sea against
Pollution was signed in 1976, and has been in force since 1978. In 1995 it
was replaced by an amended version taking into account recommendations

of the 1992 Rio Conference on Environment and Development and it was
recorded as the Convention for the Protection of the Marine Environment
and the Coastal Region of the Mediterranean, being in force since 2004.
The amended version of the Barcelona Convention introduces new prin-
ciples such as Environmental Impact Assessment (EIA), the polluter pays
principle and the precautionary principle and also suggests time limits
for environmental regulations (UNEP/MAP, 2005a). The 22 Contracting
Parties to the Barcelona Convention are: Albania, Algeria, Bosnia and
Herzegovina, Croatia, Cyprus, Egypt, the European Community, France,
Greece, Israel, Italy, Lebanon, Libya, Malta, Monaco, Morocco, Serbia
and Montenegro, Slovenia, Spain, Syria, Tunisia and Turkey.
As described in Article 1.1 of the Convention (UNEP/MAP, 2005a),
geographically it covers:
the maritime waters of the Mediterranean Sea proper, including its gulfs and
seas, bounded to the west by the meridian passing through Cape Spartel light-
house, at the entrance of the Straits of Gibraltar, and to the east by the south-
ern limits of the Straits of the Dardanelles between Mehmetcik and Kumkale
lighthouses.
As is obvious from the above definition, the internal waters of the

Contracting Parties are excluded in the provisions, as are the Black Sea,
the Sea of Marmara and the Bosphorus, since the ‘demarcation line’ is the
southern limit of the Straits of the Dardanelles. In the following provi-
sions, the Convention may be extended to include coastal areas as defined
by each Contracting Party within its own territory, and also any Protocol
to the Convention may extend geographical coverage to that which the
particular Protocol applies.
In Article 2(a) pollution is defined and described as:
the introduction by man, directly or indirectly, of substances or energy into the

marine environment, including estuaries, which results, or is likely to result,
in such deleterious effects such as harm to living resources and marine life,
hazards to human health, hindrance to marine activities, including fishing and
other legitimate uses of the sea, impairment of quality for use of seawater and
reduction of amenities.
The protocols to the Barcelona Convention, also summarized in Table 7.1,

are the following:
● Dumping Protocol. The full title is ‘Protocol for the Prevention of

Pollution in the Mediterranean Sea by Dumping from Ships and
Aircraft’. It was signed in 1976 and has been in force since 1978.
Table 7.1 MAP Protocols
Protocol Entry into Force Description

Dumping Protocol Adoption: 1976 Aims at prohibiting
Protocol for the Entry into force: 1978 discharge of wastes
Prevention of Pollution Amendments: 1995 and other materials by
in the Mediterranean but oldest version in committing states to
Sea by Dumping from force ban dumping of certain
Ships and Aircraft substances – the ‘black
list’ – and issue permits
for the dumping of less
hazardous substances – the
‘grey list’
Prevention and Adoption: 2002 Focuses on promoting
Emergency Protocol Entry into force: 2004 means of combating
Protocol Concerning Replaced the oldest oil pollution through
Cooperation in version in force multilateral cooperation,
Preventing Pollution since 1976 by committing states
from Ships, and, in to notify each other in
Cases of Emergency, case of an oil spill and to
Combating Pollution of cooperate in the clean-up
the Mediterranean Sea
LBS (Land-Based Adoption: 1980 Focuses on eliminating
Sources) Protocol Entry into force: 1983 persistent toxic substances
Protocol for the Amendments: 1995 by committing states to
Protection of the but oldest version in ban or strictly limit a
Mediterranean Sea force number of compounds
against Pollution from such as organohalogen,
Land-Based Sources organophosphorus and
organotin compounds,
heavy metals and
chlorinated hydrocarbons,
inter alia
SPA (Specially Adoption: 1995 Encourages creation and
Protected Areas) and Entry into force: 1999 development of marine
Biodiversity Protocol Replaced the oldest parks to safeguard
Protocol Concerning version in force representative types
Specially Protected since 1982 of coastal and marine
Areas and Biological ecosystems and their
Diversity in the biodiversity, endangered
Mediterranean habitats and species
and sites of aesthetic or
cultural importance
Table 7.1 (continued)
Protocol Entry into Force Description

Offshore Protocol Adoption: 1994 This Protocol requires
Protocol for the Not yet in force authorization by national
Protection of the authorities for any
Mediterranean Sea offshore activity, which
against Pollution should be granted only
Resulting from after the examination of
Exploration and a study of the activity’s
Exploitation of the potential effects on
Continental Shelf and the environment
the Seabed and its (Environmental Impact
Subsoil Assessment)
Hazardous Wastes Adoption: 1996 The Protocol requires
Protocol Not yet in force Parties to take all
Protocol on the appropriate measures
Prevention of Pollution to eliminate pollution
of the Mediterranean resulting from the
Sea by Transboundary transboundary movement
Movements of and disposal of hazardous
Hazardous Wastes and wastes to the fullest
their Disposal possible extent and to
eliminate such movements
if possible
This Protocol was amended and recorded as the ‘Protocol for the
Prevention and Elimination of Pollution in the Mediterranean Sea
by Dumping from Ships and Aircraft or Incineration at Sea’. It
was signed in 1995 but still awaits entry into force. The Dumping
Protocol commits states to banning dumping of certain substances
– the ‘black list’ – and issue permits for dumping of less hazardous
substances – the ‘grey list’. Factors to be considered when establish-
ing criteria governing issue of permits include characteristics and
composition of wastes or other matter, features of the dumping site
and method of deposit of matter to the site. An exception to the
Protocol’s provisions is the case of force majeure due to stress of
weather or any other cause when human life or the safety of a ship
or aircraft is threatened.
● Prevention and Emergency Protocol. The full title is ‘Protocol
Concerning Cooperation in Preventing Pollution from Ships, and,
in Cases of Emergency, Combating Pollution of the Mediterranean
Sea’. This Protocol was signed in 2002, and has been in force since
2004, replacing the existing ‘Protocol Concerning Cooperation in
Combating Pollution of the Mediterranean Sea by Oil and Other
Harmful Substances in Cases of Emergency’, which was in force from
1976. The Prevention and Emergency Protocol commits states to
notify each other in case of an oil spill and to cooperate in the clean-
up. In the event of an oil spill or other emergencies UNEP and also
any other state likely to be affected must be informed. Moreover, in
the framework of this Protocol, a regional activity centre (REMPEC
– Regional Marine Pollution Emergency Response Centre for the
Mediterranean Sea) has been established in Malta, administered by
the International Maritime Organization and the United Nations
Environment Programme to deal with the implementation of this
Protocol. Cooperation in the clean-up includes salvage or recovery
of packages containing hazardous or noxious substances released or
lost overboard. The Protocol also provides for other actions such as
dissemination of reports and information. The article about assist-
ance allows for it to be asked for and given by the regional activity
centre or by any other signatory state in the form of equipment,
products and facilities, expert advice and the costs of any action
shall be borne by the requesting Party.
● LBS (Land-Based Sources) Protocol. The full title is ‘Protocol for
the Protection of the Mediterranean Sea against Pollution from
Land-Based Sources’. It was signed in 1980 and has been in force
since 1983. This Protocol was amended as the ‘Protocol for the
Protection of the Mediterranean Sea against Pollution from Land-
Based Sources and Activities’. The amendment was signed in 1995
but still awaits entry into force. The LBS Protocol covers some
sectors of activity, including heavy metal industries, agriculture,
energy production and waste treatment, binding the countries to
adopt new industrial, agricultural and waste treatment practices.
It also commits states to ban or strictly limit a number of com-
pounds such as organohalogens, organophosphorus compounds,
organotins, heavy metals, chlorinated hydrocarbons, radioactive
substances and thermal discharges inter alia. The Protocol in itself
does not define specific emission or time limits; however, it provides
that states should progressively adopt such guidelines and meas-
ures. Following this, in 1997 the MED POL programme assisted
countries to design and adopt the ‘Strategic Action Programme
to Address Pollution of the Mediterranean Sea from Land-Based
Activities (SAP)’, which entails more specific emission and time
limits for pollution reduction.
● SPA (Specially Protected Areas) and Biodiversity Protocol. The

full title is ‘Protocol Concerning Specially Protected Areas and
Biological Diversity in the Mediterranean’. This Protocol was signed
in 1995, and came into force in 1999, replacing the existing ‘Protocol
Concerning Mediterranean Specially Protected Areas’, which came
into force in 1982. The SPA Protocol was outside the scope of the pro-
gramme as this was initially anticipated in the Barcelona Convention
and MAP and this is why it is considered different from the other
protocols, which were provided for in the Convention. It encourages
creation and development of marine parks to safeguard representa-
tive types of coastal and marine ecosystems and their biodiversity,
endangered habitats, and habitats critical to the survival of endan-
gered species. The Protocol also provides for protection of sites of
particular importance because of their scientific, aesthetic, cultural
or educational interest. It suggests the regulation of certain activities
such as fishing, hunting and trading animals, and the passage, stop-
ping or anchoring of ships. Moreover, it suggests the establishment
of a ‘List of Specially Protected Areas of Mediterranean Importance’
or ‘SPAMI List’. A regional activity centre has been established in
Tunis (SPA/RAC) to deal with issues of protected areas. However,
the Protocol only encourages development of specially protected
areas and does not oblige the signatory states to take any form of
action, so the question of whether this issue should be treated in the
form of a Protocol remains unanswered.
● Offshore Protocol. The full title is ‘Protocol for the Protection of the
Mediterranean Sea against Pollution Resulting from Exploration
and Exploitation of the Continental Shelf and the Seabed and its
Subsoil’. It was signed in 1994 but still awaits entry into force. This
Protocol requires authorization by national authorities for any
offshore activity, which should be granted only after study of the
activity’s potential environmental effects. It includes lists of harmful
or noxious materials and substances, the disposal of which is either
prohibited or requires a special permit, and provides for monitor-
ing of planned installations for environmental and safety effects. In
addition to this the Protocol provides that each Party shall prescribe
sanctions to be imposed for breach of obligations and that as soon
as possible appropriate rules and procedures for the determination
of liability and compensation for damage resulting from relevant
activities should be formulated and established. Delay in adoption
and ratification of this Protocol is attributed to involvement of off-
shore industries, especially the oil industry, in the decision-making
of the governments.
● Hazardous Wastes Protocol. The full title is ‘Protocol on the

Prevention of Pollution of the Mediterranean Sea by Transboundary
Movements of Hazardous Wastes and their Disposal’. It was signed
in 1996 but still awaits entry into force. The Protocol requires
Parties to take all appropriate measures to eliminate pollution
resulting from the transboundary movement and disposal of haz-
ardous wastes to the fullest possible extent and to eliminate such
movements if possible. Contracting Parties are obliged to generally
prohibit the export and transit of hazardous wastes to developing
countries and the Parties that are non-EU members should prohibit
all imports and transits. Moreover, the countries directly or with
the help of competent authorities should implement programmes
of financial and technical assistance to developing countries for
the implementation of this Protocol. Lists of hazardous wastes
and hazardous characteristics of substances are also described, and
provisions for liability and compensation for damage resulting from
the transboundary movement of hazardous wastes are also included
in the Protocol. The delay in the adoption and ratification of this
Protocol is also considered to be occurring for the same reasons as
for the Offshore Protocol, that is, due to conflicting interests with
the oil industry.
In addition there is a seventh Protocol under preparation concerning

Integrated Coastal Zone Management (ICZM). In most cases the pro-
tocols have been revised and supplemented. Most of the amendments,
including the new Barcelona Convention, are still in the process of
ratification as summarized in Table 7.1.
The Barcelona Convention and protocols raise the issue of dealing
with a legally and institutionally complex scheme, because it concerns an
international environmental order, which develops ‘diachronically rather
than synchronically and contextually rather than in isolation from its
relational foundation’ (Raftopoulos, 1993, p. 42). The legal component
of MAP is divided in two broad categories, the common environmental
norms and rules and the community membership norms and rules. The
former relate to specific environmental provisions, whereas the latter give
standard ‘membership’ powers and duties to each ‘Contracting Party’
(Raftopoulos, 1993).
The institutional component of MAP The institutional component of

MAP, as defined within the framework of the Barcelona Convention, is
structured in such a way as to give authority to two organs: the Meetings
of the Contracting Parties and the Secretariat. The highest authority in
decision-making is given to the Meetings of the Contracting Parties, which

occur every two years, and reflect shared interests of all the Parties. They
also make sure that current legal obligations are met, and oversee forma-
tion of new rules. The second authority is the Secretariat of MAP, based
in Athens, which supports its operation by carrying out all the adminis-
trative tasks that secure its smooth implementation, but which also helps
to integrate stakeholder interests into the legislative goals (Raftopoulos,
1993, p. 73).
Moreover, following launch of the MAP II process and a shift towards a
‘sustainable development’ orientation, the Mediterranean Commission on
Sustainable Development (MCSD) was set up as an advisory body to MAP
in 1996 as a think-tank on policies for promoting sustainable development
in the Mediterranean Basin. Moreover, the operation of MAP is supported
through six Regional Activity Centres (RACs) in six Mediterranean cities,
which help operations in a more decentralized way under supervision
of the Secretariat, each offering expertise in specific fields of action for
facilitating the operation of MAP, as shown in Table 7.2.
The environmental assessment component of MAP The environmental

assessment component of MAP, stated in the official text of UNEP (1978)
as the ‘Co-ordinated Pollution Monitoring and Research Programme in
the Mediterranean’ is widely known as MED POL. It is the most straight-
forward technical aspect of MAP and has played ‘an important cohesive
role for the development of a concrete, scientifically based, regional
approach to the problems of the Mediterranean pollution’ (Raftopoulos,
1993, p. 5). MED POL operates in phases. Its first phase, MED POL –
Phase I, lasted from 1975 until 1980. At that time there was not enough
scientific expertise either in the number of trained scientists or in terms
of facilities established, therefore it was constructed upon pilot projects.
This was considered a necessary condition, bearing in mind that full-scale
regional assessments require identified pollution problems common to all
the participating states (Raftopoulos, 1993). Initially there were seven pilot
projects approved in 1975 followed by several others to support the pro-
gramme. States had designated national research centres to participate in
the pilot projects, and the planning and carrying out of necessary actions
was a collaborative effort of UNEP with several international organiza-
tions (ECE, UNIDO, FAO, WHO, WMO, UNESCO, IAEA and IOC of
UNESCO). According to Raftopoulos (1993, pp. 8–9) MED POL – Phase
I proved largely successful in transferring technology and scientific exper-
tise to many Mediterranean states, especially in less-developed countries
since UNEP at the time followed a policy of allowing most of the resources
to those needing them most.
Table 7.2 Regional Activity Centres (RACs)
Regional Activity Establishment Description

Centre
REMPEC Year: 1976 Aims at preventing and
Regional Marine Place: Manoel Island, combating pollution from oil
Pollution Emergency Malta and other harmful substances
Response Centre for Status: Centre by helping Mediterranean
the Mediterranean Sea under IMO/ coastal states to be prepared
UNEP agreement, for major marine pollution
administrated by incidents and to cooperate in
IMO the clean-up
BP/RAC Year: 1977 Adopts a systemic and
Blue Plan Regional Place: Sophia prospective approach to
Activity Centre Antipolis, France Mediterranean environment
Status: National and development issues using
Centre, with an observation and evaluation
NGO status, with tools, generating indicators
regional function and publishing several studies
accordingly
PAP/RAC Year: 1980 Aims to improve the
Priority Actions Place: Split, Croatia Mediterranean
Programme Regional Status: National environmental situation by
Activity Centre Centre with addressing priority actions.
regional function It is mainly concerned
with integrated coastal
area management to lessen
development problems in built
up coastal areas
SPA/RAC Year: 1994 Focuses on biodiversity
Specially Protected Place: Tunis, Tunisia issues and is involved in the
Areas Regional Status: National protection of Mediterranean
Activity Centre Centre with species, their habitats and
regional function ecosystems by producing, inter
alia, strategies for biodiversity
conservation
CP/RAC Year: 1995 Focuses on promoting
Cleaner Production Place: Barcelona, and disseminating cleaner
Regional Activity Spain production technologies for
Centre Status: Public industrial sector in order to
company put at reduce industrial waste at
the disposal of source of the Mediterranean
MAP industrial sector
Regional Activity Establishment Description

Centre
INFO/RAC Year: 2005 INFO/ Aims to provide information
Information and RAC and communication services
Communication Year: 1993 ERS/ and technical support to MAP,
Regional Activity RAC also by enhancing public
Centre Place: Rome and awareness (Initially ERS/RAC
previously ERS/RAC Palermo, Italy would promote and introduce
Environment Remote Status: Public body remote sensing and GIS for
Sensing Regional put at the disposal environmental monitoring and
Activity Centre of MAP sustainable development)
Phase II of MED POL lasted from 1981 until 1990 and was named the
‘Long-term Pollution Monitoring and Research Programme’. For effective
implementation of its specific objectives it was divided into four distinct
components: monitoring, research and study topics, data quality assur-
ance, and assistance. Overall coordination of Phase II was in the hands of
the Mediterranean Action Plan Coordinating Unit (the Secretariat of the
Barcelona Convention) acting on behalf of UNEP, even though the coun-
tries were fully responsible for monitoring activities as stated in Article 12
of the Barcelona Convention and in Article 8 of the Land-Based Sources
Protocol.
MED POL has recently finished its Phase III, which started in 1996 and
lasted until 2005. Just before the end of Phase II important events at both
international and regional levels took place, which guided MED POL to
change its direction. These events were the adoption of Agenda 21 in Rio
1992 and the Global Plan of Action (GPA) in 1995 in Washington (UNEP,
1995a) to address pollution from land-based sources and activities, and
creation of the Mediterranean Commission for Sustainable Development
(MCSD) together with the amended LBS Protocol at regional level. Hence
there was a slow change from pollution assessment to pollution control,
with MED POL becoming a tool for the countries to properly manage
their marine and coastal areas. MED POL Phase III, adopted in 1995 and
called the ‘Programme for the Assessment and Control of Pollution in
the Mediterranean Region’, was directly concerned with implementation
of the two relevant protocols (Dumping and LBS), since it focused more
on management of pollution control (UNEP, 1999). It included activities
such as pollutant trend monitoring and assessing effects of contaminants
on living organisms as well as inventory of pollution sources and loads
and finally the setting up of a database. Regarding control, compliance
of the countries is monitored by an annual report discussing the country’s

existing action plans, programmes and measures for pollution control and
how well these comply with national, regional or international legislation.
All the above activities have to be described in agreements between each
country and MED POL.
From 2005 until 2013, a new phase of MED POL has come into opera-
tion as put forward in the 13th Ordinary Meeting of the Contracting
Parties to the Barcelona Convention (UNEP/MAP, 2003). However, the
starting points for its objectives and goals are those set out in Phase III,
which was considered adequate for supporting the overall objectives of the
Convention and the protocols. In that respect it will continue to operate
with the same tools (monitoring, compliance monitoring, assessments,
capacity building, and so on). However, taking into account recommen-
dations of the evaluation of Phase III (UNEP/MAP, 2005b), it focuses
more on some aspects of Control and Assessment and Public Participation
and it tries to use the Ecosystem Approach more widely in all its aspects
(UNEP/MAP, 2005c).
The environmental management component of MAP The MAP envi-

ronmental management component is called ‘Integrated Planning of the
Development and Management of the Resources of the Mediterranean
Basin’ (UNEP, 1978) and was the first of the main aspects of MAP to be
implemented. Its aim is also to protect the Mediterranean marine environ-
ment but instead of focusing only on pollution sources, it integrates devel-
opment issues of the region in the sense of environmental management.
From the beginning it was divided into a long-term research and study
programme, the Blue Plan and a more straightforward and immediate
programme aiming at performing specific actions, the Priority Actions
Programme (Raftopoulos, 1993).
To assist implementation of the Blue Plan, a Regional Activity Centre
was established in France, namely the BP/RAC. Initially the Blue Plan
performed 12 investigative thematic studies with the help of experts from
both North and South Mediterranean in each study. Later on a more thor-
ough and complete scientific study was performed in order to examine the
potential for integrating social and economic development in the region
to enhance environmental protection. A synthesis and presentation phase
was also planned in order to guarantee dissemination of the results of the
above studies, nevertheless the Blue Plan was criticized for not being able
to achieve that goal (Raftopoulos, 1993, p. 27). According to Raftopoulos,
it has not succeeded in getting through to the non-expert Mediterranean
community such as stakeholders, or the wider public, mainly due to a poor
communication network.
For the Priority Actions Programme (PAP), another Regional Activity

Centre was established, namely the PAP/RAC. Contrary to the Blue Plan,
it involved particular actions to be taken on issues considered as priorities
at the time. Following the example of MED POL it was designed to be
implemented through demonstration and pilot projects. At that time there
was inadequate scientific awareness on the integration of environment
and development for the purposes of environmental management so this
approach was the only solution (Raftopoulos, 1993).
However, according to Raftopoulos (1993, p. 32), the environmental
management component, consisting of the Blue Plan and the Priorities
Action Programme, even though a rather large and important aspect of
MAP, was clearly not covered in the Barcelona Convention. This means
its ideas and findings were not translated into legal provisions, so to a large
extent integration of environment and development was only in the form
of words and not action.
The financial component of MAP Finally, the financial component of

the Mediterranean Action Plan is mainly covered by the Mediterranean
Trust Fund. This is a fund that all the Contracting Parties to the
Convention contribute to, according to their respective national wealth.
The Contracting Parties may also contribute to the operations of MAP
through in-kind contributions (for example, through participation of their
national institutes in the MED POL programme especially in MED POL
Phase II). Additionally some Contracting Parties may provide extra vol-
untary contributions to the Mediterranean Trust Fund, even on a regular
basis such as, for instance, the European Union. The financial arrange-
ments of MAP are also supported on certain occasions by UNEP through
project funding, and this was the case especially in the first years of MAP’s
operation.
Effectiveness of the Mediterranean Action Plan

Effectiveness of the Mediterranean Action Plan has not been extensively
studied by international relations academics. A few exceptions include
Haas, who brought MAP to the attention of the academic community by
praising it as a success, and some others like Skjaerseth and Kütting who
were more critical. Other types of studies carried out discussed certain
aspects of MAP or tried to assess specific features (for example, legal per-
spectives) of its operation (Boxer, 1978; Raftopoulos, 1993, 1997; Jeftic,
1996; Pavasovic, 1996; Vallega, 1996; Massoud et al., 2003; Raftopoulos
and McConnell, 2004, inter alia).
Haas’s study of ‘epistemic communities’ (Haas, 1989, 1990) did much
to bring the Mediterranean Action Plan to the attention of the academic
community. He suggests that many studies focused on regime negotiations

and their creation but few attempts have been made to investigate their
real and practical significance and their direct impact on the behaviour of
actors (states). He proposes that MAP derives its effectiveness from the
influence of ‘epistemic communities’. He considers it a success because it
‘altered the balance of power within the Mediterranean governments by
empowering a group of experts who then contributed to the development
of convergent state policies in compliance with the regime’ (Haas, 1989,
p. 377). He concludes that MAP may signal the emergence of an entirely
new international political order for the environment and he stresses the
role of ‘epistemic communities’ in promoting stronger national pollu-
tion controls (Haas, 1990). Nevertheless, more than 15 years after Haas’s
study, this enthusiasm is missing from other researchers of MAP.
Skjaerseth (1996, 2002) also studied MAP but he was not convinced
about its success. He notes that the reasons for signing up to the Barcelona
Convention did not always have much to do with environmental concern.
For the less developed countries it was an opportunity to receive training
and equipment for monitoring pollution, since the financial burden, at
least until 1979, was carried by UNEP. Also it was a diplomatic opportu-
nity to establish political/diplomatic ties between countries traditionally
in conflict. Therefore the states probably had mixed motives that were not
necessarily entirely environmental. Moreover, the Barcelona Convention
goals were vague, and even though a main goal of MAP in its second
phase was to produce specific targets with specific deadlines for the Parties
to the Convention, it failed to do so. In addition, the states have not been
very willing to provide adequate reporting on the national implementa-
tion of their commitments. Therefore, due to the lack of clear targets and
the inadequate state reporting, it is difficult to estimate whether there has
been behavioural change among target groups. Skjaerseth also considers
the MAP budget to be very limited compared with the wide scope of its
demands. It is even more difficult to assess the impact of MAP on the state
of the marine environment since there is a lack of reliable and continu-
ous pollution and water quality data. It has to be noted though that the
collection of these scarce data is largely a result of MAP’s establishment.
However, even if there is an improvement in the marine environment it
is rather difficult to attribute it all to the regime, since other factors such
as general socioeconomic and technological change or natural environ-
mental variation have to be taken into account. Moreover, for many
countries, much environmental national legislation was also required by
other organizations such as the European Union. Skjaerseth concludes
that MAP is considered a collaborative political success since it produced
a complete plan for de-polluting the Mediterranean Sea and furthermore
because it increased the general environmental awareness and prepared-

ness through regional cooperation and transfer of knowledge. However,
its impact on behavioural change among target groups is not so clear
(Skjaerseth, 1996, 2002).
Kütting (2000a, ch. 5) is also critical of MAP. Even though admit-
ting that the regime has been successful in starting and maintaining a
cooperation process for a significant period of time in a region tradition-
ally characterized by many political conflicts, she finds that overall it can
not be considered as successful in terms of either institutional or environ-
mental effectiveness. Moreover, she argues that basically MAP faced the
typical North–South divide that underpins so many global environmental
problems, although in this case at a regional scale. She also considers that
MAP has been rather disregarded by the international relations academic
community because traditional international relations research focuses
on matters of national economic interest when examining international
agreements and this was not the case in MAP, as it was formed due to
environmental concern. She even asks the question why MAP ‘exists at
all since there is an apparent lack of motivation?’ (Kütting, 2000a, p. 7).
Overall, Kütting suggests that MAP may have been a political success but
in terms of amelioration of the environmental problem, it has not offered
a lot (Kütting, 2000b).
MAP may have succeeded in completing some activities but it is not
clear how much can really be assigned directly to it. As mentioned earlier,
the most important part of the Mediterranean Action Plan is that which
deals with combating pollution from land-based sources, since these are
the main polluters of the Mediterranean marine environment. More than
20 years after the LBS Protocol’s entry into force, its effectiveness cannot
be clearly estimated. There have been several noteworthy actions, such as
construction of sewage treatment plants in many Mediterranean cities,
nevertheless, it is quite likely that some of these actions would have been
taken anyway.
In conclusion, the focus should be on areas where the Mediterranean
Action Plan has undoubtedly been successful. Even if it has not achieved
an enormous change in the state of the biophysical environment, it has
certainly enhanced cooperation, stability and security in a traditionally
unstable and politically heterogeneous region. Moreover, MAP has pro-
moted environmental awareness and capacity-building especially in the
less developed countries of the southern Mediterranean. In some ways
the political, rather than scientific, success of MAP is ironic as it was the
expert scientific ‘epistemic community’ that first created the international
collaboration responsible for launching the Barcelona Convention. But
the legacy is diplomatic rather than scientific.
A new approach to defining and measuring effectiveness

In the first part of this chapter various theories about international envi-
ronmental regime effectiveness and a range of efforts to define and measure
this effectiveness in applied cases were reviewed and examined. What is
evident is that there is no one way to define and measure such a concept,
especially when dealing with complex interactive systems consisting of
socioeconomic factors, policy and politics and the global environment.
The second part presented an overview of an international regime and
a critical discussion of its effectiveness as a case study. The assessments
of MAP made by different academics and practitioners largely varied
according to which criteria they used in assessing the regime.
The theories of realism and neorealism are primarily concerned with
state security and national interest. They do not include environmental
concerns in their analysis, and assume that states have given interests,
which is not the case in environmental issues. Looking at the formation
of MAP, as Haas noted, an explanation through the hegemonic stabil-
ity strand failed once France declined as the hegemon. MAP continued
to exist and be supported by both the lead and weaker states. Historical
materialism and international political economy, especially with the
dominance of economic globalization, can in some cases explain envi-
ronmental cooperation better, but in the case of MAP these theories also
failed since interests of both sides, developed and less-developed states, are
represented equally in the regime. Therefore, neoliberal institutionalism
and strand regime theory are the most suitable traditions to explain inter-
national environmental cooperation. The distinction between Krasner’s
different orientations is not important because regimes may matter under
certain conditions, meaning that effective regimes do matter.
Concerning the different approaches used when defining and measuring
effectiveness, most of the regime theorists focus on institutional perform-
ance of a regime. Even those that consider the environmental problem do
not clearly define how this aspect can be assessed. A different approach
by Mitchell gives an example of such an assessment, but it leaves out of
the calculation factors that cannot be easily measured by numbers, such
as the political benefits of cooperation. Kütting makes clear the need
for a distinction between institutional and environmental performance,
although looked at from a regime theory perspective. As far as qualitative
and quantitative techniques used in the study of regime effectiveness are
concerned, the former usually explains a case well, since time and effort
are spent in researching that particular case, however generalizability
poses problems. On the other hand, quantitative approaches can be valid
for many situations, but they might miss important case-specific factors.
For instance, in counterfactual analysis it is difficult to estimate the
hypothetical situation of the absence of a regime, and Mitchell’s econo-

metric approach depends heavily on the availability of data, which renders
it difficult to apply. Therefore a new approach, which would take into
account both institutional and environmental parameters using comple-
mentary qualitative and quantitative techniques would be ideal to assess
the effectiveness of regimes.
Looking at the case of the Mediterranean Action Plan, the handful of
important studies on its effectiveness show a varied set of opinions, dem-
onstrating that assessing effectiveness depends primarily on defining the
criteria used for this process. Haas’s prominent study on ‘epistemic com-
munities’ found the regime successful and argued enthusiastically that it
would introduce a new concept in international environmental cooperation.
His theory can provide a satisfactory explanation for the role of scientific
groups in the creation of MAP, but its continued success was mostly politi-
cal. It remains highly questionable whether these scientific groups are the
power behind its implementation, or if these groups are instigating such
processes in other international environmental regimes as well. On the
contrary, the study of Skjaerseth is more critical about the achievements
of MAP. Even though he recognizes its political contributions to coopera-
tion and its overall enhancement of general environmental awareness, he
notes that the desired change in behaviour of the actors is not very evident.
Finally, Kütting, distinguishing between institutional and environmental
effectiveness, concludes that unfortunately MAP was not successful in the
long term in either of the two aspects. According to her criteria, its only real
achievement is the instigation and continuation of a cooperation effort in
a politically very difficult region of the world. Drawing from the previous
three studies, and their different outcomes on the same case, it is essential
to define effectiveness before any attempt to assess it.
Undoubtedly, for a scientist, only improvement of the environment is
the raison d’être of an environmental regime. However, the regime’s insti-
tutional performance is equally important as an indirect way to achieve
this as a means to an end and not as an end per se. Hence, the institutional
and environmental aspects of effectiveness do not need to be separated,
but rather integrated in order to provide a holistic view. The need for an
interdisciplinary approach is the first and foremost rule in that respect. So
far academics that study environmental regime effectiveness have come
mainly from a political science background rather than a scientific one.
On the other hand scientists might not rigorously research international
relations issues. ‘Epistemic communities’ drawing expertise from all dis-
ciplines and using both qualitative and quantitative methods of analysis
might prove useful in order to design and implement these regimes. For
instance, in the Mediterranean Action Plan the first question would be: is
the Mediterranean cleaner than before? Or at least cleaner than it would

be without MAP? Then methodological problems such as how to measure
cleanliness would arise, which could only be superseded by proper design
of long-term environmental assessment and more importantly by a proper
feedback mechanism between science and policy. In the absence of a clear
scientific answer the question might be asked, how well is MAP perform-
ing? Then the political aspects would come into play, combining all the rel-
evant issues, whether the regime enhances international cooperation and
security, creates structures, changes the behaviour of the actors, allows for
multi-stakeholder participation and so on. Such a holistic approach could
be the first rule for effective international environmental agreements.
Furthermore, practice has frequently deviated from theory. High expec-
tations, ambitious plans and disregard of social and economic considera-
tions have sometimes led to the establishment of regimes that are difficult
to implement. A general drawback of international law is its voluntary
nature, as it cannot legally bind any state, apart from those willingly
participating in the regimes. For this reason a regime should provide
incentives to its members for participation, and also for compliance in the
long term, irrespective of whether these incentives would be of a politi-
cal or economic nature. Even the imposition of rules such as sanctions
might deter countries from agreeing, thus achieving even poorer results.
Economic considerations should also be taken into account in terms of
financial resources for all the parties to implement the provisions of the
agreement, but also in terms of fair social policy. It may be that the envi-
ronment is the object of protection, but in no way should this happen at
the expense of human needs. People in developing countries need bread
to eat before saving the earth and the sea, and even in developed ones
governments might not accept strict agreements requiring, for instance,
the closure of polluting industries, for fear of unemployment. Hence the
environmental and time limits of an agreement should be specific but at
the same time realistic. Only regimes with a pragmatic vision have more
chance to succeed in the long term.
Ultimately, so far the discussion has focused on the criteria used when
assessing effectiveness of environmental regimes. Various scholars define
various criteria accordingly. Hence they examine each case by using this
set of criteria and how the regime performs in each one of them at a given
moment in time. Nevertheless, times change and with them whole new
concepts in the environmental and political sphere arise. Some regimes
have a life of more than 30 years such as the Mediterranean Action Plan.
Which leads to the logical question: how can the effectiveness of MAP be
assessed today, since other criteria were used for its design 30 years ago?
Even concepts such as marine pollution had a different meaning before the
introduction of ideas such as habitat degradation or coastal zone manage-

ment. In that respect a regime should always be ready to adapt properly
and quickly to new needs, new definitions and new realities. It should
have such an institutional structure that would allow for right and rapid
amendments, and would eradicate any trace of bureaucracy. It should
influence other international or national policies and politics and be open
to be influenced by them. Effective regimes are the living ones, which can
move through time being older and wiser, not older and weaker. Hence
regime effectiveness could not be assessed by static criteria, the only excep-
tion to this being the criterion of the regime’s dynamic nature.
This new perspective on effectiveness would require a regime to use
a holistic approach based on science, policy and their interaction, have
a pragmatic vision for its ultimate goals and be of a dynamic nature to
evolve through time.
Meeting all the above conditions is hard but perhaps it might prove
successful in the quest for effective international environmental regimes.
Bearing in mind that the above definition presents a very broad approach,
it will be further developed in forthcoming studies and particularly applied
in a specific case study, that of the Mediterranean Action Plan.
Conclusion
Environmental problems, instead of a solution per se, demand an effective
management through time. Since this management especially in the case
of global or transboundary environmental problems is very often in the
hands of international environmental regimes, special attention should be
paid to the design and implementation of these regimes, as well as to their
assessment. The new perspective on international environmental regime
effectiveness might perhaps prove a helpful tool towards this direction.
Nevertheless, further research is needed in order to specify new ways
that would bridge the gap between science and policy, which would
provide realistic solutions reconciling conflicting interests and that would
give life to human-made institutions.
Note
1. The author is very grateful to Neil Carter for his valuable comments on earlier drafts
of this study. Special thanks go to Evangelos Raftopoulos for his inspiring ideas during
several discussions on the research. Thanks also go to Jon Lovett and Gabriela Kütting
for their overall support. The Firos Foundation is gratefully acknowledged for funding
the author’s doctoral studies.
References
Ausubel, J.H. and D.G. Victor (1992), ‘Verification of International Environmental
Agreements’, Annual Review of Energy and Environment, 17(1), 1–43.
Barrett, S. and R. Stavins (2003), ‘Increasing Participation and Compliance in International

Climate Change Agreements’, International Environmental Agreements: Politics, Law and
Economics, 3(4), 349–76.
Biermann, F. (2000), ‘The Case for a World Environment Organization’, Environment, 42(9),
22–31.
Biermann, F. (2006), ‘Global Governance and the Environment’, in M.M. Betsill,
K. Hochstetler and D. Stevis (eds), International Environmental Politics, Basingstoke:
Palgrave.
Boxer, B. (1978), ‘Mediterranean Action Plan: An Interim Evaluation’, Science, 202(4368),
585–590.
Brown Weiss, E. and H.K. Jacobson (1998), Engaging Countries: Strengthening Compliance
with International Environmental Accords, Cambridge, MA: MIT Press.
Carter, N.T. (2001), The Politics of the Environment: Ideas, Activism, Policy, New York and
Clapp, J. (2006), ‘International Political Economy and the Environment’, in M.M. Betsill,
K. Hochstetler and D. Stevis (eds), International Environmental Politics, Basingstoke:
Palgrave.
Clapp, J. and P. Dauvergne (2005), Paths to a Green World: The Political Economy of the
Global Environment, Cambridge, MA: MIT Press.
Coase, R.H. (1960), ‘The Problem of Social Cost’, Journal of Law and Economics, 3(2),
1–44.
Deudney, D. (1990), ‘The Case against Linking Environmental Degradation and National
Security’, Millennium, 19(3), 461–76.
Haas, P.M. (1989), ‘Do Regimes Matter? Epistemic Communities and Mediterranean
Pollution Control’, International Organization, 43(3), 337–403.
Haas, P.M. (1990), Saving the Mediterranean: The Politics of International Environmental
Cooperation, New York: Columbia University Press.
Haas, P.M., R.O. Keohane and M.A. Levy (1993), Institutions for the Earth: Sources for
Effective International Environmental Protection, Cambridge, MA: MIT Press.
Helm, C. and D.F. Sprinz (2000), ‘Measuring the Effectiveness of International
Environmental Regimes’, Journal of Conflict Resolution, 44(5), 630–52.
Homer-Dixon, T. (1999), The Environment, Scarcity and Violence, Princeton: Princeton
University Press.
Hovi, J., D.F. Sprinz and A. Underdal (2003a), ‘The Oslo–Potsdam Solution to Measuring
Regime Effectiveness: Critique, Response, and the Road Ahead’, Global Environmental
Politics, 3(3), 74–96.
Hovi, J., D.F. Sprinz and A. Underdal (2003b), ‘Regime Effectiveness and the Oslo–Potsdam
Solution: A Rejoinder to Oran Young’, Global Environmental Politics, 3(3), 105–7.
Jeftic, L. (1996), ‘Integrated Coastal and Marine Areas Management (ICAM) in the
Mediterranean Action Plan of UNEP’, Ocean & Coastal Management, 30(2–3), 89–113.
Keohane, R.O. (1986), Neorealism and its Critics, New York: Columbia University Press.
Keohane, R.O. (1989), International Institutions and State Power: Essays in International
Relations Theory, Boulder, CO: Westview Press.
Keohane, R.O., P.M. Haas and M.A. Levy (1993), ‘The Effectiveness of International
Environmental Institutions’, in P.M. Haas, R.O. Keohane and M.A. Levy (eds), Institutions
for the Earth: Sources for Effective International Environmental Protection, Cambridge,
MA: MIT Press.
Krasner, S.D. (1983), International Regimes, Ithaca, NY: Cornell University Press.
Kullenberg, G. (2002), ‘Regional Co-development and Security: A Comprehensive
Approach’, Ocean and Coastal Management, 45(11–12), 761–76.
Kütting, G. (2000a), Environment, Society and International Relations: Towards more
Effective International Environmental Agreements, New York: Routledge.
Kütting, G. (2000b), ‘Distinguishing between Institutional and Environmental Effectiveness
in International Environmental Agreements: The Case of the Mediterranean Action Plan’,
The International Journal of Peace Studies, 5(1), 15–33.
Levy, M.A., O.R. Young and M. Zürn (1994), ‘The Study of International Regimes’,
Working Paper No. 94-113, Laxenburg, Austria: International Institute for Applied
Systems Analysis.
Levy, M.A., O.R. Young and M. Zürn (1995), ‘The Study of International Regimes’,
European Journal of International Relations, 1(3), 267–330.
Massoud, M.A., M.D. Scrimshaw and J.N. Lester (2003), ‘Qualitative Assessment of
the Effectiveness of the Mediterranean Action Plan: Wastewater Management in the
Mediterranean Region’, Ocean & Coastal Management, 46(9–10), 875–99.
Miles, E.L., A. Underdal, S. Andresen, J. Wettestad, J.B. Skjaerseth and E.M. Carlin (2002),
Environmental Regime Effectiveness: Confronting Theory with Evidence, Cambridge, MA:
MIT Press.
Mitchell, R.B. (2004), ‘A Quantitative Approach to Evaluating International Environmental
Regimes’, in A. Underdal and O.R. Young (eds), Regime Consequences: Methodological
Challenges and Research Strategies, Dordrecht: Kluwer Academic.
Mitchell, R.B. and T. Bernauer (2002), ‘Qualitative Research Design in International
Environmental Policy’, Working Paper No. 2-2002. Zurich: Swiss Federal Institute of
Technology.
Morgenthau, H. (1978), Politics Among Nations, 5th edition, New York: Knopf.
North, D.C. (1990), Institutions, Institutional Change and Economic Performance, New York
and Cambridge, UK: Cambridge University Press.
Paterson, M. (1996), Global Warming and Global Politics, London: Routledge.
Paterson, M. (2000), Understanding Global Environmental Politics: Domination, Accumulation,
Resistance, London: Macmillan.
Pavasovic, A. (1996), ‘The Mediterranean Action Plan Phase II and the Revised Barcelona
Convention: New Prospective for Integrated Coastal Management in the Mediterranean
Region’, Ocean & Coastal Management, 31(2–3), 133–82.
Raftopoulos, E. (1993), The Barcelona Convention and Protocols. The Mediterranean Action
Plan Regime, London: Simmonds and Hill.
Raftopoulos, E. (1997), Studies on the Implementation of the Barcelona Convention: the
Development of an International Trust Regime, Athens: Ant. N. Sakkoulas.
Raftopoulos, E. and M.L. McConnell (2004), Contributions to International Environmental
Negotiation in the Mediterranean Context, Athens and Brussels: Ant. N. Sakkoulas,
Bruylant.
Sands, P. (2003), Principles of International Environmental Law, 2nd edition, Cambridge,
UK: Cambridge University Press.
Skjaerseth, J.B. (1996), ‘The 20th Anniversary of the Mediterranean Action Plan: Reason to
Celebrate?’, in H.O. Bergesen and G. Parmann (eds), Green Globe Yearbook 1996, Oxford,
UK: Oxford University Press.
Skjaerseth, J.B. (2002), ‘The Effectiveness of the Mediterranean Action Plan’, in E.L.
Miles, A. Underdal, S. Andresen, J. Wettestad, J.B. Skjaerseth and E.M. Carlin (eds),
Environmental Regime Effectiveness: Confronting Theory with Evidence, Cambridge, MA:
MIT Press.
Stevis, D. and V.J. Assetto (2001), The International Political Economy of the Environment:
Critical Perspectives, Boulder, CO: Lynne Rienner.
Swatuk, L.A. (2006), ‘Environmental Security’, in M.M. Betsill, K. Hochstetler and D. Stevis
(eds), International Environmental Politics, Basingstoke: Palgrave.
Underdal, A. (1992), ‘The Concept of Regime Effectiveness’, Cooperation and Conflict, 27(3),
227–40.
Underdal, A. (2002), ‘One Question, Two Answers’, in E.L. Miles, A. Underdal, S. Andresen,
J. Wettestad, J.B. Skjaerseth and E.M. Carlin (eds), Environmental Regime Effectiveness:
Confronting Theory with Evidence, Cambridge, MA: MIT Press.
UNEP (1978), Mediterranean Action Plan and the Final Act of the Conference of
Plenipotentiaries of the Coastal States of the Mediterranean Region for the Protection of the
Mediterranean Sea, New York: United Nations.
UNEP (1995a), Report of the Ninth Ordinary Meeting of the Contracting Parties to the
Convention for the Protection of the Mediterranean Sea against Pollution and its Protocols,
UNEP(OCA)/MED IG.5/16, Athens: UNEP.
UNEP (1995b), Global Programme of Action for the Protection of the Marine Environment
from Land-Based Activities, UNEP(OCA)/LBA/IG.2/7, Washington, DC: UNEP.
UNEP (1999), MED POL – Phase III, Programme for the Assessment and Control of
Pollution in the Mediterranean Region, MAP Technical Reports Series No. 120, Athens:
UNEP.
UNEP/MAP (2003), Report of the Thirteenth Ordinary Meeting of the Contracting Parties
to the Convention for the Protection of the Mediterranean Sea against Pollution and its
Protocols, UNEP(DEC)/MED IG.15/11, Athens: UNEP/MAP.
UNEP/MAP (2005a), Convention for the Protection of the Marine Environment and the
Coastal Region of the Mediterranean and its Protocols, Athens: UNEP/MAP.
UNEP/MAP (2005b), Evaluation of MED POL Phase III Programme (1996–2005),
UNEP(DEC)/MED WG.270/Inf.10, Athens: UNEP/MAP.
UNEP/MAP (2005c), Recommendations for 2006–2007, UNEP(DEC)/MED IG.16/5,
Athens: UNEP/MAP.
Vallega, A. (1996), ‘Geographical Coverage and Effectiveness of the UNEP Convention on
the Mediterranean’, Ocean & Coastal Management, 31(2–3), 199–218.
Victor, D.G. (2003), ‘International Agreements and the Struggle to Tame Carbon’, in J.M.
Griffin (ed.), Global Climate Change: The Science, Economics and Politics, Cheltenham,
UK and Northampton, MA, USA: Edward Elgar.
von Moltke, K. (2000), ‘Research on the Effectiveness of International Environmental
Agreements: Lessons for Policy Makers?’, in Final Conference within the EU-financed
Concerted Action Programme on the Effectiveness of International Environmental
Agreements and EU Legislation held in Barcelona 9–11 November 2000, Lysaker, Norway:
The Fridtjof Nansen Institute.
Waltz, K.N. (1979), Theory of International Politics, Reading, MA: Addison-Wesley.
Weale, A. (1992), The New Politics of Pollution, Manchester: Manchester University Press.
Wettestad, J. and S. Andresen (1991), The Effectiveness of International Resource
Cooperation: Some Preliminary Findings, Lysaker, Norway: The Fridtjof Nasnen
Institute.
Young, O.R. (1989), International Cooperation: Building Regimes for Natural Resources and
the Environment, Ithaca, NY: Cornell University Press.
Young, O.R. (1999), The Effectiveness of International Environmental Regimes: Causal
Connections and Behavioral Mechanisms, Cambridge, MA: MIT Press.
Young, O.R. (2001), ‘Inferences and Indices: Evaluating the Effectiveness of International
Environmental Regimes’, Global Environmental Politics, 1(1), 99–121.
Young, O.R. (2003), ‘Determining Regime Effectiveness: A Commentary on the Oslo–
Potsdam Solution’, Global Environmental Politics, 3(3), 97–104.
8. The price of fish and the value of seagrass
beds: socioeconomic aspects of the
seagrass fishery on Quirimba Island,
Mozambique1
Fiona R. Gell
Introduction
Marine resource use is an important component of the local economy
in many tropical coastal areas. The importance of marine resources to
a community depends on the geographic and economic situation of the
area: the level of development, the role of tourism and the availability of
alternative sources of income (Ruddle, 1996a). In areas near international
airports and large cities, marine resources can be exported to an interna-
tional market or sold to tourists and can fetch high prices. In more isolated
places without a developed transport infrastructure, marine resources may
only be used on a subsistence level in the immediate local area and thus
have a much lower economic value (White et al., 1994; Lindén and Lundin,
1996; Birkeland, 1997). However, in such isolated places the local value of
marine resources is high because they are often the most important source
of income and animal protein.
To manage marine resources it is important to assess the ecological
status of the habitats and organisms that are being exploited. It is also
important to assess how local people use marine resources and their role
in the local economy. As the socioeconomic structures of coastal com-
munities develop and change, the intensity with which marine resources
are exploited also changes. In the past this generally happened on a local
or perhaps national level. However, at the beginning of the twenty-first
century, the phenomenon of ‘globalization’ means that few places really
are remote or inaccessible any longer. Economic growth and social change
in one place can affect the extent to which marine resources are exploited
in other parts of the same country, in other countries and even in other
continents. For instance, rapid economic growth in Asia has had a direct
impact on the sustainable use of marine resources throughout the Pacific,
from Taiwan to the Galapagos, with large Asian companies buying live
reef fish, sea cucumbers and other marine products to supply the growing
Far Eastern markets (Birkeland, 1997).
241
Socioeconomic investigations are increasingly being incorporated in

biological studies of tropical coastal fisheries (Sri Lanka – Dayaratne
et al., 1995; Fiji – Jennings and Polunin, 1996; Kenya – Juma, 1998),
as it has become apparent that ecological and socioeconomic aspects of
resource use must be considered together. Increasing emphasis is being
put on the role of user participation and the use of local knowledge in
the management and development of natural resource use in developing
countries (Chambers, 1997; Sillitoe, 1998). However, including an element
of ‘indigenous knowledge’ in natural resource projects is rare, with just 1.1
per cent of all projects funded by the UK Department for International
Development (DFID) in the past including such research (Sillitoe, 1998)
Fishing and other types of marine resource use are typically last resort
sources of food and employment for people who have no alternative
(McManus, 1993). In Mozambique half the population’s protein intake
is estimated to be from fish (Van der Elst et al., 2005). This is one of the
reasons that marine resources and particularly those that are easily acces-
sible on foot are put under such severe pressure in very poor, often highly
populous areas. Accessible sheltered areas such as reef flats and seagrass
beds often support large numbers of very poor fishers who cannot afford
boats (McManus, 1993). Seagrasses are therefore often more likely to be
intensely exploited than other habitats.
Study area
This study was conducted as part of the Frontier-Moçambique Quirimba
Archipelago Marine Research Programme. Much of the socioeconomic
data presented here were collected on Quirimba Island, one of the two
most densely populated islands in the Archipelago, during 13 months’
fieldwork over a period of two years (1996 and 1997). Quirimba Island
is 6km long by 2km wide. It is situated within a few kilometres of the
Mozambican mainland and is part of Ibo District in the province of Cabo
Delgado. At low spring tides it is possible to walk between the island and
the mainland and also to Ibo Island to the north and Sencar Island to the
south. Quirimba has 3000 inhabitants, most of whom live in Quirimba
village at the northern tip of the island. A few hundred people live outside
the village, the majority in the Kumilamba area in the south of the island.
The seagrass beds of the island have a high diversity of fish with 249
species identified (Gell and Whittington, 2002), which is more than 10 per
cent of the total number of fish species in the Western Indian Ocean.
The history of the Quirimbas

The Quirimba Archipelago has a long and colourful history, but there
are few existing historical records and the islands have been little studied
Socioeconomic aspects of the seagrass fishery on Quirimba Island 243
(Boxer, 1963). This is surprising because they were a key part of the Arab
and Portuguese Indian Ocean empires. Davidson (1961 in Boxer, 1963)
suggests that research on the Quirimbas would enable new insights into
the wide-ranging scope of the Indian Ocean trade. Before the Portuguese
arrived in the sixteenth century, the islands were important as prosperous
Arab trading posts for ivory and slaves. In the seventeenth century they
suffered in the wars between the Portuguese and the Omani Arabs, and
at the end of the seventeenth century the Omanis destroyed most of the
buildings in the stone towns on the islands. In the early nineteenth century
the islands were devastated by Madagascan raiders (Sousa, 1960; Boxer,
1963). Unlike Zanzibar, Mombasa, Kilwa or Mozambique Island, the
Quirimbas did not continue to exist as important Indian Ocean trading
ports or become known for their long histories. The Catholic church in
Quirimba village (now used as the school) dates back to 1894 and there is
also a ruined church, Nossa Senhora do Rozario, that was built around
1580 (Sousa, 1960). Santa Maria beach, where the Frontier-Moçambique
project was based, was said to be where some of the first Portuguese set-
tlers in the Quirimba Archipelago had lived (J. Gessner pers. comm. –
from C.R. Boxer, the historian who visited Quirimba in the 1950s).
Historically the islands seem to have been very productive agriculturally
and able to support residents and visiting traders. They were well placed for
trading goods from the interior such as ivory and gold. Ibo and Quirimba
were also important centres for the Indian Ocean slave trade. Records
from early visitors (mainly Portuguese and British) describe the islands as
productive in terms of agricultural produce, goats and, of course, fish, and
a good place to stop for supplies (Sousa, 1960; Boxer, 1963).
For over 400 years there were Portuguese inhabitants on the islands and
a strong Catholic presence, with parish priests on Ibo and Quirimba. Now
the residents are virtually all of African or mixed African, European and
Arab origin. The coastal people traditionally associated with the coast
of Cabo Delgado are the Mwani who speak a dialect of Kiswahili called
Kimwani, and were traditionally traders and fishers. The coastal Mwani
people are mainly Muslim. They are not as strictly Muslim as many coastal
Tanzanian communities and there is still a strong system of traditional
pre-Islamic beliefs. Witch-doctors or curandeiros are important figures
in the community and are consulted for a huge variety of problems. The
other main group on the island are people of Makua origin, the biggest
ethnic group in Mozambique (West, 1998) living throughout the northern
provinces of the country. A third ethnic group present in small numbers
are the Makonde, from the area around the Tanzania–Mozambique
border. The Makonde are known for their skilled wood carving.
Until 1975 Mozambique was under Portuguese colonial rule, with many
colonial plantations and factories in the rural north. The Portuguese were
strict enforcers of fisheries regulations such as mesh size regulations and in
colonial times there were regular incinerations of illegal nets on Quirimba
(J. Gessner pers. comm.). In 1975 Mozambique gained independence and
the 17-year civil war started. The civil war was fought predominantly in the
ordinary villages of rural Mozambique and the ordinary people of Cabo
Delgado province suffered brutal guerrilla warfare. Tens of thousands of
people became deslocados, refugees within their own country, fleeing vil-
lages that were being destroyed. Many of the deslocados fled to the coast,
and in particular the islands, because the guerrillas would not cross water
and so would not pursue refugees to the islands. This influx of people from
the mainland led to an increase in population on the coast (Massinga and
Hatton, 1997) and on the islands (J. Gessner pers. comm.) and a greater
mix of ethnic groups. Many of the refugees were Makuas and some were
Makondes from the northern Mozambican interior.
In the years following the end of the civil war in 1992, Mozambique
gained a level of stability with the government rebuilding the politi-
cal and economic structure (Macia and Hernroth, 1995). The Human
Development Index of the country as determined by UNDP has risen
over recent years, but Mozambique is still the world’s tenth poorest
country (UNDP, 2006). Development is now happening very quickly in
all sectors of Mozambican life, from tourism to heavy industry. However,
the Quirimbas have been somewhat neglected. They are over 2000km
from the national capital Maputo and difficult to reach and are therefore
rarely visited by politicians and other decision-makers. Infrastructure is
poor and there are few opportunities for the people of the Quirimbas to be
represented or heard at a national level.
Methods
Socioeconomic data were collected on Quirimba using several methods:
● informal interviews with fishers on fishing trips, at landing sites and

in the village;
● formal interviews with boat owners;
● accompanying fishers on fishing trips that employed all major
methods;
● Visiting fishers’ households, talking to families and observing daily
routines;
● informal workshops in the village with fishers.
A number of Quirimba fishers and other residents were key advisers

for this study. They were interviewed at length on numerous occasions,
suggested other people to talk to, took me to their houses and the houses
of family and friends to learn more about life on Quirimba and showed
me their allotments (machambas) and their other daily work. The informa-
tion presented in the results is therefore assembled from a large number of
mainly informal interviews with a wide spectrum of people in the commu-
nity. The key informants are acknowledged at the end of this chapter.
Some biological information about the women’s invertebrate fishery
was collected in the first year of the project and has been presented in
Barnes et al. (1998). Additional social and economic information about
these fisheries was collected through informal discussion with women in
the village, by accompanying them fishing and through two workshops.
Workshops were organized with the local OMM (Organizaçao da Mulher
Moçambicana – Mozambican Women’s Organization) representative in
the village. The workshops were conducted in Portuguese, Makua and
Kimwani. Most women in Quirimba were illiterate and many did not
speak Portuguese so it was important to use visual methods that were
accessible to all participants.
Results and discussion
Quirimba during the study period (1996–97)

Quirimba village has a good system of water pumps from wells through-
out the village. There is a medical clinic staffed by a trained nurse, but no
doctor, a small school, a market selling a very limited range of goods, a few
small shops and a bakery. The houses in the village are fairly large (two to
four rooms), and well built with a mangrove structure filled out with mud
and rock. Most have traditional mecute (woven coconut palm) roofs but
there are some stone or cement houses with corrugated iron roofs. The
houses have large yards where cooking is done and also small toilet and
washing shelters. There is no sewerage system in the village. A few houses
have shelters with a ‘long-drop’-style toilet but many people still use the
beach or mangrove area. There is no electricity supply to the village. A
few residents have oil-powered generators for their homes, used for a few
hours each day. There is a mosque in the village for the predominantly
Muslim community.
In Kumilamba in the south of the island, the residents have macham-
bas, small allotments used for mainly subsistence agriculture. Houses in
Kumilamba are of a less permanent nature than those in the village, and
generally smaller (typically just one room). The people who live in the
machambas have easy access to their agricultural land but they are a 6km
walk from the village where they have to go to get drinking water. Also,
many children who live in the machambas do not go to school because
the nearest school is in Quirimba village. Most households on the island

have a chamba, a piece of land, where they grow sweet potatoes, cassava,
papaya, corn and beans. The land tenure system was informal and slightly
vague at the time of study and appears to be entering a process of mod-
ernization and formalization. Many families have plots that their family
had farmed for years, which they think of as their own but have no legal
documentation. Others have cleared bush and scrub by slash and burn to
claim the newly cleared land as their own.
A large area of the middle section of the island is covered in a coconut plan-
tation owned by farmers of German origin whose family have had a planta-
tion on the island since the beginning of the century. The plantation provides
jobs for about 80 people as labourers, guards and processing workers. The
coconuts are produced for the copra market. The plantation owners rear
cattle to sell in Pemba. There is also a local herd of cattle owned by a number
of people in the village. The cattle graze in the coconut plantation.
Although people on Quirimba are fairly self-sufficient in terms of
everyday requirements such as food, water and building materials, many
people make the trip to Pemba, the provincial capital, to sell produce for
a small cash income, to pay for medical treatment and to buy household
goods such as cooking pots and clothes. Three motor boats sporadically
provide transport for the day-long journey from Quirimba to Pemba. It is
also possible to go to Quissanga on the mainland on the other side of the
Montpuez Bay by dhow then go by road to Pemba, but this often takes
even longer and is more expensive.
Types of marine resource use

The main fishing methods on Quirimba Island are seine netting and
marema trapping (woven bamboo fish traps) in the subtidal seagrass beds,
the collection of invertebrates from the intertidal seagrass beds and a
variety of smaller-scale fisheries and resource collection on the coral reefs
and in the mangroves. The marine resources captured by these methods
can be put into three categories based on their use (see also Table 8.1):
Locally consumed resources The majority of fish caught locally are con-
sumed locally by fishers and their families or by other local people who
trade goods or services for them.
Locally sold resources Some fresh fish and shellfish are sold locally.
There is no fish or seafood market in the village as such – buyers usually
go to the beach where fish or shellfish that have been collected that day are
landed. Dried produce is occasionally sold between Quirimba residents on
a casual basis.
Table 8.1 Marine species exploited on Quirimba Island with method of capture, user group, habitat captured from,
market and use for the species
Scientific Common Kimwani Method of User Group Habitat Market Use

Name Name Capture
Fish
Siganus sutor African Safi Seine, gill net Local men Seagrass and Local Eaten fresh
whitespotted and marema reef flat Mainland/ Dried
rabbitfish trap Pemba
Leptoscarus Seagrass Bonju Marema trap Local men Seagrass Local Eaten fresh
247
vaigiensis parrotfish and seine Mainland/ Dried
Pemba
Lethrinus Variegated Sololo Suri trap and Local men Seagrass Local Eaten fresh
variegatus emperor seine net Mainland/ Dried
Pemba
Gerres oyena Blacktip Sala Seine net and Local men Seagrass and Local Eaten fresh
mojarra gill net reef flat Mainland/ Dried
Pemba
Lethrinus Pink ear Njana Seine net Local men Seagrass and Local Eaten fresh
lentjan emperor coral Mainland/ Dried
Pemba
Scientific Common Kimwani Method of User Group Habitat Market Use

Name Name Capture
Large reef Gill net, Itinerant fishers Coral reefs Mainland Dried
fish e.g. spearfishing
acanthurids,
scarids
Carangidae Jacks Njolwe Fence trap, line Local men, Open water, Local Eaten fresh
fishing itinerants seagrass, reef
Molluscs
Octopus Octopus Mweza Stick, hook Local women Reef flat Local Eaten fresh and
248
vulgaris dried
Barbatia Ark shell Ombay Hand Local women Intertidal sand Mainly Dried
fusca and seagrass mainland/
Pemba
Pinna Pinna shell Macaza Hand Local women Intertidal sand Mainly Dried
muricata and seagrass mainland/
Pemba
Pinctada Oysters Mbari Hand Local women Intertidal Mainly Dried
nigra seagrass mainland/
Pemba
Pleuroploca Tulip shell Kome nlume Hand/collected Men Seagrass Operculum Animal eaten
trapezium on net or trap sold in fresh
fishing trips Tanzania
Chicoreus Murex Makome Hand/collected Men Seagrass Operculum Animal eaten
ramosus on net or trap sold in fresh
fishing trips Tanzania
Terebralia Mangrove Hand Men/children Mangrove/ None Used for
palustris whelk intertidal flat baiting traps
Echinoderms
Holo- Sea cucumber Magajojo Hand Local finfishers Seagrass and Export Dried for food
thuriidae and itinerants intertidal and medicinal
Mask and Local finfishers Seagrass and Export uses
canoe and itinerants shallow reef
SCUBA Itinerant fishers Reefs Export
Crustacea
249
Panulirus Crayfish Hand/spear/ Local fishery/ Reef Pemba Restaurants
spp. hook some
itinerant/
tourist
Scylla serrata Mangrove Hand Local fishers Mangrove Local Eaten fresh
crabs European
Portunus Swimming Hand/bycatch Children/trap Intertidal flats, Local Eaten fresh
pelagicus crab in traps fishers shallow water
Reptiles
Chelonidae Turtle Assa Incidental catch Local and Reef and Local Eaten fresh
in net itinerant seagrass
Source: From personal observation and interviews and workshops with fishers.
Resources sold outside Quirimba Large quantities of dried fish are sold
off the island in two main ways. First, island-based traders buy fish from
fishers, particularly trap fishers, accumulate a large quantity of dried
fish then sell it in the market in Pemba or in markets elsewhere in Cabo
Delgado province. They buy fish, often a lot of Lethrinus variegatus and
Leptoscarus vaigiensis, for 3000MZN (meticais) per kilo wet weight, and
sell it for 15 000–20 000MZN dried (12 000MZN was equal to approxi-
mately US$1 during the study period, so traders paid US$0.25 per kilo
for fish and sold it for US$1.25–1.67). This seems like a large profit but
in fact the dried fish will have lost up to three-quarters of its wet weight
through drying, gutting and cleaning. Net fishing boat owners also buy
fish to sell on the mainland. They have an arrangement with their boat’s
crew whereby they either buy the majority of fish caught at a reduced price
or they pay their crew a certain wage and receive the majority of the fish as
their share of the operation. They dry this fish and accumulate it in storage
until they have enough to make the trip to Pemba to sell it.
The other main route to sale for Quirimban fish is through fishers
from the mainland who come to Quirimba and the other islands in the
Archipelago to fish. A small proportion of these fishers are from villages
on the nearby mainland such as Mahate and as far south as Pemba. Many
of them have been coming to the Quirimbas to fish for many years. Most
are farmers in the wet season and fish the Quirimbas in the dry season.
The dried fish supplements their family’s diet of maize and cassava, and
can also be sold for cash to buy any goods that the family cannot produce
themselves, such as clothes, cooking oil and medicines.
The majority of the visiting fishers are from the next province south,
Nampula, and have only been fishing the Quirimbas in large numbers for
a few years (from interviews with Nampula fishers and J. Gessner pers.
comm.). These fishers are in a similar situation to those described above –
they have agricultural work in the wet season and come to the Quirimbas
in the dry season to catch fish to dry and take back to eat over the wet
season and to sell. Traders also come to Quirimba to buy dried fish,
which they said is ‘better quality’ than the fish they can buy in Nampula.
These traders deal in hundreds of kilos of fish, which they drive down to
Nampula.
Cash commodities for export The only really lucrative commercially

driven fishery on Quirimba is the sea cucumber fishery. In the mid-1990s
a large commercial sea cucumber fishing operation based in Pemba was
active in the whole of the Quirimba Archipelago. The operation was
run by a Chinese export company who processed the sea cucumbers and
exported them to the Far East along with other commercially valuable
marine resources caught locally, such as shark fin. At the end of the study
period this fishery had temporarily ceased. At least one Tanzanian sea
cucumber fishing operation was active in the Archipelago, using SCUBA
gear. Traditionally, local fishers collect sea cucumbers by snorkelling, thus
limiting the depth range over which they are vulnerable. Ordinary net or
trap fishers who collect sea cucumbers while they are fishing can sell them
individually to traders for up to 1000MZN (less than US$0.10) each. Other
incidental catches that can be sold to traders include seahorses for which a
fisher can get 5000MZN (US$0.40) for a 7cm individual and the opercula of
some shells, which can be sold by the kilo for 150 000MZN (US$12.00) or
individually for a few hundred meticais. These traders sell these resources
on in bulk across the border in Tanzania to be sold on to the Far East.
The prices for fish caught in the seagrass and coral reef fisheries of
Quirimba were remarkably consistent. Fresh fish were sold for a set
price of between 4000 and 5000MZN per kilo. Occasionally large, good
quality lethrinids caught in baited traps or by hook and line were sold
for a higher price (reportedly up to 10 000MZN per kilo) because of
the good quality of lethrinid flesh. Otherwise, everything from a kilo of
7–10cm Lethrinus variegatus or other small fish, to a kilo of a large jack
(Carangidae) or snapper (Lutjanidae) caught in a fence trap was sold for
around 4000MZN. 5000MZN per kilo of fish was the standard price for
fish throughout rural coastal Cabo Delgado. The same price was charged
in the rural Mecufi district south of Pemba, where 5000MZN was quoted
as the lowest price of fish in Mozambique (Loureiro, 1998). Set prices are
common for a variety of commodities in northern Mozambique, from
shellfish to sweet potatoes.
The only way to preserve fish on Quirimba is to dry it in the sun. This is
done on drying racks outside houses in the village. Dried fish was sold on
the mainland for between 15 000 and 20 000MZN per kilo. Small fish are
dried closed (fechado) and were sold for 15 000MZN per kilo and larger fish
are dried open (aberto – like a kipper) and sold for the slightly higher price
of 16 000MZN per kilo. Dried fish often have a slightly ‘off’ taste, which
does not prevent their sale or consumption. Occasionally seagrass parrot-
fish (Leptoscarus vaigiensis) die in the marema traps, and are retrieved in a
partially rotted state. These fish are not discarded and are dried as normal
and sold for consumption. The highest prices, some exceeding 25 000MZN
per kilo were given for dried fish sold in villages a long way inland. The
prices of fish and marine invertebrates on Quirimba are shown in Table
8.2, along with a selection of other local prices for comparison. Note the
high prices per kilo of fresh invertebrates (10 000–15 000MZN) compared
with that of fish (3000–5000MZN), and also the high price of imported
goods such as sugar (20 000MZN per kilo) and onions (25 000 per kilo).
Table 8.2 Prices paid on Quirimba Island for fish, agricultural produce and other goods, their source, mode of
production and how they are made available to the community
Item Cost on Cost on Source Mode of Market

Quirimba Quirimba Production/
Meticais US$ Exploitation
Fresh fish/kg 3000 to 5000 0.25–0.40 Local Caught by men Sold direct at landing
site
Dried fish/kg 12 000 to 15 000 1.00–1.25 Local Caught by men Sold from home
Squid/kg 3000 0.25 Local Caught by men Sold direct at landing
site
Octopus/kg fresh 5000 0.42 Local Caught by women Sold direct at beach
252
dry 16 000 1.30 Local and men
Pen shells fresh/kg 15 000 1.25 Local Collected by women Sold direct at beach
Ark shells fresh/kg 10 000 0.83 Local Collected by women Sold direct at beach
Oysters/kg 12 000 1.00 Local Collected by women Sold direct at beach
Mangrove crab/kg 5000 0.42 Local Collected by men
Sweet potato/kg 2000 0.17 Local Grown in Sold in machambas
machambas
Cassava leaves/kg 5000 0.40 Local Grown in Sold in machambas
machambas
Peanuts/kg 4000 0.33 Mainland Sold in village market
Coconut (each) 1000 0.08 Local Machambas and Sold in village market
plantation
Onions/kg 25 000 2.08 Mainland Imported Sold in village market
occasionally/
Quissanga market
Eggs (each) 1000 0.08 Local From chickens kept Sold from home
in village and
machambas
Bread flour/kg 10 000 0.80 Mainland Sold in village shop
Maize flour/kg 6000 0.50 Mainland/local Maize from Sold in village shop
machambas
Rice/kg 10 000 0.80 Mainland Sold in village shop
Papaya (each) 500–2000 0.04–0.16 Local Grown in Sold in machambas
machambas
Mango (each) 100–1000 0.01–0.08 Local Machambas and in Sold in village market
plantation seasonally
Sugar/kg 20 000 1.60 Malawi Imported Sold in village shop
253
Condensed milk 15 000 1.25 Maputo/South Imported Sold in village shop
Africa
Chicken (each) 15 000–25 000 1.25–2.08 Local Reared in village or Sold from home
in machambas
Capulana (sarong – 40 000–80 000 3.33–6.66 Maputo/Tanzania Imported (cheapest Sold in village shop
two pieces) from India)
T-shirt 10 000 0.80 Mainland Imported Sold in village market
occasionally
Trousers 20 000 1.60 Mainland Imported Sold in village market
occasionally
TB medication 40 000 3.33 Mainland Imported Sold at village clinic
(course of
treatment)
Note: Prices are shown in the local currency of Mozambican meticais and US dollars.
The different economic levels of marine resource use in the Quirimbas

Marine resource collection in the Quirimba Archipelago can be divided
into a number of categories depending on the level of investment needed
by the fisher:
Little or no personal investment The main group in this category are the
crew of the seine net fishing boats who need to make very little invest-
ment to start fishing. Some have their own strong shoes to protect against
urchins, others have their own masks but the majority have neither. Young
boys go straight into the seine net fishery and strangers to the island are
also free to enter this fishery.
The women’s invertebrate fishery requires very little personal invest-
ment, needing only a bucket or other container. Women collect inverte-
brates from the intertidal at Santa Maria and Pantopi, and also make trips
on boats out to exposed sand banks at low spring tides. Young children
who collect small species of mollusc from the upper intertidal, and women
who collect these shells for occasional meals can also be included in this
category.
Moderate personal investment Trap fishers need a dugout or plank-

constructed canoe with two outriggers (worth between 100 000 and
800 000MZN, see Table 8.3), 30 to 50 (usually 40) marema traps, weights
(stones collected locally), ‘hippo fat’ to waterproof the boats, a pole for
punting and a bailer. The total cost for trap fishing equipment ranged
from a minimum of 220 000MZN (US$18) for a second-hand dugout
canoe and 30 traps, to over 1 000 000MZN (US$83) for a large plank-
Table 8.3 The average costs of the major items of gear for seine net and
marema trap fishing
Item Cost in Meticais Cost in US$ Source

Seine net – new 5–10 million 417–833 Imported
Seine net – 3 million 250 Imported – bought in
second-hand Pemba
Fishing boat 2–3 million 166–250 Made on Quirimba
3–4m with wood from the
mainland
Marema trap 4000 (need 40) 0.33 Bamboo panels from
(13.20 for 40) the mainland
Canoe – dugout 100 000–400 000 8–33 Made on Quirimba
Canoe – timber 400 000–800 000 33–67 Made on Quirimba
constructed canoe and 40 traps. The average initial expenditure was

probably around 300 000MZN for a standard canoe and 160 000MZN
for 40 traps, a total of 460 000MZN (US$38). Maremas also have to be
replaced every few months. The average salary for a day’s formal work in
Quirimba, for example on the coconut plantation or as an ordinary fisher,
was 10 000MZN (US$.83 – see Table 8.4). Most of this small salary was
needed for food and other necessities and saving was not an option for
most ordinary people on Quirimba.
Weeks could pass when a fisher could not go fishing because of the
weather, because of problems with the net or the boat, or because of illness
so any money saved would be used then. Unforeseen expenses included
money for medicines for the fisher himself or a member of the family, for
example, treatment for an elderly relative with TB might cost 40 000MZN
(see Table 8.2). Quinine and aspirin for malaria also had to be bought
fairly frequently. Most people grew their own staple foods such as maize,
sweet potatoes and cassava in their machambas, so for most people food
was rarely a problem.
Groups of women from the district of Mecufi, south of Pemba, were
reported to come to Quirimba every year and fish in the shallows of the
reef flat in front of the village using large capulanas (the patterned cotton
wrap worn by local women and with dozens of other uses). This method
of fishing for very small fish is found throughout the region (Comores –
Dahalani, 1997; Tanzania – Andersson and Ngazi, 1998). Some village
women said that the reason they did not do this type of fishing was that
it required a special large capulana that they couldn’t afford. This type
of fishing therefore requires a level of investment difficult for Quirimban
women who have few opportunities for earning cash.
Gill nets were used by a few groups of people for fishing on the reef
flat on the east coast of Quirimba. These nets are shorter than seine nets
and a lot cheaper. Three people are needed to do gill netting so either all
contributed to the cost of the net or one person bought it. The gill nets on
Quirimba were all owned by ordinary families.
A large initial personal investment The main group in this category were
the seine net fishing boat owners. The same person usually owned the net
and the boat, which together could be an investment of over 6 million
MZN (US$500, see Table 8.3). These boat owners were the entrepreneurs
of Quirimba. They had the largest stone-built houses and were amongst
the only people to own motorbikes and other major consumer items. Some
of the boat owners also had other business concerns apart from trading
fish; some had shops in the village, others traded across the border in
Tanzania.
Table 8.4 The major forms of employment on Quirimba, average time worked, salaries and location of the work
Employment Hours Days Daily Salary Daily Salary Annual Annual Salary Location
Worked Worked per Meticais US$ Salary US$
Month Meticais
Trap fisher 4 hours 22–24 15 000–40 000 1.25–3.33 4–10 million 333–833 Montepuez Bay
Net fisher 5 hours 18–20 10 000 0.83 2 160 000 180 Montepuez Bay
Net fishing captain 5 hours 18–20 14 000 1.17 3 024 000 252 Montepuez Bay
Net fishing boat Various Various 111 000 9.25 23 976 000 1998 Montepuez Bay
owner
Work in machamba Various Various – could Food Machambas
(mainly – seasonal be whole day (Kumilamba)
subsistence)
256
Worker in coconut 8 hours 6 days a week 10 000 0.83 3 138 600 262 Plantation
plantation
Fish processor 1–2 hours 18–20 (same as net 4000 0.33 864 000 72 Quiwandala
fishers)
Intertidal 3 hours 6–8 days around 20 000 + 1.67 1 680 000 140 Santa Maria, Saja, etc
invertebrate spring tides
collector
Making bread Various 18–20 (same as net Fish Quiwandala
and cakes to barter fishers)
for fish
Trader Various Various Various Quirimba, Pemba,
Tanzania
The stories of how fishing boat owners made the money to initially
buy the fishing boats varied. Some were local men who had worked their
way up from being a fisher crewing a large seine net fishing boat, to trap
fishing, to buying their own large boat. A few had family money from land
ownership, others had come from the mainland specifically to fish.
The seine net fishing boats were built in Quirimba village with wood
imported from the continent. A 3m boat cost from 2 to 2.5 million meticais
(around US$200). A 4m boat cost around 3 million meticais. One of the
boat owners interviewed had bought his current fishing boat for 1.5 million
four years previously. The nets were bought on the mainland, usually in
Pemba. A new net complete with floats, weights and ropes reportedly
cost 10 million meticais but all the boat owners interviewed had bought
their nets second-hand for around 3 million meticais. They expected the
nets to last for 10–15 years. Various systems of payment were used on the
fishing boats. Some of the boat owners paid their crew 4000MZN per kg
of fish caught and said their catch could range from a few kilos to a few
hundred kilos. The pay per trip for the crew worked out at 10 000MZN per
fisher and 14 000MZN for the captain, and some small share of the catch,
usually 1kg. The owners also had to pay four or so people to clean the fish.
They were usually paid one kilo of fish or 4000MZN each day. (See Table
8.4 for salaries for the various fishing jobs on Quirimba and Table 8.5 for
the costs and earnings in the seagrass trap and net fisheries.)
Economic gain from fishing

Net fishing The average daily catch per boat was 75kg. If this was sold
at 4000MZN per kilo it would be worth 300 000MZN (US$24). Out of
this the boat owner paid an average of eight crew 10 000MZN each, the
captain 14 000MZN and five people 4000MZN to clean the fish. This
would give a total daily expenditure of 114 000MZN, making a total profit
of 186 000MZN (US$14.90). Boat owners had responsibility for maintain-
ing boats and nets, buying new ropes and other materials, so some money
would go towards this. Ropes in particular needed replacing regularly,
perhaps once a year. Around neap tides some boats catch just 25kg of fish,
worth 100 000MZN, which is less than the daily expenditure calculated
above. Occasionally, when things went wrong and the sail broke or the
boat ran aground, the crew would come back with no fish and presumably
on these days everyone would go home empty-handed.
Other boat owners organized the fishing in a different way. They took
a certain amount of fish every day as their share and the value of the rest
was divided between the crew with the captain getting slightly more. In
this case it was in the interest of the crew to catch more fish to increase
their wage but there did not seem to be an inclination for fishers on boats
Table 8.5 Summary of average values for various aspects of the seagrass
trap and net fisheries
Trap Fishing Net Fishing

Mean daily catch per 7kg 75kg
boat
Total daily catch for fleet 280kg 2250kg
Mean value of daily catch 21 000–35 000MZN 225 000–375 000MZN
US$1.75–2.91 US$18.75–31.25
Initial expenditure Boat 300 000MZN None for individual fishers
US$25
Traps 160 000MZN
US$13
Total 460 000MZN
US$38
Crew’s wages – 8 crew at 10 000MZN per
day, total 80 000MZN,
US$6.67
Captain’s wages – 14 000MZN, US$1.17
Processors’ wages – 5 processors at 4000MZN/
US$0.33 per day, total
20 000MZN/US$1.67
Total daily expenditure 3000MZN/US$0.25 114 000MZN/US$9.5
Mean daily profit for 21 000–35 000MZN 111 000–261 000MZN
boat owner US$1.75–2.91 US$9.25–21.75
Total daily value of catch 630 000–1 400 000MZN 3 375 000–11 250 000MZN
for fleet US$52.5–116.67 US$281.25–947.5
Days fished per year 264 216
Total annual catch 60 tonnes 440 tonnes
Total annual value of 180–300 million meticais 1320–2200 million meticais
catch US$15 000–25 000 US$110 000–183 333
Total annual catch for 1500–2500 million
both methods meticais
US$125 000–208 333
Note: The exchange rate used was US$1 to 12 000MZN.
managed this way to work harder to maximize their earnings. The boats
that seemed to work the most efficiently were those with small crews of
five or so. They had to work harder than larger crews to haul the nets but
they often seemed more efficient at getting their boats out, catching the fish
and returning to port. Many of the boats with large crews of 12 or more
seemed to be primarily a social occasion that just happened to result in
something to eat and some money.
Trap fishing The average daily catch per fisher for trap fishing was 7kg,
an average daily earning of 28 000MZN, nearly three times the average
salary as a labourer or as fishing crew. Trap fishers fished alone and
cleaned their own fish and so had no expenditure on a day-to-day basis. If
fishers could sell their catch fresh in the village they would get 4 000MZN
whereas if they sold it all in one go to a fish trader they would only get
3000MZN per kilo (21 000MZN for an average catch) but they could
organize this in advance and would be guaranteed to sell it all on the day
they caught it. Trap fishers usually retained about 1kg of their catch for
their own consumption. Very few trap fishers dried their own fish.
Socio-cultural aspects of fish use on Quirimba

Fish was highly valued as food on Quirimba. Although some species were
favoured for their taste, particularly lethrinids and siganids, all edible
species fetched a similar price. Fish was often used for barter. Small fish
(around 10cm) caught by net fishers were bartered for bread rolls or peanut
biscuits made by local women, which they brought out to the boats as they
came into the landing site at Quiwandala at the end of the day’s fishing
trip. The women who made these biscuits were usually single or had hus-
bands who were not involved in fishing. There was an inflexible exchange
rate of four small fish (often variegated emperors or three-ribbon wrasse)
for one biscuit. Fish was commonly shared between family and friends in
the complex Quirimban ‘kinship’ networks (discussed below). Fish were
also commonly given as gifts or in exchange for other goods or services.
The small size of the fish eaten was very striking. The capture and con-
sumption of small fishes that would be considered ‘trash fish’ in many
places is a common feature of some tropical artisanal fisheries (Pauly,
1979; Gayanilo and Pauly, 1997). As Munro (1996) states, unless a fish is
poisonous, the concept of trash fish does not exist in many poor develop-
ing countries. People on Quirimba utilized extremely small fish that would
have been discarded elsewhere, for example 5cm-long butterfly fishes,
emperors and damsel fishes. Fish was usually added to stews contain-
ing coconut milk, chillies and sweet potatoes. The small fish could add
flavour and add nutritionally in terms of animal protein, but did not sig-
nificantly increase the cooking time. Firewood was always in short supply
on Quirimba and collecting wood, whether from the mangroves or from
scrubland or from around the machambas, was hard work done mainly by
the women. Small fish could be cooked quickly and required much less fuel
than large fish.
There were some unusual uses of fish and fish products. Boxfish
(Ostraciidae), not eaten in many places because of skin toxins, were well-
liked in Quirimba. They were stuffed with rice and cooked directly in the
fire, their tough skins protecting the flesh from burning. Other toxic fish
such as scorpionfish (Scorpaenidae), catfish (Plotosidae) and most puffer-
fish and tobies (Tetraodontidae) were discarded but some large pufferfish
were retained for sale to Makua people in Pemba and Nampula. Some
Makua people eat these fish and know the special methods for preparing
them to avoid fatal poisoning.
Fishing traditions
Fishing communities all over the world have long traditions and complex
systems of beliefs and superstitions (Ruddle, 1996b), from the South
Pacific (Johannes, 1981) to the Isle of Man (pers. obs.; Manx Heritage
Foundation, 1991). In southern Kenya, where the habitats and fishing
methods are similar to those used on Quirimba and the coastal people
have a broadly similar cultural history with strong Arab influences, strong
fishing traditions still exist (McClanahan et al., 1997). There were few
traditions associated with fishing in Quirimba but it was difficult to ascer-
tain whether this was because there had been a complex system of beliefs
that had gradually disappeared or whether there had never been such a
belief system. The seine net fishers did not seem to have any traditions or
superstitions directly associated with fishing apart from a vague concept
of ‘luck’. One fisher sometimes wore a seamoth fish (Pegasidae) round
his neck as a talisman to bring luck. Seeing dolphins during fishing was
considered good luck. On the other hand, fishers in some parts of Kenya
still hold ritual ceremonies and have sacred sites at sea where offerings
are made to spirits to improve fishing. These rituals are gradually being
lost, in many instances because Islam is increasing in strength and the
younger generation of Muslim fishers see the traditional rituals as against
Islam. Although there has been a lot of population movement around the
Kenyan coastal area, it probably does not approach the large-scale migra-
tion of people around the north of Mozambique as a consequence of the
war. McClanahan et al. (1997) suggest that the more recent Islamization
of culture in Kenya is linked to the decay of the rich and elaborate cul-
tural traditions of coastal management. This may also be the case to some
extent in Quirimba.
There were a number of traditions associated with eating fish or other
marine products. When dugongs were caught, and they were fished in the
area up until the 1990s, they had to be taken to the mosque for blessing
before they were eaten. Dugong meat was given to pregnant women to
make their babies ‘beautiful’. Other fish species were by tradition not eaten
by pregnant women because they were said to harm the baby. (In a place
where a large proportion of women lost babies at late stages of pregnancy
and where infant mortality was high it was not surprising that there were
a lot of superstitions surrounding pregnancy and the health of the unborn

baby.)
Although there was no formal system of rules and traditions for fisheries
management, conflicts within and between user groups were rare. There
were a number of potential sources of conflict – between trap fishers and
net fishers using the same fishing sites, between local Quirimban fishers
and fishers from Quissanga on the mainland side of the bay, and between
Quirimban fishers and Nampula fishers. There were also many opportu-
nities for theft and for ‘cheating’. Seine nets, perhaps the most valuable
possessions on Quirimba, were always left anchored in the water off the
fish landing site overnight and were never stolen during the study period.
Traps were rarely stolen or emptied by people other than their owner.
In interviews, trap and net fishers were asked if theft was a problem
and about conflicts between user groups, but none of those interviewed
expressed any concerns. Most net fishers tried to avoid fishing over areas
containing traps or lifting nets over traps although some were reported to
take the fish out of traps caught in their nets. Trap fishers did not come
into conflict over sites fished or empty the traps of other fishers. Dynamite
fishing has not been witnessed in the Quirimba Archipelago, but in the
past it has been a serious problem in coral areas of Southern Tanzania just
across the border (Guard and Masaiganah, 1997).
In southern Kenya, McClanahan et al. (1997) found that where old tra-
ditions of respect for sea spirits and customs for behaviour at sea had been
lost, theft from set nets and traps and of the nets and traps had increased.
This implies that without the specific superstitions in place that were feared
by all sectors of the community, people (in that case mainly the younger
people to the anger of the elders) would behave without respect for others.
It was therefore remarkable how rare conflicts were on Quirimba, where
religious and superstitious beliefs seemed to be disjointed and did not form
a coherent code of living and where people of a range of different back-
grounds, ethnic groups and beliefs were living in the shadow of a 17-year
civil war. One of the reasons for the high levels of trust and low incidence
of conflict may have been the strong sense of ‘kinship’. People had large
networks of extended families and friends with whom they shared food
and exchanged help and services. Many people married more than once
(some ordinary Quirimban fishers had up to three wives, each living in a
different household) and had siblings with different sets of parents, linking
together disparate parts of the community. This type of complex kinship
was also reported from southern Mozambique (Gengenbach, 1998).
Parents often do not bring up some of their own children, but entrust them
to relatives who may not have children. This complex network of kinship
in a community of 3000 meant that anti-social behaviour such as stealing
from other people’s fishing gear would, first, be likely to be witnessed by

a member of the community and, second, might well affect the ‘kin’ of the
thief. Theft was remarkably rare on Quirimba and most thieves came from
outside the island; for example large-scale coconut theft in the plantation
was almost always done from boats from the mainland.
The main sources of conflict during the study period were land disputes
and land-use issues such as the problem of one person’s goats grazing
on another person’s crop. These were taken to the district administra-
tor to resolve. Land disputes are an acknowledged problem in post-war
Mozambique (Gengenbach, 1998; West, 1998). All land in Mozambique
belongs to the state, and officially can only be acquired on 50-year leases
(Massinga and Hatton, 1997). In most cases in rural areas formal leases
were not issued and land ownership or the right to farm the land was an
informal arrangement. During the war displaced people often took over
land left by others who had moved elsewhere because of the conflict. The
gradual return of refugees to their homes (‘rural resettlement’) has led
to serious land disputes in many parts of Mozambique (Gengenbach,
1998). The current administration system of Quirimba seemed to be in
the process of change. Unlike Tanzania, a few hundred kilometres to
the north, local elders and chiefs did not make the important decisions,
and the whole system of village elders that was in place before independ-
ence does not exist anymore. This was rooted in the decision of the new
FRELIMO government at independence in 1975 to abolish the system
of chiefs in Mozambique (West, 1998). Traditional local hierarchies of
elders were replaced by the system of a Provincial Governor and a District
Administrator (or Chef de Post).
The strongest tradition adhered to by virtually all Quirimbans, whether
of Mwani or Makua origin, was the belief in the power of witch-doctors
and witchcraft. Most people had a little sewn up cotton packet that they
wore round their neck or kept in a pocket containing a good luck charm
from a witch-doctor. People with illnesses went to a witch-doctor for treat-
ment before they went to the clinic and spells were bought for a variety of
reasons, from protecting property against theft to giving people energy.
There were well-known witch-doctors in the area, such as Tanzoor on
Quisiva Island, and well-known thieves who had powers of witchcraft to
help them. However, none of these beliefs were related to the sea and it
seems likely that many of them were brought from mainland villages and
Makua traditions and superimposed on whatever belief system existed
previously in the Quirimbas. It is possible that the Mwani people never
had the strong belief system found in some other fishing communities.
Many of the communities with the strongest belief systems are isolated
island communities such as those in the Pacific (Ruddle, 1996b), where
communities had strong cultural and linguistic identities. The Quirimbas

have been invaded, colonized and formed an important part of trade
routes for Arab and Portuguese traders for over 500 years. This, together
with the large-scale destabilization caused by Portuguese colonial rule
until 1975 and then the 17 years of civil war that followed directly after do
not make for a community with a strong sense of identity or the preserva-
tion of tradition. The current isolation of the Quirimba Archipelago is a
modern phenomenon.
Total value of the fishery

Total annual catch for net fishing was estimated at 440 tonnes, which
would be worth 1320 million meticais or US$110 000 (at the minimum
price of 3000MZN per kilo), and for the trap fishery 60 tonnes, worth
180 million meticais or US$15 000. The total value of the seagrass fin
fishery (not including the women’s invertebrate fishery) of Montepuez
Bay was therefore estimated at 1500 million meticais or US$125 000 (1997
exchange rates) per year (see Table 8.5). About half of this, 750 million
meticais or US$61 560, went directly back into the local economy in the
form of wages to fishers and fish processors. The area of seagrass fished
by the Quirimba fleet was estimated at around 35km2. If the total annual
value of fish caught was US$125 000, then the minimum annual value per
square kilometre of seagrass was estimated at US$3570.
Approximately 400 people on Quirimba Island were employed in
the seagrass fisheries. The annual individual wages for those involved
in the fishery range from 864 000MZN (US$72) for fish processors,
2 160 000MZN (US$180) for ordinary fishing crew, to 3 024 000MZN
(US$252) to 7 392 000MZN (US$616) for trap fishers. Boat owners could
in theory earn from US$1458 to US$4698 annually. Whether they actually
did this was unclear – the boat owners themselves were not clear about the
kind of annual profit they made, and it is likely that a lot of the ‘profit’
went back into maintaining boats and gear.
A provincial and recent historical context

In 1986 a beach seine could be purchased in Cabo Delgado for the
equivalent of 2731 kilos of fish. In 1994 this had risen to 66 915 kilos of
fish (Republic of Mozambique State Secretariat of Fisheries, 1994b). In
the study described here, the 1997 price of a new seine net was given at
between 5 and 10 million meticais. Using the lowest price it was possible
to get for fish on Quirimba, 3000MZN, this was the equivalent of between
1667 and 3333 kilos of fish. Using the maximum price for fish, 5000MZN
this was the equivalent of 1000 to 2000 kilos of fish, more akin to the 1986
relative prices than those in 1994. The price of fish has not risen much
since 1994 but the prices of nets have evidently fallen as supply networks
to the area have improved. In the 1980s the price of fish in Cabo Delgado
was kept artificially elevated; the price of fish decreased nine-fold since
1986 from US$3.37 for a kilo of fish in Pemba to just US$0.39 in 1993
(during the 1980s ordinary people in Pemba did not eat local fish, but
ate fish imported from around Africa – there was a notorious poisoning
incident from freshwater fish – M. Carvalho pers. comm.). Fishers in the
area have very low financial incentives and little access to new gear, pre-
cluding anything more than subsistence fishing (Republic of Mozambique
State Secretariat of Fisheries, 1994a) and also face the storage and
transportation problems mentioned above.
National per capita consumption of fish was estimated at 5.1kg per year
in 1994 (Republic of Mozambique State Secretariat of Fisheries, 1994b). It
is likely that per capita consumption of fish on Quirimba greatly exceeded
this national value. Fishers and their families shared about 1kg of fish per
fishing day. Each person ate approximately 100g per day on fishing days
(16 days per month), giving 1.6kg per month and a minimum of 19.2kg
of fish per year. In addition to this dried fish was also often eaten on the
non-fishing days.
In 1989 Cabo Delgado had 4539 fishers accounting for 0.7 per cent of
the working population, the same proportion of fishers in the population
as was seen on a national scale. Quirimba Island had approximately 400
fishers in a population of 3000, so fishers there accounted for at least 10 per
cent of the population and a much higher percentage of the working popu-
lation. Between 1989 and 1994 the number of fishers nationally increased
from 52 000 to 80 000 and a similar increase would be expected in Cabo
Delgado. This increase has mainly been attributed to the displacement
of the population towards the coast because of the civil war (Republic of
Mozambique State Secretariat of Fisheries, 1994b; Hatton and Massinga,
1994). With this migration there has been a disproportionate increase in
the numbers of people living off subsistence fishing, often the only source
of food and employment to displaced people. The pressure on artisanal
coastal fisheries has been particularly intense because, although there was
a surge of people to the coast in the 1980s and early 1990s, there was no
accompanying development of the local fishing industry or related diver-
sification of activities. This situation is described as having led to the local
population and the refugees collectively mining as opposed to husbanding
the littoral resources (Hatton and Massinga, 1994).
Employment opportunities and salaries

Workers on the coconut plantation or in other daily employment worked
six days a week all year, a total of 312 days per year. Their potential
annual salary would be just over 3 million meticais. or US$260, signifi-

cantly higher than that of the net fishers (see Table 8.4). However, fishing
crew had up to ten days off per month to do other work such as on their
machambas and some were able to supplement their income from net
fishing with other types of fishing, for example spear-fishing. Jobs in the
coconut plantation were limited, but many people claimed to prefer fishing
to plantation work. Fishing may have been chosen despite the less regular
income for social reasons. On the fishing boats there was an atmosphere
of cooperation and everyone, regardless of age or experience, appeared to
be treated equally (with the exception perhaps of the young boys who start
off as apprentices and have to do some of the unpleasant jobs). Although
the hauling of the fishing nets was extremely hard work, much of the
time spent out fishing was spent sailing between fishing sites when fishers
talked, sang and seemed to enjoy themselves. The coconut plantation,
the only source of formal daily employment on the island offered regular
work that did not depend on the weather or the state of a leaky boat. It
also offered extra benefits such as the opportunity to buy some foodstuffs
cheaper and some free medical treatment. However, the plantation was
run by white farmers of European origin and there was some historical
resentment of working for white plantation owners, with the memory of
ruthless Portuguese plantation and sisal factory owners still fairly recent.
The parents of some of the young fishers were employed by the Portuguese
in this way before independence.
Fisheries elsewhere in the region, in Tanzania, the Comores and
Seychelles for example, have developed as local economies have expanded.
Fishers have left traditional fishing methods behind to use outboard
engines and more sophisticated gear. The fishers of Quirimba did not use
outboard engines on their boats. The fishing boat owners on Quirimba
appeared to have been earning enough to potentially be able to buy boat
engines in the future, although it wasn’t something any of them talked
about doing. The main obstacle to this may not have been so much the
cost (although this is likely to be extremely high – import tax of 300 per
cent is commonly charged on certain imports and the cost of transporting
a boat engine from South Africa to the north of Mozambique may also be
enormous) but the availability. If no one is importing engines to Pemba,
which was the case during the study period, no one will be able to buy
them. Much larger sums of money than the local boat owners have access
to would be needed for actually importing numbers of engines from South
Africa.
Another reason for the lack of engines may have been that even with
engines it would not necessarily be efficient to fish on the outer reefs
instead of in the seagrass beds. Weather conditions in the area were very
changeable and it was often not possible to anchor the project’s research
boat off the outer reef of Quirimba to do survey dives. It was regularly not
possible to get the boat out onto the east coast of the island at all because
of strong winds and rough seas. Traditional sailing dhows were almost
certainly better suited to fishing in the shallow seagrass beds than boats
with outboard engines. There was thus no incentive to ‘develop’ fishing
techniques.
Social structure of the fishing fleet

The crews of seine net fishing boats were a mixture of ages, from young
boys of 12 or less to older men in their 50s and 60s. However, the majority
of fishers were young, aged between 15 and 25. Many young men came
to Quirimba as refugees from the fighting around inland villages during
the war. Many came as children and fishing had been their only job, even
though it was not an option that was open to their parents in the inland
villages they came from. In some more developed fishing communities,
for example in the Seychelles (Wakeford et al., 1998) or in parts of the
Caribbean (J. Hawkins, pers. comm.), most of the fishers are from older
generations whereas the young men try to avoid fishing and attempt to
get better salaries for modern, cleaner jobs. In Quirimba there was clearly
no lucrative alternative to fishing so it remained the most common job
for young men. Some of the Quirimban fishers left fishing to work as
traders across the border from Mtwara in Tanzania to Moçimboa on the
Mozambican border, or as crew on boats that did this run, but this was
often temporary or seasonal work.
Older men were more likely to be trap fishers. Trap fishing was a skilled
method of fishing requiring an apprenticeship. It was difficult to learn and
required a large investment in terms of learning the skills involved in making
and maintaining traps and setting them. If someone bought, assembled and
set some traps without guidance from an expert they could quite literally
catch nothing. In the first year of study there were signs that trap fishing was
in decline. The fishery was dominated by elderly men and very few young
men seemed to be going into the trap fishery. However, in 1997 marema
traps became available locally. Instead of fishers having to make a special
journey to the mainland to buy traps, they could buy them in Quirimba
village. There was a sudden increase in people buying traps, and even people
who did not have a boat to set them from started making traps.
The division of fisheries work between the sexes

Women never worked in the seagrass seine net or trap fisheries as fishers
but provided the main labour force in the processing of the fish from the net
fishery. The idea of a woman involved in either of these two fisheries was
completely unacceptable to the men and women interviewed. Men cited

the weakness of women as a reason, although Quirimban women regularly
carried 30-litre water barrels from one end of the island to the other, felled
mangroves, carried the wood long distances and did most of the heavy
agriculture work. Men also cited women’s inability to swim. There was a
traditional reluctance to teach girls to swim even though they were likely to
spend a lot of time on boats, and I was told by a number of different people
that women often drowned in boat accidents. The women interviewed
did not have any aspirations to fish in the trap or net fishery but when
pressed did suggest that economic factors made it impossible to consider
trap fishing. In Tanzania a clear distinction was made between women’s
territory and men’s for marine resource collection (Mtwara – M. Guard
pers. comm.; Bagamoyo – Semesi et al., 1998). Women fished in water up
to waist deep. In deeper water, where a boat would be needed, men took
over. In some Tanzanian coastal communities these designated areas led to
conflicts of interest in the community and to members of the same family
working against each other over these rights. In Kwale District, Kenya,
there was a similarly inflexible division of labour with men fishing and
felling mangroves and women collecting invertebrates, growing food and
collecting firewood. Men there maintained that this was because women
weren’t strong enough to do the ‘men’s’ work, whereas women maintained
that the reason they didn’t do this work was financial (Juma, 1998).
There were a few options for catching fin fish open to women. They
could do the capulana fishing (mentioned previously), they could catch fish
that had been trapped in intertidal pools as part of intertidal gleaning and
they could empty their family’s luwando or fence trap. Women collected
three main groups of invertebrates in the intertidal seagrass beds of the
Montepuez Bay:
● Mbari (Pinctada nigra). Mbari were collected at low spring tides

from the stalks of the large seagrass species Enhalus acoroides.
They were sold for fresh consumption and also dried for sale on the
mainland.
● Macaza (Pinna muricata). Macaza were also collected at low spring
tides from intertidal areas of sand and the seagrass Thalassia hemp-
richii. The majority of the animal was discarded and only the small
adductor muscle was retained to eat.
● Ombe (Barbatia fusca). Ark shells were collected on the intertidal
flats and in Thalassia hemprichii at low tide, the women skilfully
detecting their presence beneath the surface of the sand using their
toes. Ark shells were also either consumed fresh locally or dried and
sold in Pemba.
Some other invertebrates were collected by women and men. Makome

(Chicoreus ramosus – murex shells) were one of the few invertebrate species
routinely collected by men. They were collected incidentally by trap and
net fishers in the seagrass beds. Abandoned marema traps appeared to
provide a good aggregation device for murex shells. Some fishers also
specialized in makome fishing, using canoes and paddling out to deeper
seagrass sites to dive for the shells. Shells caught in deeper water by full-
time murex fishers were much bigger than those caught incidentally by
other fishers. Murex shells were used in a variety of ways. The animal was
cooked in the shell for immediate consumption, the opercula were sold by
the kilo to Tanzania and the shells were burnt on large pyres with other
shells to make lime for painting houses. Despite the large size of many of
these shells there was no evidence of their use in the curio trade.
Kome nlume (Pleuroploca trapezium – tulip shells) were collected almost
entirely incidentally in other fisheries. They were sometimes eaten but
their main value was for their opercula, which were also sold by the kilo
to Tanzania. A remarkably wide variety of other molluscs were caught by
children. These were mainly small molluscs that are found higher than the
level of low spring tides and even those found near the top of the beach
and in small areas of mangrove. The value of these shells appeared to be
solely to supplement the diet of the family and none were sold. Women
who are not normally involved in invertebrate collection also collect these
species on a casual subsistence basis.
Many of the intertidal invertebrates such as mbari (oysters), macaza
(pen shells) and ombe (ark shells) were mainly found at the low tide level
and so were only accessible at low spring tides, just three or four days
per fortnight. This meant that women who collected these molluscs had
a less regular income than net fishers but their catches on these few days
were often more valuable than a day’s net fishing. Women could also fit a
range of activities around the tidal cycle. Some women collected octopus
and other organisms associated with the exposed reef flat on the east coast
of Quirimba around neap tides, then moved round to the lower littoral
species around tides closer to springs. During completely unsuitable tides
(high neaps) women worked on their machambas. Some women also col-
lected the higher littoral molluscs during these tides, such as Strombus sp.
(tiger sand conch).
Alternative sources of income

On Quirimba the main alternative source of employment was subsist-
ence agriculture. Most people on the island were involved to some extent
in agriculture, tending a chamba to provide their family with papaya,
sweet potato, cassava and beans. Most agricultural activity took place
in the wet season (November to April) but many people were involved
all year round. The food available in the local market was very limited
and imported shop-bought food was prohibitively expensive, so most
people relied on food grown in their machambas for the bulk of their food,
with fish or shellfish providing the animal protein. The machambas were
constantly being expanded into the few remaining areas of forest and
scrub by slash and burn clearing methods. On Inhaca Island in southern
Mozambique this method has produced land that is high in nutrients
for the first year of cultivation but that soon becomes poor in nutrients
and becomes impossible to cultivate after 15 years (Serra King, 1995).
Erosion is also a problem in land cleared by slash and burn, and this may
be something that will impact on the coastal habitats. The cash economy
was very limited so there were few shops and it was virtually impossible
to buy a meal or cooked food of any kind in the village, so there were few
opportunities in service jobs. This situation may be typical of a largely
cashless economy for poor rural Mozambicans. In Tanzania many rural
areas are much more developed and there are more opportunities for jobs
with regular salaries and cash (Andersson and Ngazi, 1998). Street food
is common and provides an extra employment opportunity. There was no
tourist industry mainly because of the inaccessibility of the island and the
lack of basic infrastructure.
The gender roles in the fishery at Quirimba were fairly typical of those
found in tropical artisanal fisheries around the world. Women throughout
East Africa, South East Asia and the South Pacific glean intertidal flats
for invertebrates and the fishery for fish is dominated by men (Palau –
Matthews and Oiterong, 1995; South East Asia – Bailey and Pomeroy,
1996; Kenya – Juma, 1998). Although the biology of the women’s inver-
tebrate fishery was not included in the present study and can be found
in Barnes et al. (1998), the socioeconomics were studied to some extent.
Women collected the three main species of mollusc on the intertidal
(mainly Thalassia hemprichii) and upper subtidal seagrass beds (mainly
Enhalus acoroides).
The community on Quirimba utilized a wide variety of resources.
Although seine net fishing in the seagrass beds was the main single
employer, most households used a wide range of sources of subsistence
food and additional income, typically a mixture of men’s income from
fishing by net or trap, food from the machambas grown predominantly
by women, invertebrates gleaned from the seagrass intertidal by women
and some additional invertebrates collected by children. Consequently,
although many people were involved in the seagrass-based fish and inver-
tebrate fishery very few people, except maybe some boat owners, were
entirely reliant on the seagrass fishery. As Bailey and Pomeroy (1996)
found in South East Asia, households and, therefore, communities reliant

on a wide range of resources that changed with seasons and with chang-
ing weather conditions were much more adapted to cope with long-term
changes in the availability of resources. The use of a wide variety of
resources by people in coastal communities was also found in Tanzania.
On Mafia and Zanzibar (Unguja) islands only 11 per cent of people
interviewed relied on a single source of income, with most people com-
bining their main source of income, some sort of fishing, with small-scale
agriculture (Andersson and Ngazi, 1998).
On Quirimba most people did combine some sort of marine resource
use, whether fishing or invertebrate collection, with subsistence agricul-
ture. However, within the invertebrate fishery a change was taking place.
The molluscs (Pinctada nigra, Pinna muricata and Barbatia fusca) were
originally collected by women as a subsistence commodity that was eaten
locally. In the community workshops the women who collected them
reported that in the last few years the prices they could get for them in
Pemba had risen sharply as demand for dried seafood increased. It was
therefore much better to sell them on the mainland rather than eat them
themselves or sell them on Quirimba. Others in the community associated
this change in use of invertebrates with a decline in their numbers and
said that it was virtually impossible for an ordinary Quirimban person to
afford to buy these preferred species of shellfish. The shellfish produced
by the women were therefore the main commercial product of ordinary
people in Quirimba, whereas the fish caught by the trap and net fishers,
although also forming an important commodity for the boat owners and
for dried fish traders, was the main source of subsistence food for ordinary
households on Quirimba.
In Tanzanian coastal communities a similar pattern has been observed,
with an increase between 1993 and 1998 of the local consumption of fish
and vegetables and the complete disappearance of intertidal invertebrates
from local diets. There too, prices have increased because of a combina-
tion of increased demand due to growing populations and the developing
tourism industry and increased scarcity as marine resources are overfished
(Andersson and Ngazi, 1998). The situation in Tanzania is perhaps ten or
more years ahead of what was recorded in Quirimba in terms of economic
development. Cabo Delgado is one of the least economically developed
provinces in Mozambique (Hatton and Massinga, 1994). In Inhaca Island,
southern Mozambique, the invertebrates collected by the women are eaten
by the family and it is the fish that is sold for cash (Wynter, 1990), perhaps
reflecting the close proximity of a tourist market for fish and the national
capital.
Integrated coastal zone management in northern Mozambique

A comprehensive review of integrated coastal zone management can be
found in the proceedings of a national workshop on integrated coastal
zone management in Mozambique (Lundin and Lindén, 1997). The only
example of integrated coastal zone management in Cabo Delgado prov-
ince is that of the Mecufi Coastal Zone Management Project (Massinga,
1997).
Along with the other East African and Western Indian Ocean Island
States who signed the ‘Arusha Resolution on Integrated Coastal Zone
Management in Eastern Africa (including Island States)’, Mozambique
has pledged to develop and implement integrated coastal zone man-
agement (ICZM) programmes. Within the main government depart-
ment involved in coastal zone management, MICOA (Ministry for the
Coordination of Environmental Affairs) there is now a Coastal Zone Unit
with specific responsibility for coastal management issues. Active coastal
zone management has largely been restricted to some large marine pro-
tected area projects in the southern sector of the country, for example the
Inhaca Island marine park, the Bazaruto Archipelago marine protected
area and the Xai-Xai Integrated Coastal Area Management project.
However, projects are gradually being developed throughout the country,
for example the Mecufi Coastal Zone Management Project just south of
Pemba or the Island Management project on Isla de Moçambique, which
are currently in progress. City coastal zone management projects are also
planned for the city of Nacala in Nampula Province and the city of Beira
in Sofala Province.
Having only emerged from the civil war in 1992, Mozambique has the
difficult task of reconciling coastal zone management programmes with
the need for rapid development of large areas left by the war without
even the most basic infrastructure and facilities. It also has to address the
problem of a population that, during the civil war, gravitated towards
coastal areas. Many refugees have remained and are now settled in
their coastal lifestyles. It is estimated that more than 40 per cent of the
Mozambican population live in coastal districts (MICOA, 1997). The pres-
sures on coastal resources are manifold: increased populations through
migrations, through the natural increase in population that is being expe-
rienced throughout the region and also increased exploitation of resources
by outside user groups such as foreign fishing fleets and recreational users
(Hatton and Massinga, 1994).
Itinerant fishers
Small numbers of itinerant fishers from local mainland villages have
fished seasonally in the Quirimbas, but in the five years preceding the
study increasing numbers of fishers from Nampula started coming to

Quirimba and other islands in the Archipelago. They inhabited the islands
in groups of tens to hundreds of fishers on a seasonal basis, usually staying
between three and six months during the dry season. These fishers said
that they have been forced to fish in the Quirimbas by the depletion of
their own nearshore fish stocks. Nampula has indeed had a much more
intensive artisanal fishery sector than Cabo Delgado. In 1994 there were
reportedly 180 beach seines in the Province of Cabo Delgado and 10 882
in Nampula. The coast of Nampula is about 50 per cent longer than that
of Cabo Delgado but it had a seine netting intensity 600 per cent greater
(Republic of Mozambique State Secretariat of Fisheries, 1994b). Fishers
working for companies in Nampula fished with an intensity and a com-
mercial intent that was very different from the Quirimbans’ subsistence
approach. Nampula fishers returned to their home towns with boats full
of the dried fish and invertebrates they had collected in the Quirimbas,
part of which they kept for their families over the wet season and part they
sold commercially in Nampula. Fishers from Nacala (provincial capital of
Nampula) were recognized as a major component of the fishing industry in
the Macomia district, further north from Quirimba in Ibo district (Wilson
et al., 1996).
Up to the end of 1997 no active regulation of fishing activity in the
Quirimbas was in place. Far from trying to control the systematic
depletion of their marine resources themselves, Quirimban fishers were
welcoming the invasions of Nampula fishers despite the gradually emerg-
ing problems of wells running dry, traps being caught in the nets of the
visiting fishers and other small but significant signs of potential impend-
ing conflict. Possible reasons for the Quirimbas welcoming behaviour
include a desire to maintain contact with people from a more developed
part of Mozambique. One local fisher said he enjoyed the company of the
Nampula fishers because they were more educated in Islam than most
local people and he found them interesting. The Quirimban fishing com-
munity also seemed generally to be a very open, welcoming community
(the way they welcomed this study is a good example), maybe an extension
of the ‘kinship’ idea addressed previously.
Highly mobile fishing fleets are not unusual in the East African region
and have led to conflicts elsewhere. For instance, in southern Kenya, itin-
erant fishers from Pemba Island in Tanzania, having reputedly depleted
their own fish stocks, fished in traditionally managed reef areas against
local management regulations (McClanahan, et al., 1997). In Tanzania,
iceboats from Dar es Salaam fish right the way down the coast of
Tanzania, often using dynamite and other damaging methods.
One of the problems with this situation of over-exploitation by itinerant
fishers in the Quirimbas is that, for the purpose of future management or

legislation, it will be very difficult to say who is a Quirimban resident and
to whom the marine resources of the Quirimbas actually belong. During
the war people were forced to move around a lot to escape the intense
fighting in rural areas. Some refugees stayed where they fled to, others
have moved back to their home villages. Many people have also contin-
ued to move, whether between different opportunities to work or between
partners or family members displaced throughout the area. There is also a
lot of movement between towns and rural areas. Much trading is done by
boat, so as well as the movement of people between the islands and main-
land towns and villages there is an additional group of people who are
continuously moving up and down the coast between ports from Nampula
to Mtwara in southern Tanzania. The Mwani people of the coast are his-
torically fishers and traders and the trading is still an important part of
coastal culture.
It seemed to be difficult for people on the islands to think of the sur-
rounding marine resources as theirs. The majority had moved around so
much in their lifetime that the idea of having to move on again because
of resource depletion did not seem as serious as it perhaps may have
seemed to a more settled community. One fairly typical fisher interviewed
on Quirimba Island was just 24 years old but had already moved numer-
ous times and had changed his fishing method on each occasion. He had
moved from freshwater fishing as a boy in an inland village, to line fishing
from a coastal village, octopus fishing on Quisiva, through a variety of
other locations and fishing methods to trap fishing on Quirimba. He had
family members on four of the other islands and in numerous villages in
the mainland. To him the idea of having to leave Quirimba because the
‘fish had run out’ was not distressing at all. The threat of fish or inverte-
brates running out was not normally considered, and one interviewee said
‘There will be fishers at Quiwandala [the main fish landing site] until the
end of the world’.
Threats to sustainable resource use

This study has identified four main threats to sustainable resource use in
the area, namely:
Population growth Coastal populations throughout Mozambique are on

the increase. During the 1990s the population of the Quirimbas increased
through a higher birth rate and immigration of refugees. Quirimba Island
had a population of nearly 3000 but no provision for waste manage-
ment. A growing population will inevitably put increasing pressure on the
marine resources. The need for firewood from mangrove areas and for
fish to provide protein in local diets will increase. The pressure to exploit
resources for sale outside the community to bring in much-needed cash for
medicines and other manufactured goods will increase and these commu-
nities may already be exceeding sustainable levels of exploitation.
Influx of visiting itinerant fishers from outside the local area The influx
of itinerant fishers was a recent phenomenon and had only really had a
major impact on the islands since the beginning of the 1990s (Gessner pers.
comm.). It is likely, however, that the numbers of itinerant fishers and the
intensity with which they fish will increase as the pressures on fisheries and
other resources in their home villages and towns increase and more people
seek the reputed wealth of resources in the Quirimbas. The commercial
fishing sector is also likely to grow to feed the expanding tourist market.
The difficulties of controlling and managing resource use and develop-

ment The Quirimba Archipelago is a remote area and has been neglected
for a long time in terms of infrastructure, development and effective law
enforcement. There was a low official authoritative presence in the area
and no officials in the capacity of marine resource protection and manage-
ment. Fishing gear regulations were not enforced and most people were
not even aware of their existence. The local administration system did
seem to be improving, but even local administrators with the best inten-
tions for the protection of resources and the implementation of legislation
are unlikely to have the time or resources to do very much about managing
resource use on their own.
Because of the lack of official intervention there was great potential
for unscrupulous outsiders exploiting the marine resources of the islands
to the detriment of the needs of the local people (Massinga and Hatton,
1997). The sea cucumber and shark fin fisheries run in the Archipelago
by large companies for export to the Far East are a good example of
this. Intermediaries exporting these commodities are likely to have made
a lot of money but very little was paid to the local people involved. In
both cases the methods of extraction were so intensive that the potential
for local people to exploit these resources sustainably in the future may
already have been lost. Local fishers reported that sites where they once
skin-dived for abundant sea cucumbers now do not have any at all. Sharks
were also a rare sight now throughout the Archipelago.
Many foreign business people are interested in exploiting some resource
or other that they have heard is abundant in the Quirimbas, from cray-
fish to the opportunities for tourism. Their attitude seems to be that the
resources of the Quirimba Archipelago are there for the taking and it is
unlikely that anyone will interfere actively. It is important that the rights
of the Quirimban residents to their resources are established. Once they

are assured, people of the Quirimbas can begin to take responsibility for
the protection or guardianship of those resources before, for example,
land is sold to a businessperson as a theme park or sea cucumbers are dried
and sent to the Far East as an aphrodisiac.
Indirect threats to the marine ecosystems An important new growth

industry in the north of Mozambique was forestry and with two major
rivers entering the sea in the vicinity of the Archipelago the potential
for the indirect impacts of siltation of reefs, seagrass beds and man-
groves was great. As with all other development in the remote north
of the country, logging of coastal forests was difficult to monitor and
control and it is likely that illegal logging was a serious problem in the
area. Large areas of old forest were logged and this was likely to lead to
increased erosion leading to silt being washed downriver and into the sea
around the Archipelago. Slash and burn agriculture is widespread on the
mainland and this will also lead to increases in soil erosion and potential
siltation problems. Coral reefs are particularly susceptible to the effects
of heavy siltation but seagrasses too are unable to tolerate high levels of
sedimentation with low light levels (Bach et al., 1998).
Management implications and problems

The four major threats identified above do not feature strongly in the
concerns expressed by the residents of the Archipelago. The majority
of the threats have yet to manifest themselves in the form of dramatic
decreases in fish or invertebrate catches. The fishers did not complain
about the outside fishers using their fishing sites but did complain about
the high prices of fishing gear. The main concern of women was that
with the growing demand for their dried shellfish products in Pemba and
other mainland towns, the favoured species of shellfish were becoming an
unaffordable luxury. The majority of people interviewed who were resi-
dent permanently in the islands seemed satisfied with the present situation.
They had access to agricultural land to grow basic foods such as sweet
potatoes, cassava and papayas and they supplemented this diet with fish
and shellfish.
On a day-to-day basis, people throughout the Quirimbas suffered
through lack of basic health care. Paying for medicines was often dif-
ficult in a virtually cashless economy. Minor injuries and illnesses often
went untreated and led to more serious conditions and the long-term
loss of income. For example, a fisher on Quirimba with a small infected
sore did not receive treatment, the injury became more serious and he
was unable to fish for a period of months, inflicting hardship on himself
and his family. Malaria and tuberculosis are both common potentially
life-threatening illnesses, and both can be controlled using medicines that
are cheap by Western standards, but even the most basic treatments can
be prohibitively expensive. There will be an increasing pressure on the
people of the Quirimbas to find ways of earning cash so that they can deal
with these emergencies and also so that they have the freedom to buy the
consumer goods that are becoming available.
Based on this analysis, four basic needs can be identified that could
begin to address the potential problems in the Quirimbas:
Education The general level of education is very low, particularly for

women. Boys generally receive a few years of education and learn to read,
write and speak Portuguese. Many girls are educated for less than one
year and remain illiterate. It is important that all children are educated to
a standard where they are literate and able to access educational materi-
als and the media. Children should either be given a firm grounding in
Portuguese (whilst it is ensured that local languages and cultures are not
suppressed), or educational materials, radio programmes and newspapers
should be produced in local languages, particularly Kimwani, which is the
main island language and is marginalized at present. It would, of course,
be ideal to give children both of these opportunities. If people were given
the opportunity to reach a higher level of education the ecological and eco-
nomic arguments for them to conserve their own resources may be more
easily communicated. A higher level of education may also offer people
the opportunity to find employment other than fishing and invertebrate
collection and would make people more employable in service industries
that will accompany the development of tourism. Literate, Portuguese-
speaking Quirimban residents could then by employed in the inevitable
tourist developments, rather than imported labour from the cities as has
happened elsewhere.
Legislation The ever-increasing numbers of itinerant fishers in the

Archipelago need to be regulated in some way. On some of the smaller
islands in the Archipelago every available space is covered with their
makeshift shelters. All are there both to catch fish to sustain their family
over the wet season and also to make a profit selling their fish inland.
Without some sort of limit or regulation this component of the fishery
could lead to its decline. This problem may be one that has to be addressed
at its source, in the majority of cases in Nampula, and is therefore in
essence a political problem requiring political will to solve. Social or envi-
ronmental problems that have forced these fishers to look outside their
province for additional income on such a huge scale need to be addressed
within the province. The use of unacceptably small mesh sizes, in some
cases even finer material such as sacking, and other fishing gear infringe-
ments also need to be addressed. Active fisheries monitoring in the islands
is definitely required.
Research and monitoring More research is needed into the sustainability

of the resource use methods currently in use and into potential alternative
use of resources. A great positive step would be the establishment of a
marine research centre on one of the larger islands. A large GEF (World
Bank Global Environment Facility) loan has been allocated to the north-
ern Mozambican coastal area and there are currently plans for a marine
station at Pemba to the south of the Archipelago or in Moçimboa da Praia
to the north (S. Bandeira pers. comm.). This would give scientists who are
based in Maputo a good base from which to conduct research and perhaps
most importantly to monitor changes and developments in the area and
their impact on the marine ecosystems. At present there is no scientific
monitoring. For example, the inaccessibility of the area and lack of marine
scientists have meant that the status of the Quirimbas coral reefs during
1998’s catastrophic coral bleaching events remains unknown. The estab-
lishment of a marine research centre would also play an important role in
the education programme. Such a centre could serve as a base for training
environmental educators and provide educational facilities for schools
and community groups, and an opportunity for local people to actively
participate in research and decision-making.
Implementation To implement any gear regulations it would be neces-

sary to provide financial and practical help to fishers to help them pur-
chase more favourable gear types. The education programme must instil
in the island residents the idea of the marine resources of the Archipelago
as theirs to conserve and use sustainably now so that they will continue
to provide food and employment to their children and grandchildren. In
this way, enforcement of legislation could come in part from within the
community. Only in the role of guardians of their own future resource
use will the people of Quirimba be motivated to participate in manage-
ment. The training of local people as environmental educators to work in
schools and with fishers on a long-term basis would be one way of devel-
oping environmental awareness in the area and would also overcome the
language problem mentioned earlier. An integrated programme of educa-
tion, research in close cooperation with local people, legislation to protect
vulnerable resources and to prevent destructive practices, and a long-term
programme of implementation and monitoring are needed.
How would possible changes in fishing patterns affect the economy?

One potential management option would be to move away from seine
netting as the main fishing method and increase trap fishing. Each seine
fishing boat covers a large area of Enhalus acoroides on each trip and the
small trap fishing areas required by individual trap fishers would be much
smaller than the available area. How would an increase in trap fishing and
a decrease in net fishing affect the people involved? From an economic
point of view everyone employed in the seine net fishery, with the excep-
tion of the fishing boat owners, would have a higher income if they were
employed as trap fishers (after the initial expenditure on boats and traps).
Net fishers could potentially earn double their salary as trap fishers. They
would also catch fewer fish per person so if the same number of men were
involved in the fishery, the total fishing intensity would drop (trap fishers
catch an average of 6.7kg per trip whereas net fishers catch an average of
9.4kg per person per trip – even taking into account the extra days fished
by trap fishers, fewer fish would be caught) so putting less pressure on
resources. There would, however, be some problems with this change in
fishing strategy. There would be the initial problem of fishers being able to
raise the capital to buy a canoe and traps. Trap fishers also had an outlay
of about 160 000MZN every few months, for replacing their traps. This
worked out as an expense of approximately 3350 MZN per fishing day.
If fishers moved from net fishing to trap fishing this would restrict the
business of the seine net boat owners. As the net fishing boat owners were
the only businesspeople on the island, if they were to lose their livelihood
it could have a negative effect on the overall development of the area.
Boat owners invested money from fishing in other local businesses such as
shops and market stalls. These provided employment for members of the
owners’ families and other local people, and provided a service. Each boat
owner provided a reliable income for anything from five to 12 fishers and
also part-time income for the women and children who cleaned fish.
Conclusions
In conclusion, the study reported here showed that fishing was the single
most important source of food and employment on Quirimba Island,
along with subsistence agriculture and the women’s invertebrate fishery,
but the majority of people and households relied on a combination of
marine resource use and agriculture. Seagrass beds were the main habitat
to be exploited on Quirimba and therefore the most valuable habitat to
the community. There was a hierarchy of earnings in the seagrass fishery
in Quirimba, with boat owners being the highest earners, then trap fishers
followed by boat captains and boat crew. Net fishing was the biggest single
source of employment on the island and yielded the lowest wages of jobs
for young men. Net fishing may have been preferred to better-paid trap
fishing or plantation work because it was sociable, flexible and seemed to
be enjoyable. Small fish may have been preferred for social and economic
reasons. They could be used easily for barter and were easy to cook, dry and
sell. Interestingly, there were remarkably few conflicts between the users of
marine resources despite the fact that there were overlapping uses of the
resource and many recently arrived fishers. However, an increasing popula-
tion and a declining resource base will inevitably lead to competition and
environmental degradation unless there is investment in the community.
Note
1. There were many people in Quirimba village and in Kumilamba who gave generously
of their time to assist in the collection of the information in this chapter. Much of the
information was collected with the help of trap fisher Anibal Amade who suggested
many useful interviewees, provided translation assistance where necessary and also pro-
vided detailed information on the trap fishery. Seine net fishers Mussa and Lamu from
Quirimba spent hours explaining the workings of the Quirimba fisheries and introduced
me to many other fishers, including reef flat gill netters. Saidi Ndiki, an older gill net
fisher, was interviewed on a number of occasions about this fishery. Captains Ibrahimo,
Manueli and Jabira from the seine net fishery provided a great deal of useful informa-
tion. Oscari, Zaidu and Momadi from Santa Maria were key informants on the fisheries.
For general information on the economics of life on Quirimba, Awaje Shale and her
family were very helpful. The workshop with the invertebrate fishers was organized with
Fatima Mussa and other women from OMM (Mozambican Women’s Organization) in
the village. Mario Carvalho and Alex Corrie of The Frontier-Moçambique project pro-
vided assistance with translation and setting up interviews and so on, and the rest of the
Frontier staff, visitors and volunteers provided invaluable support at various times. The
Quirimbas National Park was designated in 2002.
References
Andersson, J. and Z. Ngazi (1998), ‘Coastal communities’ production choices. Risk diver-
sification and subsistence behaviour: responses in periods of transition’, Ambio, 27(8),
686–93.
Bach, S.B., J. Borum, M.D. Fortes and C.M. Duarte (1998), ‘Species composition and
plant performance of mixed seagrass beds along a siltation gradient at Cape Bolinao, the
Philippines’, Marine Ecology Progress Series, 174, 247–56.
Bailey, C. and C. Pomeroy (1996), ‘Resource dependency and development options in coastal
Southeast Asia’, Society and Natural Resources, 9(1), 190–99.
Barnes, D.K.A., A. Corrie, M.W. Whittington, M.A. Carvalho and F.R. Gell (1998),
‘Coastal shellfish resource use in the Quirimba Archipelago, Mozambique’, Journal of
Shellfish Biology, 17(1), 51–8.
Birkeland, C. (1997), ‘Disposable income in Asia – a new and powerful external pressure
against sustainability of coral reef resources on Pacific Islands’, Reef Encounter, 22,
9–13.
Boxer, C.R. (1963), ‘The Querimba Islands in 1744’, Studia, 11, 343–60.
Chambers, R. (1997), Whose Reality Counts? Putting the First Last, London: Intermediate
Technology Publications.
Dahalani, Y. (1997), L’Impact de la pêche au ‘djarifa’ sur le recrutement des populations
des poissons et des crustacés en face des mangroves du littoral côtier de Mayotte (baie de
Chiconi), Centre d’Océanologie de Marseille and Service des Pêches et de l’Environment
Marin de Mayotte.
Dayaratne, P., O. Linden and R. De Silva (1995), ‘Puttalam Lagoon and Mundel Lake, Sri
Lanka: A study of coastal resources, their utilization, environmental issues and manage-
ment options’, Ambio, 24(7), 391–401.
Gayanilo, F.C. and D. Pauly (1997), FAO Stock Assessment Tools: Reference Manual,
Rome: ICLARM/FAO.
Gell, F.R. and M.W. Whittington (2002), ‘Diversity of fishes in seagrass beds in the Quirimba
Archipelago, northern Mozambique’, Marine and Freshwater Research, 53(2), 115–21.
Gengenbach, H. (1998), ‘“I’ll bury you in the border!”: Women’s land struggles in post-war
Facazisse (Magude District), Mozambique’, Journal of Southern African Studies, 24(1),
7–36.
Guard, M. and M. Masaiganah (1997), ‘Dynamite fishing in southern Tanzania: Geographical
variation, intensity of use and possible solutions’, Marine Pollution Bulletin, 34(10),
758–62.
Hatton, J. and A. Massinga (1994), The Natural Resources of Mecufi District. Mecufi Coastal
Zone Management Project, Maputo, Mozambique: Ministry of Environment.
Jennings, S. and N.V.C. Polunin (1996), ‘Fishing strategies, fishery development and socio-
economics in traditionally managed Fijian fishing grounds’, Fisheries Management and
Ecology, 3, 335–47.
Johannes, R.E. (1981), Words of the Lagoon, Berkeley, CA: University of California Press.
Juma, S.A. (1998), ‘Men, women and natural resources in Kwale District, Kenya’, Ambio,
27(8), 758–9.
Lindén, O. (1993), ‘Resolution on integrated coastal zone management in East Africa signed
in Arusha, Tanzania’, Ambio, 22(6), 408–9.
Lindén, O. and C.G. Lundin (1996), ‘The journey from Arusha to the Seychelles: Successes
and failures in integrated coastal zone management in Eastern Africa and island states’,
Proceedings of the Second Policy Conference on Integrated Coastal Zone Management
in Eastern Africa and Island States, Seychelles, 23–25 October: World Bank and Sida/
SAREC.
Loureiro, N.L. (1998), ‘Estudo da ictiofauna coralina e pesqueira do distrito de Mecúfi,
Província de Cabo Delgado’, Tese de Lienciatura, Universidade Eduardo Mondlane,
Maputo, Mozambique.
Lundin, C.G. and O. Lindén (eds) (1997), ‘Integrated coastal zone management in
Mozambique’, Proceedings of the National Workshop, Inhaca Island/Maputo,
Mozambique 1996: World Bank and SIDA/SAREC.
Macia, A. and L. Hernroth (1995), ‘Maintaining sustainable resources and biodiversity while
promoting development – a demanding task for a developing nation’, Ambio, 24(7–8),
515–17.
Manx Heritage Foundation (1991), Manx Sea Fishing, Douglas, Isle of Man: Manx Heritage
Foundation Publications.
Massinga, A. (1997), ‘Mecufi coastal zone management project’, Proceedings of the National
Workshop on Integrated Coastal Zone Management in Mozambique, Inhaca Island and
Maputo, Mozambique, May 1996.
Massinga, A. and J. Hatton (1997), ‘Status of the coastal zone of Mozambique’, Proceedings
of the National Workshop on Integrated Coastal Zone Management in Mozambique,
Inhaca Island and Maputo, Mozambique, May 1996.
Matthews, E. and E. Oiterong (1995), ‘Marine species collected by women in Palau,
Micronesia’, Micronesia, 28(1), 77–90.
McClanahan, T.R., H. Glaesel, J. Rubens and R. Kiambo (1997), ‘The effects of traditional
fisheries management on fisheries yields and the coral-reef ecosystems of southern Kenya’,
Environmental Conservation, 24(2), 105–20.
McManus, J.W. (1993), ‘Managing seagrass fisheries in Southeast Asia: An introductory
overview’, in M.D. Fortes and N. Wirjoatmodjo (eds), Seagrass Resources in South
East Asia, technical papers from the advanced training course/workshop on Seagrass
Resources, Research and Management (SEAGRAM 2), Quezon City, Philippines, 1990.
MICOA (Ministry for the Coordination of Environmental Affairs, Mozambique) (1997),
‘Territorial planning of the coastal zone of Mozambique – methodology’, Proceedings of

the National Workshop on Integrated Coastal Zone Management in Mozambique, Inhaca
Island and Maputo, Mozambique, May 1996.
Munro, J.L. (1996), ‘The scope of tropical reef fisheries and their management’, in N.V.C.
Polunin and C.M. Roberts (eds), Reef Fisheries, London: Chapman and Hall.
Pauly, D. (1979), ‘Theory and management of tropical multispecies stocks: A review with
emphasis on the Southeast Asian demersal fisheries’, ICLARM Stud. Rev. No. 1, 35 pp.
Republic of Mozambique State Secretariat of Fisheries (1994a), Master Plan, Maputo,
Mozambique.
Republic of Mozambique State Secretariat of Fisheries (1994b), Sector Report, Maputo,
Mozambique.
Ruddle, K.R. (1996a), ‘The geography and human ecology of reef fisheries’, in N.V.C.
Polunin and C.M. Roberts (eds), Reef Fisheries, London: Chapman and Hall.
Ruddle, K.R. (1996b), ‘Traditional management of reef fishing’, in N.V.C. Polunin and C.M.
Roberts (eds), Reef Fisheries, London: Chapman and Hall.
Semesi, A.K., Y.D. Mgaya, M.H.S. Muruke, J. Francis, M. Mtolera and G. Msumi (1998),
‘Coastal resources utilization and conservation issues in Bagamoyo, Tanzania’, Ambio,
27(8), 635–44.
Serra King, H.A. (1995), ‘Estudo da dinâmica da matéria organica do solo após corte e quei-
mada, Ilha da Inhaca’, Tese de Licenciatura, Universidade Eduardo Mondlane, Maputo,
Mozambique.
Sillitoe, P. (1998), ‘The development of indigenous knowledge: A new applied anthropology’,
Current Anthropology, 39(2), 223–40.
Sousa, A.G. (1960), ‘As Ilhas Quirimbas’, unidentified Portuguese historical journal.
UNDP (2006), Human Development Report 2006, http://hdr.undp.org/en/reports/global/
hdr2006/, accessed 24 August 2009.
Van der Elst, R., E. Bernadine, J. Narriman, G. Mwatha, P. Santana Afonso and D. Boulle
(2005), ‘Fish, fishers and fisheries of the Western Indian Ocean: Their diversity and status.
A preliminary assessment’, Philosophical Transactions of the Royal Society A, 363(1826),
263–84.
Wakeford, R.C., G.P. Kirkwood and C.C. Mees (1998), ‘Management of the domestic
fishery of the Seychelles’, poster at the FSBI Annual International Symposium on Tropical
Fish Biology, University of Southampton, UK, July 1998.
West, H.G. (1998), ‘“This neighbour is not my uncle!”: Changing relations of power and
authority on the Mueda Plateau’, Journal of Southern African Studies, 24(1), 141–61.
White, A.T., L.Z. Hale, Y. Renard and L. Cortesi (eds) (1994), Collaborative and Community-
based Management of Coral Reefs: Lessons from Experience, West Hartford, CT:
Kumarian Press.
Wilson, J.D.K., D.A. Pinto, J.B. Gomes and A. Sumale (1996), ‘Comercialização e dis-
tribuição de pescado, Distrito de Macomia, Província de Cabo Delgado, Relatório da 1a e
2a Fases do Estudo’, Instituto de Desenvolvimento da Pesca de Pequena Escala, Maputo,
Mozambique.
Wynter, P. (1990), ‘Property, women fishers and struggles for women’s rights in
Mozambique’, SAGE, 7(1), 33–7.
9. The link between ecological and
social paradigms and the sustainability of
environmental management: a case study
of semi-arid Tanzania
Claire H. Quinn and David G. Ockwell
Introduction
Over the last two decades, sustainable development has become widely
accepted as the goal that environmental management should strive to
achieve. The idea of sustainable development has evolved due to the inherent
tensions between economic development and the desire to protect the envi-
ronment. The most commonly used definition comes from the Brundtland
Report, which defines sustainable development as: ‘Development that
meets the needs of the present without compromising the ability of future
generations to meet their own needs’ (WCED, 1987, p. 43).
The argument in favour of sustainable development has often been
characterized as a response to the fact that economic development
in a capitalist market economy leads to environmental degradation
(Carter, 2001; Anand, 2003). This, however, fails to recognize that some
approaches to protecting the environment can result in negative economic
impacts where such protection prevents previous economic uses of natural
resources – a particularly important issue for developing countries where
the poorest people often depend most directly on natural resources for
their livelihoods. In such situations, the challenge for environmental man-
agers working to achieve sustainable development is how to protect and
conserve natural resources at the same time as protecting the livelihoods of
poor people. Furthermore, management approaches also need to enable
poor people to develop opportunities to improve their economic situation
and lift themselves out of poverty in the long term. Recognition of this
link between sustainable environmental management and the eradication
of poverty is central to the thinking behind goals 1 and 7 of the United
Nations Millennium Development Goals, which aim, respectively, to erad-
icate extreme poverty and ensure environmental sustainability. Within this
context, therefore, ‘environmental management for sustainable develop-
ment’ needs to protect the environment at the same time as protecting and
developing the livelihoods of those people who depend on it.
282
A case study of semi-arid Tanzania 283
Using the case study of semi-arid Tanzania, in this chapter we highlight

the reciprocal relationship between the environment and society. We
demonstrate how the paradigms that define environmental managers’ and
policy-makers’ conceptions of ecological and social problems are integral
to defining the policy discourses (see Ockwell and Rydin, Chapter 6, this
volume) that shape their choice of management solutions. Furthermore,
we argue that if these paradigms fail to recognize the reciprocal link
between the environment and society, the resulting management solutions
will fail to protect and enhance the livelihoods of poor people. The chapter
begins with an analysis of the reciprocal relationship that defines and
shapes both the ecology and socioeconomics of semi-arid regions. It then
moves on to explore the different paradigms that have dominated concep-
tions of ecological and social problems in the region and how these para-
digms have shaped policy discourses and environmental management. The
chapter concludes by highlighting how recognition of the reciprocal rela-
tionship between people and the environment within the paradigms that
shape policy and management decisions is critical to achieving sustainable
development.
The reciprocal relationship between environment and society

Understanding what defines and shapes a particular ecosystem has tradi-
tionally been assumed to be a purely ecological question. More recently,
however, as well as recognizing the importance of ecological factors in
defining semi-arid regions, observers have begun to recognize the defining
nature of human activities. This stems from an increased understanding
of the reciprocal relationship that exists between the environment and
human society. Focusing on East Africa, and Tanzania in particular, to
demonstrate this reciprocal relationship we begin by examining the cli-
matic and vegetative characteristics of semi-arid regions. We then show
how people and their livelihoods play an important role in shaping the
ecology of semi-arid regions and how, in turn, the ecology of the region
shapes people’s livelihoods.
Climatic and vegetative characteristics of semi-arid regions

Climate, the average weather conditions such as temperature, day length
and rainfall (O’Brien, 1993), forms a key attribute of semi-arid regions.
More precisely, variability in climate, especially rainfall, since temperature
and day length are less variable near the equator, has a huge influence
on the nature of semi-arid regions. In sub-Saharan Africa the semi-arid
regions are influenced by the Intertropical Convergence Zone (ITCZ), a
broad low pressure zone in which the trade winds meet and pick up mois-
ture from the equator and move it north and south towards the tropics
(Griffiths, 1972; Mussa, 1978). In East Africa rainfall is influenced by the

movement of trade winds over the Indian Ocean and additionally by the
El Niño Southern Oscillation (Lovett, 1993). These major air and ocean
current circulations mean that the semi-arid regions in East Africa are
subject to highly variable rainfall concentrated in one or two rainy seasons
a year separated by relatively long dry seasons (Van Keulen and Seligman,
1992). While parts of East Africa have a single rainy season that stretches
from November/December to April/May (Griffiths, 1972; Peberdy, 1972),
most of the semi-arid region in Tanzania experiences a short break in
the rains so that there are, in fact, two rainy seasons (Kingdon, 1971;
Hamersley, 1972). These peaks in rainfall correspond with the passage of
the ITCZ as it tracks the sun’s movement in the sky caused by the earth’s
axial tilt (Kingdon, 1971; Griffiths, 1972; O’Brien, 1998). The equatorial
location of Tanzania also means that mean monthly temperatures vary by
only 3–4°C (Griffiths, 1972; Mussa, 1978). The seasons therefore broadly
divide into a hot and dry season from December until March, a cooler
rainy season in April and May followed by a long cool dry season with
the short rains in November corresponding with a rise in temperature
(Hamilton, 1989; Lovett et al., 2002). In addition to large-scale influences
on climate, the position of lakes and mountain ranges as well as altitude
can create localized weather conditions (Kingdon, 1971; Griffiths, 1972;
Maro, 1978, Hamilton, 1982; Lovett, 1993), so that rainfall is highly
variable spatially as well as temporally.
Semi-arid regions are subject to both inter-annual and intra-annual
variations in rainfall. This means that rainfall fluctuates between years
but also during seasons (Mortimore, 1998). Large volumes of rain can fall
over quite short periods so that while seasonal and yearly averages may
be maintained the patterns of rainfall between seasons and years might be
very different. The short rains in November are particularly variable and
may not appear at all in some years (Griffiths, 1972). Failure of the rains
occurs when the trade winds fail to meet in the right place at the right time
(Walter, 1939). Rainfall can also fluctuate by as much as 60 per cent from
the average between years (Peberdy, 1972), with both single and multi-year
droughts a common occurrence. However, long-term studies of rainfall
patterns in sub-Saharan Africa have not shown conclusively a downward
trend in the amount of rainfall received in these regions overall, although
sub-Saharan Africa has been subject to a period of particularly low rain-
fall since the 1980s. This period of aridity is not unprecedented but rises in
sea temperature caused by anthropogenic climate change could contribute
to a situation of long-term aridity (Nicholson, 1993, 2001). It is also pos-
sible that an increase in aerosols in the atmosphere from pollution may be
causing reduced rainfall by reducing surface solar radiation (Ramanathan
et al., 2001; Stanhill and Cohen, 2001; Roderick and Farquhar, 2002).
Lower surface solar radiation reduces the energy available for evaporation
leading to reduced rainfall and disruption of tropical circulation patterns
(Ramanathan et al., 2001).
The climate of semi-arid regions and its variability has a strong influ-
ence on the vegetation found in semi-arid regions. Vegetation distribu-
tion, diversity and productivity are all influenced in the first order by
climate, but also by soils, competition, fire and grazing (O’Brien, 1993;
HilleRisLambers et al., 2001). Two large phytogeographical regions are
recognized to cover most of the semi-arid region of Tanzania, the Somalia-
Maasai region and the Zambezian region as described by White (1983)
(see Figure 9.1). The Somalia-Maasai region is the driest vegetation type
found predominantly in north and central Tanzania and is comprised of
a range of grass-dominated savanna habitats. Savannas vary from com-
pletely grass-dominated grass savanna through bushland, thicket and
shrubland, to tree savannas and wooded savannas (Jacobs et al., 1999).
Trees and shrubs can comprise up to 50 per cent of species composition
but the canopy remains open (Trapnell and Langdale-Brown, 1972).
Included within the Somalia-Maasai region are the edaphic savanna
grasslands found on the Serengeti plains (see Figure 9.1). These grasslands
occur on soils derived from volcanic ash, which tend to be alkaline and
often have a hard calcareous pan created by the leaching of carbonates
through the soil by rainfall (White, 1983). The calcareous pan creates a
barrier to root growth, preventing the growth of woody vegetation, and
is also impermeable to water so that these areas can be prone to flooding
during the rainy season. Savannas are believed to have existed in East
Africa for 20 million years and were, in part, created by the development
of seasonal rainfall patterns that restricted the distribution of tropical
rainforests (Pratt et al., 1966). Today a mixture of climate, soils, natural
fires, grazing and anthropogenic influences maintain the open nature of
savanna ecosystems.
The Zambezian region consists of miombo woodland dominated by
trees from the genus Brachystegia (Kingdon, 1971; White, 1983; Lovett
et al., 2001). The region forms a crescent in the south of Tanzania around
the more arid Somalia-Maasai region (see Figure 9.1). It is less recognized
as a semi-arid ecosystem even though it can occur in areas with rainfall
as low as 500mm per year. Drier miombo woodland tends to be less
diverse while wetter miombo woodland vegetation has to be adapted to
periods of flooding and waterlogging followed by long dry seasons with
little available water and frequent fires. It has been argued that in some
areas miombo forests may be the result of long-term human disturbance
(Hamilton, 1982; White, 1983). Certainly, in areas with deeper soils,
300 0 300 600 Kilometres
EDAPHIC GRASSLAND
SOMALIA-MAASAI VEGETATION
ZAMBEZIAN VEGETATION
MAJOR LAKES
N
Source: White (1983).
Figure 9.1 Major vegetation zones in semi-arid Tanzania
rainfall parameters suggest that tropical dry forests could be supported

and so the presence of miombo woodland may be the result of past clear-
ance for cultivation (White, 1983). In Tanzania the Zambezian region
is currently less open to occupation and use by indigenous communities
because of tsetse fly, which is widespread and responsible for disease
transmission to both humans and livestock (Kingdon, 1971; Lovett et al.,
2001) but it is still used seasonally and often provides grazing and forage
for livestock in times of drought elsewhere.
While the phytogeographical regions are useful for describing vegetation
on a larger spatial scale they can overlook the importance of smaller-scale

heterogeneity in vegetation distribution. East Africa has been a region of
large geological upheaval with volcanoes, block faulting and erosion creat-
ing plains, mountains and rift valleys and a variety of soils (Griffiths, 1993)
so that vegetation composition can change rapidly over short distances
(Van Keulen and Seligman, 1992; Snyman and Fouche, 1993; Armitage,
1996). Water availability is also an important factor for vegetation distri-
bution in semi-arid regions. While rainfall is seasonal and variable in low-
lying areas, rainfall in the mountains is more consistent and provides an
important source of water (Quinn et al., 2001). As a result, gallery forests
along rivers and wetlands can be found in semi-arid regions and provide
important variations in vegetation as well as sources of water in the dry
season. All these factors of seasonality and geology have created a system
in which resources such as grasslands, forests and water vary both spa-
tially and temporally. Primary production tracks rainfall patterns so that
not all areas will be productive at the same time over the region.
The reliance on climate as the defining factor for semi-arid regions has
posed difficulties in producing an accurate definition. A review of govern-
mental and international organizations working in semi-arid regions has
revealed four main definitions of semi-arid regions. The UK Department
for International Development (DFID) refers to semi-arid zones in its
Renewable Natural Resources Research Strategy as one of six produc-
tion systems (ODA, 1994). In the DFID definition a semi-arid zone is one
with a mean monthly temperature above 18°C, where evapotranspiration
exceeds precipitation in one or more seasons and mean annual rainfall
is between 400mm and 1200mm. The term ‘semi-arid’ is also used in the
Food and Agriculture Organization (FAO) agroclimatological stratifica-
tion and defines an area with a growing period of between 75 and 119
days (Bourn and Blench, 1999). This does not correspond directly with the
DFID definition as the DFID rainfall parameters cover a growing season
of 75 to 180 days (Bourn and Blench, 1999). The FAO agroclimatological
definition is also more restrictive than the arid and semi-arid tropics and
subtropics definition that the FAO uses in its World Livestock Productions
Systems (growing season less than 180 days) (Sere et al., 1995). In addi-
tion a fourth definition is used by UNEP (United Nations Environment
Programme) in its studies of desertification (Mortimore, 1998). In many
empirical studies the semi-arid region has been defined by a minimum
rainfall of 400–500mm a year to exclude arid regions, which come with
their own set of particular management issues related to extremely low
rainfall and low population densities (Pratt and Gwynne, 1977; Mwalyosi,
1995; Mortimore, 1998). Above 400–500mm of rainfall a year rain-fed
agriculture, although marginal, is possible and livestock and people are
found in much higher densities than in arid regions (Mortimore, 1998).

The variation in definition means that areas are included or excluded from
the semi-arid zone depending on the definition used and the type of vegeta-
tion associated with semi-arid areas.
In Tanzania, a restricted definition based on the FAO agroclimatologi-
cal stratification, or an upper annual rainfall limit of 900mm, would limit
the area considered semi-arid to Somalia-Maasai vegetation, which forms
a strip of arid land running from the north to a tip at the junction of the
Tanganyika, Nyasa and Eastern Arc rifts (Figure 9.2). The DFID defini-
tion of semi-arid regions, on the other hand, where the mean monthly tem-
perature is above 18°C, evapotranspiration exceeds precipitation in one or
more seasons and mean annual rainfall is between 400mm and 1200mm,
covers a more extensive area of land. This definition covers an area that
includes the Zambezian vegetation zone as well as the Somalia-Maasai
vegetation zone (see Figure 9.3). For Tanzania the semi-arid region in this
context covers approximately 80 per cent of the land area of the country,
an area of approximately 704 800km2.
The role of people and livelihoods

In the same way that climate and resources are variable in semi-arid
regions, so are the people who live and maintain their livelihoods in them.
Mortimore (1998, p. 10) refers to semi-arid regions as the ‘heart of the dry-
lands’ in Africa. They are extremely important for people, livestock and
agriculture. In order to understand semi-arid regions it is not enough to
understand the ecology in isolation. It is also necessary to understand the
communities who live in these regions. These communities have had, and
continue to have, an impact on the ecology of where they live. Equally,
the ecological conditions in which these people live have impacted, and
continue to impact, on them. It is this that constitutes the reciprocal
relationship between people and their environment in semi-arid regions.
This relationship has not always been recognized and as a result attempts
at sociopolitical and ecological control, first by the colonial governments
and later by the independent socialist government, have often led to a
decoupling of people and the environment.
Many of the ecological conditions observed today are the result of past
people–environment interactions and the subsequent decoupling of the
people–environment relationship. For example, savanna ecosystems may
be the result of long-term interactions between humankind and the envi-
ronment while the distribution of tsetse fly may be the result of depopula-
tion and subsequent woodland encroachment in areas once inhabited and
cleared for livestock or cultivation (Iliffe, 1979).
East Africa is recognized as important to human history. Archaeological
Note: Rainfall upper limit of 900mm per annum, equivalent to the FAO
agroclimatological classification of semi-arid regions; growing period 75–119 days. (Seasons
are defined on the basis of climatic fluctuations in Tanzania and take into account the long
rains and short rains.)
Source: Quinn (2005).
Figure 9.2 The extent of semi-arid regions in Tanzania based on a

restricted definition
finds in Kenya and especially Tanzania suggest that modern humans

originated in this region (Leakey, 1972). However, it is not a simple task to
decipher the origins and history of current ethnic groups in Tanzania. This
is because membership of and division into tribes was not and had never
been as clear-cut as often believed. Society in East Africa was pluralist
(Shorter, 1974), meaning that society was made up of heterogeneous and
often fragmented groups of people, reflecting the heterogeneous nature
of the environment. However, this did not mean that there were static
Note: Mean monthly temperature above 18°C; evapotranspiration exceeds precipitation

in one or more seasons; mean annual rainfall between 400mm and 1200mm. (Seasons are
defined on the basis of climatic fluctuations in Tanzania and take into account the long
rains and short rains.)
Figure 9.3 The extent of semi-arid regions in Tanzania based on the

DFID definition
and discrete tribes, rather that tribal membership was fluid and the result
of adaptation to the landscape (Iliffe, 1979). Tribal membership also did
not mean isolation from other groups. Many shared a common heritage
and continued to interact through trade, intermarriage and migration
(Homewood and Rodgers, 1991).
Kieran (1972), Shorter (1974) and Iliffe (1979) have attempted to discuss
in detail the origins and history of people in East Africa. The historical
development of societies in Tanzania will be discussed here based on their
influence on the people–environment relationship. The broad divisions

used are based on linguistic origins. The first inhabitants of the region
are believed to be Khoisan in origin and were hunter-gatherers. A low,
dispersed and nomadic population using traditional low technology tools
such as bows and arrows meant that this livelihood strategy was sustain-
able in terms of land and resources, even in times of drought (Armitage,
1996). It is unlikely that hunter-gatherers changed the ecology of their
surroundings to any great extent because of their low population and
primitive technology. Hunter-gatherer livelihoods rely on exploiting wild
animal and plant resources already present rather than domestication or
cultivation of animals or crops. Once widespread in the region they were
displaced or absorbed by other tribes who began to colonize the area from
North and West Africa. Only small groups such as the Hadza, who are
believed to be the descendants of these early inhabitants, still practise a
hunter-gatherer livelihood strategy in Tanzania today (Armitage, 1996).
From about 1000 BC and possibly as early as 2000 BC Cushitic tribes
migrated south into Tanzania from a region now in modern day Ethiopia.
They were generally cattle herders and occupied the northern central
region of Tanzania dominated by Somalia-Maasai vegetation, which was
settled because of its suitability for pastoral livelihood strategies. These
early pastoralists are still represented by the Iraqw in Tanzania. Possibly
one of the largest migrations, and certainly one of the most widespread
and influential linguistically in Tanzania, was the arrival of the Bantu-
speaking tribes from West Africa. Bantu tribes are thought to have started
migrating into the lake regions of Tanzania as early as 500 BC and after
AD1000 approximately 90 per cent of the tribes in Tanzania were of
Bantu origin. The success of the Bantu tribes is attributed to their use of
cultivation, which allowed them to successfully colonize the more fertile
highlands before spreading down onto the plains (Shorter, 1974). After
the Bantu tribes the fourth major group to migrate into the region were
the pastoral Nilotic tribes from Eastern Ethiopia who arrived much later
in the 1500s. The Maasai were part of this migration and had settled in
Northern Tanzania by the late 1700s.
Both pastoral and agricultural livelihood strategies had a much greater
impact on the ecology of Tanzania than the hunter-gatherer strategies of
the original inhabitants. Pastoralism was based on transhumance rather
than being truly nomadic. This meant that the homestead, comprised
of women and very young children, remained in one place for relatively
long periods of time. Lactating cows and calves were kept in, and grazed
around, the homestead, so often homesteads were widely spaced to allow
adequate forage. Young men and boys moved seasonally with the rest
of the herd to find adequate grazing throughout the year. Areas were
designated and used for dry or wet season grazing, or only used in years
of drought. Fire was often used to clear woody vegetation and encourage
grass growth in grazing areas, reducing woody biomass and maintaining
a more open savanna system. Agricultural groups also used fire to clear
areas for growing crops, and brought in new plant species such as banana,
sorghum, millet and maize, which they grew in different areas depending
on soil quality and rainfall. As a result, settlement patterns depended on
soil quality and water availability. Only in the highlands where soils were
generally more fertile, rainfall higher and competition for land greater did
communities settle for long periods and invest in their fields. In semi-arid
regions agricultural communities moved regularly to clear new areas for
cultivation in response to declining soil fertility or inadequate rainfall. It is
likely that both pastoral and agricultural groups were responsible for clear-
ing woody vegetation in many areas, creating open savanna or secondary
miombo woodlands and reducing dry tropical forest cover (White, 1983).
Today some 110 tribes are recognized in Tanzania (Gulliver, 1972).
Although they can be grouped according to their linguistic and historical
traditions they show divergent patterns in livelihood systems and social
organization (Gulliver, 1972). Tribal identification has been fluid and
was often related to location; many tribal names can be translated to
mean ‘highlanders’, ‘northerners’ or ‘from the waterless country’ (Iliffe,
1979). Therefore, it was possible to change tribal affiliation by moving
and changing livelihood strategies. As a result, linguistic origin is not
always useful in delineating livelihood strategies. Although Cushitic and
Nilotic tribes were generally pastoralists and Bantu, cultivators, in fact
many tribes adapted their livelihood strategies according to the ecology
and environment in which they lived. The diversity of soils and variability
in climate that created ecological heterogeneity in semi-arid regions were
also responsible for the heterogeneity found in tribal identity and liveli-
hoods. Both Cushitic and Bantu tribes cultivated crops as well as owning
livestock and the relative importance of these strategies varied with loca-
tion and especially rainfall. Tribal identity was inextricably linked to the
local ecology.
Sociopolitical change has had a huge impact on livelihoods and ecology
in Tanzania. The influx of pastoral and agricultural groups changed the
way that people interacted with the environment and as a result the ecology
of semi-arid regions. In the past there was a close link between livelihoods
and ecological conditions. More recently, however, a process of decou-
pling of people and environment has occurred. This process began with
the spread of livestock and human diseases into East Africa in the early
1800s. The spread of rinderpest is recognized as being particularly impor-
tant in its impact on people and environment and the relationship between
them (Kjekshus, 1996). Rinderpest was introduced into Northern Africa

by Europeans and spread into East Africa, decimating cattle populations.
The Maasai were particularly hard hit. It is estimated that approximately
90 per cent of their cattle were killed and two-thirds of the Maasai popula-
tion died from famine and disease as a result (Iliffe, 1979; Homewood and
Rogers 1991; Kjekshus, 1996; Koponen, 1996). The spread of rinderpest
was also accompanied by epidemics of smallpox and influenza, among
other human diseases, which served to reduce the numbers of people still
further. Many areas that had been used for cultivation or livestock were
deserted as people died or left in search of food. This depopulation allowed
the expansion of miombo woodland and wildlife numbers increased. The
distribution of the tsetse fly, which carries sleeping sickness, also expanded
as its woodland habitat increased (Coulson, 1976). The spread of tsetse fly
made once inhabited areas no longer suitable for settlement.
This process of decoupling continued into the colonial era where socio-
political change created by war and conflict contributed to depopulation.
German colonial rule was characterized by tribal war and resistance. For
example, during Maji Maji1 and its aftermath, the largest and most wide-
spread resistance to the colonial government, it is estimated that some
250 000 people were killed (Iliffe, 1979). As a result of depopulation large
areas of Tanzania became uninhabited and were appropriated by the state
for plantations or wildlife reserves. Many areas that had been used for
livestock grazing or cultivation were now no longer available to local pop-
ulations. Further reductions in population occurred during the two World
Wars when many local people were forced into labour for both the British
and German armies, and food appropriated for the armies contributed to
widespread famine (Brooke, 1967).
Even during the relatively peaceful British colonial rule following World
War II the decoupling of the people–environment relationship continued.
People were moved and concentrated into villages in order to facilitate
greater socioeconomic control. The process of indirect rule through
tribal chiefs advocated by the British also led to social upheaval as power
through tribal leadership was contested. The process of land appropria-
tion for wildlife reserves and plantations continued as the British govern-
ment sought to control the resources of Tanzania as well as its people. All
these factors together meant that the relationship between communities
and their environment was disrupted. For example, the area now covered
by the Selous Game Reserve was once densely populated. Depopulation
due to disease and war led to widespread woodland encroachment and
expansion of tsetse fly. The population who were left were scattered in
small, mobile communities. People were concentrated into villages so that
the district government could have greater socioeconomic control and the
Selous Game Reserve was created in order to prevent people from moving
back into the tsetse-infested area (Matzke, 1976). The creation of the game
reserve also meant that the land was now in the control of the state rather
than outside it. Only later did the remit of the reserve change to include
conservation aims.
Even with independence the process of social control over rural commu-
nities has continued, with its resultant impact on the environment. Ujamaa
villages created by the process of villagization were originally conceived as
cooperative villages where people would work together to increase agri-
cultural production (Hyden, 1975; McHenry, 1979; Kikula, 1997). These
villages would also facilitate the delivery of social and agricultural services
to often remote and dispersed communities. However, the independent
government, along with the colonial governments who preceded it, failed
to recognize the influence of ecology on patterns of settlement. Variable
rainfall, soils, wildlife and disease all determined settlement patterns,
livelihood choices and population distributions. In particular, rainfall
distribution meant that 60 per cent of the population lived in 20 per cent
of the land area of Tanzania (Maro and Mlay, 1978). Villagization settled
agricultural groups in areas marginal for farming and so contributed to
the conflict that already existed between pastoral and agricultural ethnic
groups, as these villages were often located on or near to important water
sources. Pastoralism was perceived to be a more primitive livelihood strat-
egy and so pastoral communities were encouraged to settle into villages,
regardless of the ecological need to move livestock. Although villagization
was planned as a way to increase production it was a failure and led to
declines in food and cash crop production (Smith, 2001).
To summarize, the ecology of semi-arid regions is characterized by both
inter-annual and intra-annual variability in rainfall. Large volumes of
rain can fall over short periods, while single and multi-year droughts are
common. This variability in climate, along with variability in geology and
soils, results in spatial and temporal heterogeneity in vegetation composi-
tion, primary production and water availability. Semi-arid regions are
therefore characterized by variability. It is now more widely recognized,
however, that the ecological conditions presently found in semi-arid
regions are, at least in part, the result of long-term human activity in these
areas. At the same time, the climatic variability in semi-arid regions is a
huge challenge for people in trying to sustain their livelihoods. As a result,
in semi-arid Tanzania both pastoral and agricultural livelihoods strate-
gies are found and each has developed flexibility in order to cope with
variability. Traditional pastoral livelihoods common to the region allow
for mobility of people and livestock so that adequate forage can be found
throughout the year, in much the same way as wild ungulate populations
migrate seasonally throughout East Africa. Pastoral livelihoods need

access to a mosaic of vegetation types in order to provide adequate grazing
for their livestock and reduce their vulnerability to variability in climate.
In more agricultural livelihoods diversity in location with access to key
water resources, crop selection and combinations of both agricultural and
pastoral livelihood strategies have been employed as methods for dealing
with ecological variability.
This reciprocal relationship between people and the environment has
historically been central to the sustainable management of semi-arid
Tanzania. Over the last two centuries, however, long-term sociopolitical
change has resulted in a decoupling of the people–environment relation-
ship. Depopulation due to disease and war left large areas uninhabited
and allowed the spread of woodland and tsetse fly. Depopulation also
allowed successive governments to appropriate land for the state through
plantations or wildlife reserves, removing land from local control and
denying access to key resources. Villagization also encouraged settlement
of agricultural groups in traditionally pastoral areas, adding to conflict
over important resources such as water. If agricultural groups take over
crucial water resources then pastoral communities are denied access and
can become more vulnerable, especially during drought years. Semi-arid
regions therefore represent a zone of conflict between different livelihood
strategies over access to resources, a conflict often exacerbated by changes
in land tenure and appropriation by the state that often results in vital
resources becoming inaccessible to local communities.
Despite wider recognition in contemporary times of the importance of
this reciprocal link between people and their environment, government
policy in semi-arid regions is still broadly influenced by past colonial
governments and protectionist and preservationist approaches, which
perceive traditional livelihood activities as damaging to the environment.
In the next section we move on to consider the dominant paradigms
that conceptualize thinking around ecological and social problems in
semi-arid regions and demonstrate how the nature of these paradigms
defines whether or not resulting policy discourses recognize the reciprocal
relationship between people and the environment.
Managing semi-arid regions: ecological and social paradigms

The idea of a ‘paradigm’ refers to the way in which a certain issue is
perceived or understood. The paradigms that define policy-makers’ and
managers’ understanding of the ecological and social dynamics of semi-
arid regions will also define their perception of appropriate policy or
management solutions. In Tanzania, two opposing discourses dominate
policy approaches to managing semi-arid regions. One is the ‘people
versus environment’ discourse where local people are seen as damaging

to the environment. This discourse implies a need to manage people and
the environment separately. The other is the ‘people and environment’
discourse that seeks to encourage integrated management of people and
resources by recognizing the importance of the reciprocal relationship
between people and their environment. These policy discourses are under-
pinned by two different ecological and social paradigms upon which is
defined the conceptualization of management problems.
Ecological paradigms
Until relatively recently the predominant paradigm for understanding
the ecology of semi-arid regions, and rangelands in particular, has been
equilibrium theory. This paradigm developed from a European and
US perspective of rangelands in temperate climates (Clements, 1916).
Equilibrium theory is based on the assumption that vegetation composi-
tion will progress towards a climax community, depending on climate and
soil conditions, through the processes of succession (Begon et al., 1990;
Behnke and Scoones, 1993). The climax community is therefore consid-
ered to be the most stable plant community that could occur. Disturbance,
either natural or anthropogenic, will result in the vegetation being ‘pushed
back’ to an earlier successional stage and once the disturbance ceases or
is removed then natural processes will lead back to the climax condition
(Briske et al., 2003). The most commonly used example of succession is
the clearing of forest for agriculture in temperate regions. Once the land
is abandoned then it is assumed that a climax forest community will
eventually return (Behnke and Scoones, 1993). These ideas were applied
to grazed rangelands in sub-Saharan Africa, as well as elsewhere, where
different grass communities were considered to represent different suc-
cessional stages with certain ecological communities representing climax
(Brown and MacLeod, 1996).
An important element of equilibrium theory as applied to grazed range-
lands is the concept of carrying capacity. If, as suggested by equilibrium
theory, secondary production is linearly related to the successional status
of vegetation then it is possible to calculate carrying capacity in terms of
wild herbivores or livestock that can be supported on the range. In equi-
librium theory the interaction between plants and animals is an important
negative feedback mechanism that regulates the ecosystem (Briske et al.,
2003). This density dependence relationship means that at ecological
carrying capacity herbivore numbers are controlled by primary produc-
tion so that the number of births equals the numbers of deaths and the
vegetation is held back from reaching the climax composition that would
occur without such disturbance from grazing. For managers of rangelands
the aim would be to maintain the vegetation at some optimal sub-climax

composition so as to maximize secondary production, depending on the
management objectives. What is important to note is that carrying capac-
ity in this context is dependent on the management objectives and is not
an inherent feature of the ecosystem. Different carrying capacities will
be desired, for example, for commercial meat production or subsistence
pastoralism.
Equilibrium theory as applied to sub-Saharan Africa has been criticized
for not adequately describing the reality of rangelands and recently this
criticism has led to the development of an alternative paradigm for range
ecology: the non-equilibrium theory (Behnke and Scoones, 1993; Warren,
1995; Brown and MacLeod, 1996; Briske et al., 2003). This paradigm sug-
gests that there is no single successional pathway and that rangeland can
exist in multiple stable states. Rangelands are then seen as mosaics of plant
communities created by a variety of factors rather than evidence of early
succession stages caused by disturbance (Sullivan and Rohde, 2002). What
is important for non-equilibrium theory is that density-dependent inter-
action between plants and animals is weak or non-existent and instead
abiotic factors, and in particular rainfall, are the dominant driving factors
of vegetation production, distribution and composition. This is an impor-
tant distinction from equilibrium theory because it recognizes that the
highly variable climate and rainfall patterns are more important in deter-
mining primary production and composition than grazing by herbivores.
In non-equilibrium theory fluctuations in rainfall create fluctuations in
vegetation productivity, which in turn cause herbivore numbers to fluctu-
ate. Rainfall is such an important driver of the system that vegetation and
herbivores do not reach equilibrium and so it is not possible to calculate a
carrying capacity.
There is still controversy over whether equilibrium or non-equilibrium
paradigms apply to African rangelands. Evidence has been presented that
supports both views (Illius and O’Connor, 1999; Sullivan and Rohde,
2002). Illius and O’Connor (1999) in particular present the view that there
are still density-dependent interactions between herbivores and vegetation
in key resource areas. Therefore carrying capacity is maintained for these
areas even when it does not correspond to the whole range. In contrast
Sullivan and Rohde (2002) argue that even key resource areas are subject to
highly dynamic rainfall patterns so that primary production is more related
to rainfall than to grazing. It has been suggested that rather than focusing
on the dichotomy the theories should be considered as positions along a
continuum where rangeland can exhibit equilibrium or non-equilibrium
dynamics, or a combination of both (Briske et al., 2003). Empirical
studies are therefore necessary to determine where along this continuum
a particular rangeland lies. Briske et al. (2003) argue that in drier regions
where the intra-annual variation in rainfall is greatest then rangelands are
likely to exhibit non-equilibrium dynamics, however in wetter areas where
variation in rainfall is less then equilibrium theory is more appropriate.
Following this argument in Tanzania would lead to the Somalia-Maasai
region being considered a non-equilibrium system since rainfall is highly
variable. In contrast the Zambezian region would be subject to equilibrium
dynamics since rainfall averages are higher and more consistent. However,
even in the Zambezian region rainfall is highly seasonal and intra-annual
variation is still high, although total rainfall is generally greater than in the
Somalia-Maasai region. As a result non-equilibrium dynamics cannot be
discounted as important for at least some parts of this system.
Social paradigms
The two different ecological paradigms outlined above are central in
determining which social paradigm influences policy, because each focuses
on a different cause of environmental degradation. In equilibrium theory
overstocking and overpopulation are seen as the primary causes of deg-
radation of rangelands and management regimes should seek to control
stocking rates and change land tenure so as to maintain rangelands in
an optimal condition. Equilibrium theory leads to the pastoral paradigm
being the dominant theory that frames understanding of communities.
In the pastoral paradigm pastoral communities are seen as unable to
adequately manage rangeland resources, leading to degradation (Warren,
1995). Population increases accompanied by inappropriate management
are therefore considered to be key problems that require intervention in
order to prevent over-use and protect natural resources and wildlife. In
Tanzania the predominance of the equilibrium paradigm in understand-
ing rangeland ecology has helped support the view that pastoral ways of
life are primitive and economically unproductive (Armitage, 1996). Early
successional stages in rangeland composition are given as evidence that
pastoral communities overstock the rangelands leading to degradation.
Interventions that de-stock and change land tenure are therefore seen as
necessary to prevent degradation and ultimately soil erosion and desertifi-
cation (Ellis and Swift, 1988; Warren, 1995).
In contrast, abiotic factors, particularly rainfall, are attributed the key
role in determining range condition in non-equilibrium theory. As a result,
managing stocking rates would not necessarily lead to improvements in
range condition. In addition, it is now recognized that most development
efforts based on equilibrium theory and the pastoral paradigm have ulti-
mately ended in failure (Oba et al., 2000). While it could be argued that
failures have been due to resistance of pastoral communities to accept
these management changes, there is a mounting body of evidence that

suggests that pastoral forms of management do not always lead to deg-
radation and that they are stocking the rangelands at optimum levels for
their management objectives and ecological conditions (Homewood and
Rodgers, 1987; Behnke and Scoones, 1993; Behnke and Abel, 1996; Oba
et al., 2003). This shift in ecological thinking has therefore led to a shift
towards a new social paradigm that acknowledges indigenous knowledge
as important for resource management. The emphasis is on community
participation in development projects and recognition that supporting and
adapting existing traditional management may provide a more successful
long-term route for aid intervention.
Policy discourses
The choice between the ecological and social paradigms outlined above,
together with the long-term sociopolitical changes that have characterized
Tanzania’s history over the last few hundred years, have been central in
defining contemporary policy discourses. Colonial rule in Tanzania has
had a profound effect on the subsequent management of resources in
these areas and the way in which local communities are perceived. When
the British and other European nations began to colonize large parts of
East Africa in the late 1800s the received wisdom was that Africa was one
of the last great wildernesses untouched by humankind (Anderson and
Grove, 1987; Leach and Mearns, 1996; Neumann, 1996). National Parks
were created by the British colonial administration in part to protect this
wilderness, but they were also created to preserve the aristocratic lifestyle
that was diminishing in the United Kingdom as the result of social and
political changes in land ownership (Neumann, 1995, 1996). In reality, the
first National Parks were less to do with conserving wildlife as it is under-
stood today and more about protecting hunting rights and a particular
way of life, and exercising social control through exclusion and settlement
of indigenous communities (Anderson and Grove, 1987; Neumann, 1995).
These early ideas of separating people and nature became entrenched in
the colonial administration and the independent administrations that fol-
lowed. This ‘people versus environment’ discourse in policy is inextricably
linked with conservation thinking, equilibrium theory and the pastoral
paradigm. Today wildlife is perceived as a common heritage that should be
conserved for its own sake and for the sake of future generations (Lovett
et al., 2001). Maintaining biodiversity through the exclusion of livestock,
which under the equilibrium paradigm is believed to out-compete native
ungulates for forage, is seen as a necessary part of conservation strategy.
Local communities are perceived to be at odds with sustainable man-
agement of resources and top-down interventions are advocated. This
discourse focuses on compensating local communities through wildlife

conservation programmes (Lovett et al., 2001).
The problem with the initial colonial view of ‘wilderness Africa’ was
that it represented a snapshot of conditions in East Africa at a time of
unprecedented crisis in indigenous populations. As discussed earlier in
this chapter, East Africa was devastated by the rinderpest epidemic of
the 1890s. It rapidly swept through East Africa killing huge numbers of
cattle and almost two-thirds of the Maasai population alone died because
of famine and disease (Homewood and Rodgers, 1991; Kjekshus, 1996).
Oscar Baumann (cited in Iliffe, 1979) described the devastation experi-
enced by the Maasai:
There were women wasted to skeletons from whose eyes the madness of star-
vation glared. . .‘warriors’ scarcely able to crawl on all fours, and apathetic,
languishing elders. These people ate anything. Dead donkeys were a feast for
them, but they did not disdain bones, hides, and even the horns of the cattle.
(Iliffe, 1979, p. 124)
The decimation of once powerful tribes resulted in destabilization of

social structures and without cattle people moved out of certain areas in
order to find food. When the German administration arrived in the late
1890s, followed by the British after World War II, they found large areas
unsettled by people and communities unwilling to trade, intent on raiding
other tribes for food. From this they concluded that the area had never
been highly populated and that the people were primitive and economically
backward (Kjekshus, 1996). In reality many of the ecological conditions
found at that time had been created by human settlement. In this situation
of crisis large areas were easily alienated from indigenous communities
to create national parks, and the subsequent recovery in populations was
seen as evidence of overpopulation and the need for intervention.
Re-examination of historical changes in ecological and sociopolitical
conditions has created a shift in thinking about the role of people in East
Africa (Leach and Mearns, 1996). While in his independence message to
TANU (Tanzania African National Union) Nyerere said ‘from now on we
are fighting not man but nature’ (Nyerere, 1969, p. 139), as Iliffe points out
‘it [is] more complicated than that’ (1979, p. 576). The ‘people and envi-
ronment’ policy discourse has grown from the belief that indigenous com-
munities have been subject to inappropriate Western ideals of ecology and
management, dominated by the equilibrium paradigm, which has led to
land being appropriated for national parks and agriculture and, as a result,
the destabilization of traditional forms of management. Non-equilibrium
theory has highlighted the dynamic nature of production and resource
distribution in semi-arid regions and suggests that current concerns
about overstocking and degradation may be misplaced (Homewood and

Rodgers, 1987; Stocking, 1996). While it is not suggested that degrada-
tion never occurs (Homewood, 1995) more emphasis is placed on rainfall
dynamics and the effects of restricting access to rangelands through aliena-
tion for national parks and the expansion of agriculture into traditional
pasture lands. The emphasis in this discourse is on equity and the rights
of indigenous populations (Lovett et al., 2001). This discourse challenges
the pastoral paradigm in which local communities are seen as incapable of
managing resources and Western people are viewed as knowing best what
needs to be done to make better the lives and environments of communi-
ties in developing countries (Leach and Mearns, 1996; Stott and Sullivan,
2000). Instead, indigenous knowledge is recognized and local communities
are seen as vital for the sustainable management of resources. In this dis-
course the emphasis is on community participation and community-based
initiatives for wildlife and natural resource management.
Nevertheless the ‘people versus environment’ policy discourse has
historically dominated natural resource management in Tanzania.
Approximately 25 per cent of Tanzania has been gazetted as national
parks, game reserves or forest reserves (IUCN, 1987) with semi-arid
regions being particularly affected. This has been accompanied by con-
cerns with overpopulation and overgrazing in areas still used by pastoral
and agro-pastoral communities. A prime example is the HADO (Hifadhi
Ardhi Dodoma – soil conservation in Dodoma) project in Kondoa, which
was set up in 1973 in response to the perceived overstocking of rangelands
in the area and resultant soil erosion (Lovett et al., 2001; Quinn et al.,
2007). The HADO project instigated complete de-stocking of the range-
lands accompanied by improved agriculture and tree planting (Wenner,
1983) and later the introduction of zero-grazing where small livestock
are kept and fed on crop residues (Kerario, 1996). Although the HADO
project has been considered a success in improving vegetation cover there
is the suggestion that soil erosion in the area was the result of deforesta-
tion rather than population increase (Mung’ong’o, 1991). Ultimately the
HADO project could only be instigated through direct government
intervention and control and even then grazing of livestock still occurs
(personal observation).
More recent interventions have considered community participation
in natural resource management but there is concern that these only pay
lip service to local community knowledge (Leach and Mearns, 1996) and
criticism can be found of community-based conservation projects else-
where that have failed to protect wildlife or fully integrate local institu-
tions (Campbell et al., 2001). It seems unlikely that there is the political
will to open up national parks to pastoralism and governments still do
Table 9.1 Two conceptual frameworks for resource management in semi-

arid regions
Ecological Perceived Social Policy Policy Social

Paradigm Causes of Paradigm Discourse Response Relationship
Degradation to
Environment
Equilibrium Human- Pastoral People Top-down Decoupled
theory induced paradigm versus interventions from the
environment environment
Overstocking, De-stocking, Unable to
population changes in manage
increase land tenure resources
Non- Natural New social People and Bottom-up Coupled

equilibrium paradigm environment approaches to the
theory environment
Abiotic Indigenous Able to
factors knowledge, manage
community resources
participation sustainably
not trust the ability of local institutions to manage resources sustainably.

Those who support the ‘people and environment’ discourse are often
discredited for believing in the idea that there was a peaceful co-existence
between people and wildlife before colonialism (Giblin and Maddox,
1996; Kjekshus, 1996). This is not necessarily accepted as the case. The
point instead is that interventions so far have destabilized traditional
forms of management that were appropriate for the ecological context of
semi-arid regions (Oba et al., 2000). New policy responses therefore need
to take a more ‘bottom-up’ approach to integrate indigenous knowledge
and community participation into natural resource management projects
(see, for example, Ockwell, 2008).
The problem facing Tanzania, and other countries in sub-Saharan
Africa, is that there are, in effect, two alternative conceptual frameworks
for understanding and managing natural resources in semi-arid regions
(Table 9.1). The first conceptual framework is based on equilibrium
theory. Equilibrium theory is still widely supported by both researchers
and particularly government, resulting in human-induced environmen-
tal degradation being high on the agenda and interventions often being
designed to reduce the human impact on the environment (Kikula, 1999).
As a result, the pastoral paradigm still persists where local communities

are interpreted as primitive and in need of modernization through inter-
vention (Smith, 2001). This approach leads to the ‘people versus environ-
ment’ discourse and so to interventions that tend to be ‘top-down’ and set
people and the environment at odds with each other, continuing the loss of
control over the environment experienced by local communities since the
late nineteenth century (Maddox, 1996).
In contrast, the conceptual framework based on non-equilibrium theory
recognizes that local communities may not be responsible for environmen-
tal degradation and that abiotic factors are more important for determin-
ing range condition (Sullivan and Rohde, 2002). As a result the new social
paradigm recognizes that traditional forms of management can be suitable
for sustainably managing natural resources found in semi-arid regions.
This then leads to a policy discourse that seeks to recouple people to their
environment by taking ‘bottom-up’ approaches to resource management
that make use of indigenous knowledge and involves communities in the
development and running of management projects.
The ecological and social paradigms are still highly contested and as
a result the policy context is undergoing a period of flux and change.
Local communities are now challenging the authority of the government
to alienate land for conservation but it seems unlikely that in the present
political climate protected areas will be opened up to multiple land uses.
There is a high level of mistrust in local communities of policy interven-
tions that they see as not addressing local needs. However, this seems to be
matched by the government’s lack of trust in local institutions to manage
resources. In addition, the desire to protect areas for conservation and
tourism has not abated, especially since it provides much-needed foreign
capital for central government, if not for local communities. Alienation
of land for protected areas has contributed to a concentration of human
activities so that increased population densities are creating demands
for more land and creating conflicts over key resources such as water.
Government policy that favours agricultural livelihoods over pastoral
strategies has also contributed to increased resource conflict (Armitage,
1996). Uncertainty and change in the policy context of resource manage-
ment in semi-arid Tanzania adds to the ecological uncertainties under
which local communities must attempt to sustain their livelihoods.
Revisiting poverty and sustainable development

This chapter began by highlighting the centrality of poverty reduction in
achieving sustainable development. The reciprocal relationship between
ecology and people implies that ‘environmental management for sustain-
able development’ needs to protect the environment at the same time as
protecting and developing the livelihoods of those people who depend on

it. By focusing on approaches to environmental management in semi-arid
Tanzania, this chapter has illustrated how environmental policy discourses
are defined by the ecological and social paradigms that underpin them.
These different social and ecological paradigms have fundamental
implications in terms of resulting management approaches. In semi-arid
Tanzania, and East Africa more generally, there could not be a greater
difference between policy responses than between the traditional ‘people
versus environment’ policy discourse, which implies a need for de-stocking
and the exclusion of people from the land, and the ‘people and environ-
ment’ policy discourse that implies a need to engage indigenous knowledge
through community participation in environmental management.
In the past there has been a tendency for environmental managers to
focus solely on ecological theories or paradigms when formulating policy
approaches. This chapter has highlighted the fact that these ecological
paradigms are often accompanied by underlying social paradigms. If
environmental management is to be successful in achieving sustainable
development, including poverty reduction, it is therefore essential that
environmental managers are properly aware of the ecological and social
paradigms that define their thinking and that underpin the policy dis-
courses and accompanying management responses that they support.
In the absence of such reflexive, critical self-awareness, environmental
managers are exposed to the danger of making decisions based on hidden
assumptions that may or may not reflect the myriad of subjective ecologi-
cal and social realities that shape the ecosystem dynamics that they seek
to manage (Ockwell, 2008; Ockwell and Rydin, Chapter 6, this volume).
Such unreflexive decision-making is unlikely to be able to rise to the chal-
lenge of achieving development that ensures a sustainable future for both
the environment and the people whose livelihoods depend on it.
Note
1. Maji is Kiswahili for water. During the conflict water was distributed that was believed
to protect the drinkers from German weapons. Unfortunately, the water did not work
and many lost their lives as a result.
References
Anand, P. (2003), ‘Economic analysis and environmental responses’, in A. Blowers and
S. Hinchliffe (eds), Environmental Responses, Milton Keynes: Open University and
Chichester: John Wiley and Sons Ltd.
Anderson, D. and R. Grove (1987), ‘The scramble for Eden: Past, present and future in
African conservation’, in D. Anderson and R. Grove (eds), Conservation in Africa: People,
Policies and Practice, Cambridge, UK: Cambridge University Press, pp. 1–12.
Armitage, D.R. (1996), ‘Environmental management and policy in a dryland ecozone: The
Eyasi-Yaeda basin, Tanzania’, Ambio, 25(6), 396–402.
Begon, M., J.L. Harper and C.R. Townsend (1990), Ecology: Individuals, Populations and
Communities, 2nd edition, Oxford, UK: Blackwell Science.
Behnke, R. and N. Abel (1996), ‘Revisited: The overstocking controversy in semi-arid
Africa’, World Animal Review, 87(2), 4–27.
Behnke, R.H. and I. Scoones (1993), ‘Rethinking range ecology: Implications for rangeland
management in Africa’, in R.H. Behnke, I. Scoones and C. Kerven (eds), Range Ecology
at Disequilibrium: New Models of Natural Variability and Pastoral Adaptation in African
Savannas, London: Overseas Development Institute, pp. 1–30.
Bourn, D. and R. Blench (eds) (1999), Can Livestock and Wildlife Co-exist? An
Interdisciplinary Approach, London: Overseas Development Institute.
Briske, D.D., S.D. Fuhlendorf and F.E. Smeins (2003), ‘Vegetation dynamics on rangelands:
a critique of the current paradigms’, Journal of Applied Ecology, 40(4), 601–14.
Brooke, C. (1967), ‘The heritage of famine in Central Tanzania’, Tanzania Notes and
Records, No. 67, 51–9.
Brown, J.R. and N.D. MacLeod (1996), ‘Integrating ecology into natural resource manage-
ment policy’, Environmental Management, 20(3), 289–96.
Campbell, B.M., A. Mandondo, N. Nemarundwe, B. Sithole, W. De Jong, M. Luckert and
F. Matose (2001), ‘Challenges to proponents of Common Property Resource systems:
Despairing voices from the social forests of Zimbabwe’, World Development, 29(4),
589–600.
Carter, N. (2001), The Politics of the Environment: Ideas, Activism, Policy, New York:
Clements, F.E. (1916), Plant Succession: An Analysis of the Development of Vegetation,
Washington, DC: Carnegie Institute Publishers.
Coulson, A.C. (1976), ‘Crop priorities for the lowlands of Tanga region’, Tanzania Notes and
Records, Nos. 81–82, 43–53.
Ellis, J.E. and D.M. Swift (1988), ‘Stability of African pastoral ecosystems – alternate para-
digms and implications for development’, Journal of Range Management, 41(6), 450–59.
Giblin, J. and G. Maddox (1996), ‘Introduction’, in G. Maddox, J. Giblin and I.N.
Kimambo (eds), Custodians of the Land: Ecology and Culture in the History of Tanzania,
London: James Currey.
Griffiths, C. J. (1993), ‘The geological evolution of East Africa’, in J.C. Lovett and S.K.
Wasser (eds), Biogeography and Ecology of the Rainforests of Eastern Africa, Cambridge,
UK: Cambridge University Press.
Griffiths, J.F. (1972), ‘Climate’, in W.T.W. Morgan (ed.), East Africa: Its Peoples and
Resources, Nairobi: Oxford University Press, pp. 107–18.
Gulliver, P.H. (1972), ‘Peoples’, in W.T.W. Morgan (ed.), East Africa: Its Peoples and
Hamersley, A. (1972), ‘Agriculture and land tenure in Tanzania’, in W.T.W. Morgan (ed.),
East Africa: Its Peoples and Resources, Nairobi: Oxford University Press, pp. 189–98.
Hamilton, A.C. (1982), Environmental History of East Africa: A Study of the Quaternary,
London: Academic Press.
Hamilton, A.C. (1989), ‘The climate of the East Usambaras’, in A.C. Hamilton and
R. Bensted-Smith (eds), Forest Conservation in the East Usambara Mountains, Tanzania,
Cambridge, UK: IUCN, pp. 97–102.
HilleRisLambers, R., M. Rietkerk, F. van den Bosch, H.H.T. Prins and H. de Kroon (2001),
‘Vegetation pattern formation in semi-arid grazing systems’, Ecology, 82(11), 50–61.
Homewood, K. (1995), ‘Development, demarcation and ecological outcomes in
Maasailand’, Africa, 65(3), 331–50.
Homewood, K. and W.A. Rodgers (1987), ‘Pastoralism, conservation and the overgrazing
controversy’, in D. Anderson and R. Grove (eds), Conservation in Africa: People, Policies
and Practice, Cambridge, UK: Cambridge University Press, pp. 111–28.
Homewood, K. and W.A. Rodgers (1991), Maasailand Ecology: Pastoral Development and
Wildlife Conservation in Ngorongoro, Tanzania, Cambridge, UK: Cambridge University
Press.
Hyden, G. (1975), ‘Development of the cooperative movement in Tanzania’, Tanzania Notes

and Records, No. 76, 51–6.
Iliffe, J. (1979), A Modern History of Tanganyika, Cambridge, UK: Cambridge University
Press.
Illius, A. W. and T.G. O’Connor (1999), ‘On the relevance of non-equilibrium concepts to
arid and semi-arid grazing systems’, Ecological Applications, 9(3), 798–813.
IUCN (1987), IUCN Directory of Afrotropical Protected Areas, Gland, Switzerland and
Cambridge, UK: IUCN.
Jacobs, B.F., J.D. Kingston and L.L. Jacobs (1999), ‘The origin of grass-dominated ecosys-
tems’, Annals of Missouri Botanical Gardens, 86(2), 590–643.
Kerario, E. (1996), ‘A note on problems observed during the initial stages of the zero-grazing
project in the HADO project area of Kondoa’, in C. Christiansson and I.S. Kikula (eds),
Changing Environments, Research on Man–Land Interrelations in Semi-arid Tanzania,
Nairobi: Regional Soil Conservation Unit and Swedish International Development
Cooperation Agency, pp. 132–5.
Kieran, J.A. (1972), ‘History’, in W.T.W. Morgan (ed.), East Africa: Its Peoples and
Resources, Nairobi: Oxford University Press, pp. 21–34
Kikula, I. (1997), Policy Implications on Environment: The Case of Villagization in Tanzania,
Uppsala: The Nordic Institute and University of Dar es Salaam.
Kikula, I.S. (1999), ‘Lessons from twenty-five years of conservation and seven years of
research initiatives in the Kondoa Highlands of central Tanzania’, Ambio, 28(5), 444–9.
Kingdon, J. (1971), East African Mammals: An Atlas of Evolution in Africa, London:
Academic Press.
Kjekshus, H. (1996), Ecological Control and Economic Development in East African History:
The Case of Tanganyika 1850–1950, London: James Currey.
Koponen, J. (1996), ‘Population: A dependent variable’, in G. Maddox, J. Giblin and I.N.
Leach, M. and R. Mearns (1996), ‘Environmental change and policy’, in M. Leach and
R. Mearns (eds), The Lie of the Land: Challenging Received Wisdom on the African
Environment, Oxford, UK: The International African Institute in association with James
Currey.
Leakey, L.S.B. (1972), ‘Earlier prehistory’, in W.T.W. Morgan (ed.), East Africa: Its Peoples
and Resources, Nairobi: Oxford University Press, pp. 9–14.
Lovett, J.C. (1993), ‘Climate history and forest distribution in eastern Africa’, in J.C. Lovett
and S.K. Wasser (eds), Biogeography and Ecology of the Rainforests of Eastern Africa,
Lovett, J.C., C.H. Quinn, H. Kiwasila, S. Stevenson, N. Pallangyo and C. Muganga (2001),
Overview of Common Pool Resource Management in Semi-arid Tanzania, York: Centre for
Ecology, Law & Policy, University of York.
Lovett, J.C., S. Stevenson and H. Kiwasila (2002), Final Technical Report: Review of Common
Pool Resource Management in Semi-arid Tanzania, London: DFID.
Maddox, G. (1996), ‘Environment and population growth’, in G. Maddox, J. Giblin and I.N.
Maro, P.S. (1978), ‘Land: Our basic resource for socio-economic development’, Tanzania
Notes and Records, No. 83, 31–38.
Maro, P.S. and W.I.F. Mlay (1978), ‘People, population distribution and employment’,
Tanzania Notes and Records, No. 83, 1–20.
Matzke, G. (1976), ‘The development of the Selous Game Reserve’, Tanzania Notes and
Records, No. 79–80, 37–48.
McHenry, Jr, D.E. (1979), Tanzania’s Ujamaa Villages: The implementation of a rural devel-
opment strategy, Berkeley: Institute of International Studies, University of California.
Mortimore, M. (1998), Roots in the African Dust: Sustaining the Drylands, Cambridge, UK:
Mung’ong’o, C.G. (1991), Environmental Degradation and Underdevelopment in a Rural

Setting in Tanzania: The Case of Kondoa District, Dar es Salaam: Institute for Resource
Assessment, University of Dar es Salaam.
Mussa, D.T. (1978), ‘Our climate: Light, rainfall and temperature patterns’, Tanzania Notes
and Records, No. 83, 49–60.
Mwalyosi, R.B.B. (1995), ‘Agro-pastoralism and biodiversity conservation in East Africa:
The case of Maasailand, Tanzania’, in P. Halladay and D.A. Gilmour (eds), Conserving
Biodiversity outside Protected Areas: The Role of Traditional Agro-pastoralism, Gland,
Switzerland: IUCN.
Neumann, R. (1995), ‘Ways of seeing Africa: Colonial recasting of African society and land-
scape in Serengeti National Park’, Ecumene, 2(2), 149–69.
Neumann, R. (1996), ‘Dukes, earls and ersatz edens: Aristocratic nature preservationists in
colonial Africa’, Environment and Planning D: Society and Space, 14(1), 79–98.
Nicholson, S.E. (1993), ‘An overview of African rainfall fluctuations of the last decade’,
Journal of Climate, 6(7), 1463–6.
Nicholson, S.E. (2001), ‘Climatic and environmental change in Africa during the last two
centuries’, Climate Research, 17(2), 123–44.
Nyerere, J. (1969), ‘Independence message to TANU’, in Freedom and Unity: A Selection
from Writings and Speeches, 1952–65, Nairobi: Oxford University Press, pp. 138–9.
Oba, G., N.C. Stenseth and W.J. Lusigi (2000), ‘New perspectives on sustainable grazing
management in arid zones of sub-Saharan Africa’, Bioscience, 50(1), 35–51.
Oba, G., R.B. Weladji, W.J. Lusigi and N.C. Stenseth (2003), ‘Scale-dependent effects of
grazing on rangeland degradation in northern Kenya: A test of equilibrium and non-
equilibrium hypotheses’, Land Degradation & Development, 14(1), 83–94.
O’Brien, E.M. (1993), ‘Climatic gradients in woody plant species richness: Towards an expla-
nation based on an analysis of Southern Africa’s woody flora’, Journal of Biogeography,
20(2), 181–98.
O’Brien, E.M. (1998), ‘Water-energy dynamics, climate and prediction of woody plant
species richness: An interim general model’, Journal of Biogeography, 25(2), 379–98.
Ockwell, David G. (2008) ‘“Opening up” policy to reflexive appraisal: A role for Q
Methodology? A case study of fire management in Cape York, Australia’, Policy Sciences
41(4), 263–92.
ODA (1994), Renewable Natural Resources Research Strategy, London: Overseas
Development Administration.
Peberdy, J.R. (1972), ‘Rangeland’, in W.T.W. Morgan (ed.), East Africa: Its Peoples and
Pratt, D.J., P.J. Greenway and M.O. Gwynne (1966), ‘A classification of East African range-
land’, Journal of Applied Ecology, 3(2), 369–82.
Pratt, J. and H. Gwynne (1977), Rangeland Management and Ecology in East Africa, London:
Hodder & Stoughton.
Quinn, C.H. (2005), ‘Coping with ecological uncertainty in semi-arid Tanzania: Livelihoods,
risk and institutions’, PhD Thesis, University of York, UK.
Quinn, C.H., K. Forrester, H. Kiwasila, C. Muganga and N. Pallangyo (2001), ‘Fieldwork
report – village profiles’, York: Centre for Ecology, Law & Policy, University of
York.
Quinn, C.H., H. Kiwasila, M. Huby and J.C. Lovett (2007), ‘Design principles and Common
Pool Resource management: An institutional approach to evaluating community manage-
ment in semi-arid Tanzania’, Journal of Environmental Management, 84(1), 100–13.
Ramanathan, V., P.J. Crutzen, J.T. Kiehl and D. Rosenfeld (2001), ‘Aerosols, climate and
the hydrological cycle’, Science, 294(5549), 2119–24.
Roderick, M.L. and G.D. Farquhar (2002), ‘The cause of decreased pan evaporation over
the past 50 years’, Science, 298(5597), 1410–11.
Sere, C., H. Steinfeld and J. Groenewold (1995), World Livestock Production Systems:
Current Status, Issues and Trends, Rome: Food and Agriculture Organization of the
United Nations.
Shorter, A. (1974), East African Societies, London: Routledge and Kegan Paul.
Smith, C.D. (2001), Ecology, Civil Society and the Informal Economy in North West Tanzania,
Aldershot: Ashgate.
Snyman, H.A. and H.J. Fouche (1993), ‘Estimating seasonal herbage production of a semi-
arid grassland based on veld condition, rainfall and evapotranspiration’, African Journal
of Range & Forage Science, 10(1), 21–4.
Stanhill, G. and S. Cohen (2001), ‘Global dimming: A review of the evidence for a wide-
spread and significant reduction in global radiation with discussion of its probable causes
and possible agricultural consequences’, Agricultural and Forest Meteorology, 107(4),
255–78.
Stocking, M. (1996), ‘Soil erosion: Breaking new ground’, in M. Leach and R. Mearns (eds),
The Lie of the Land: Challenging Received Wisdom on the African Environment, Oxford,
UK: The International African Institute in association with James Currey, pp. 140–54.
Stott, P. and S. Sullivan (2000), ‘Introduction’, in P. Stott and S. Sullivan (eds), Political
Ecology: Science, Myth and Power, London: Arnold, pp. 1–15.
Sullivan, S. and R. Rohde (2002), ‘On non-equilibrium in arid and semi-arid grazing
systems’, Journal of Biogeography, 29(2), 1595–1618.
Trapnell, C.G. and I. Langdale-Brown (1972), ‘Natural vegetation’, in W.T.W. Morgan (ed.),
East Africa: Its Peoples and Resources, Nairobi: Oxford University Press, pp. 127–40.
Van Keulen, H. and N.G. Seligman (1992), ‘Moisture, nutrient availability and plant pro-
duction in the semi-arid region’, in T. Alberda, H. Van Keulen, N.G. Seligman and C.T.
de Wit (eds), Food from Dry Lands: An Integrated Approach to Planning of Agricultural
Development, Dordrecht: Kluwer Academic Publishers, pp. 25–81.
Walter, A. (1939), ‘A note on the seasonal rains in East Africa and their causation’,
Tanganyika Notes and Records, No. 8, 21–6.
Warren, A. (1995), ‘Changing understandings of African pastoralism and the nature of
environmental paradigms’, Transactions of the Institute of British Geographers, 20(2),
193–203.
WCED (1987), Our Common Future, Oxford, UK: Oxford University Press.
Wenner, C.G. (1983), ‘Soil conservation in Tanzania: The HADO project in Dodoma
Region’, a report on a visit in April–May 1983, Dar es Salaam, SIDA.
White, F. (1983), The Vegetation of Africa: A Descriptive Memoir to Accompany the
UNESCO/AETFAT/UNSO Vegetation Map of Africa, Paris: UNESCO.
10. Exploring game theory as a tool for
mapping strategic interactions in
common pool resource scenarios
Vanessa Pérez-Cirera
Introduction
The objective of this chapter is to introduce game theory as an analytical
tool for understanding and mapping strategic interactions amongst indi-
viduals and institutions in the management of common pool resources.
The chapter on the economics of common property resources explores
the relation between poverty and property rights in natural resource
management and emphasizes the role of transaction costs in the govern-
ance structure of common property systems and how these costs shape
the outcome of these systems (Adhikari, Chapter 5, this volume). The
chapter on the economic valuation of the different forms of land-use in
Tanzania emphasizes the importance of identifying and incorporating
non-marketed/non-priced values of environmental goods and services and
how such valuations can be undertaken so that optimal levels of land-use
are identified (Kirby, Chapter 11, this volume). A question that remains
open is if and how these optimal solutions can be reached. This section
intends to contribute towards this broad question by introducing game
theory as a useful analytical tool that helps us understand how decision-
making processes are made in the management of common pool resources.
The review explains how strategic decision-making can be mapped in
a game-theoretic fashion so that variables that are key for arriving at
socially optimal solutions can be identified.
The first section of the chapter gives an introduction to game theory as
a method for the construction of game-theoretic models, introducing the
reader to game-theoretic language and representation forms. The second
section will review the most frequently used games for depicting problems
encountered in CPR settings and illustrate the use of game theory in
analysing binding agreements as institutional solutions to CPR dilemmas.
The last section will aim to illustrate how game theory can be applied
to understanding decision-making processes and assessing the desir-
ability and viability of policy options with illustrations from semi-arid
Tanzania.
309
Game theory; language and representation forms

Game theory is a mathematical theory used to understand the outcomes
of strategic interactions in terms of quantifiable gains and losses from
different decisions, to mostly competing players, as if they were playing a
game. As a branch of applied mathematics, game theory started to be used
as a framework for analysis after the publication of Von Neumann and
Morgenstern’s (1944) Theory of Games and Economic Behavior. As such,
the main use of game theory was in economics through the early 1970s,
when its use started to spread to other sciences.1
There are two major strands within game theory: cooperative and non-
cooperative game theory. Cooperative game theory deals with situations
in which the players can negotiate before the game is played on what to
do in the game. It is assumed that these negotiations result in signing
a binding agreement (Binmore, 1994). For this type of game, what is
important is not so much the strategies available to the players, but the
preference structure of the game itself, since this is what determines which
contracts are feasible.
Non-cooperative game theory is based on a different set of principles.
Non-cooperative game theory calls for a complete description of the rules
of a game so that the strategies available to the players can be studied in
detail. The objective is to find a pair of equilibrium strategies to be des-
ignated as the solution of the game and this solution might or might not
be cooperative (Binmore, 1994). This strand of game theory gives space
to agreements. However, these are not conceived as necessarily binding.
Agreements can be made after or before a game is played and, depending
on how the payoffs and strategies available change, agreements can or
cannot be sustained.
At this stage, it is important to point out that game theory is not the
same as game models (Snidal, 1985). Game theory sets out an analytical
framework for the construction of game models, not for their design.
The results of the game will change depending on the way the model is
conceived. This means that one could construct a model and give differ-
ent features, for example, increase the number of players, and the result
would be different. Game theory thus can be conceived as a ‘metalan-
guage for [the construction of] game-theoretic models’ (Ostrom et al.,
1994, p. 24).
As most of the games used for depicting CPR problems emanate
from non-cooperative game theory, the next section intends to outline
some of the key concepts used in non-cooperative game theory to familiar-
ize the reader with game-theoretic language and forms of representation.
Exploring game theory 311
Rain (␤)
X1
Settle
N.1 Drought (1 –␤)

X2
Pastoralist.1
Rain (␤)
X3
Migrate
N.2 Drought (1–␤)

X4
Source: After Bardhan (1993).
Figure 10.1 Example of an extensive form game
Non-cooperative extensive form games

Non-cooperative games can be depicted in two forms: extensive and
strategic forms. In the extensive form, the rules of the game are laid out
in full detail by drawing a tree (Figure 10.1). To illustrate the way in
which extensive games are constructed, consider a simplified decision
problem faced by an agro-pastoralist. He or she can decide whether to
settle in an area of land, or to migrate. Her or his decision will depend on,
among other things, the rainfall patterns she or he expects to encounter
in different places. Consider then the following game tree (extensive form
game) with two players (player 1 and player 2). Player 1 will represent the
agro-pastoralist (P) and player 2 will represent nature (N).
Pastoralist.1 represents the node at the root of the tree from which the
first available strategies follow. The strategies are named branch lines in
this representation form, and depart from the small circles called nodes.
In this game, the two available strategies for the agro-pastoralist are to
settle in the current land or to migrate and move to another place. The
next nodes of the game N.1 and N.2 represent two different nodes for
nature. When two nodes are connected with a dotted line, which is not the
case in this example, it is said that both nodes are in the same information
set. That is, there is incomplete information about the actions of the last
player. Otherwise, as we move forward in the tree from left to right, infor-
mation about the last player’s moves is gained along the way.
In this game, the two available strategies for nature are rain or
drought. In reality the level of rainfall could fall at different levels along
a continuum ranging from no rain at all to a lot of rain, but to keep this
game simple, consider these two extremes. The two nodes for nature (N.1
and N.2) are chance decision nodes as there is uncertainty on what nature
will do. In chance decision nodes, all branches deriving from them have a
probability distribution equal to 1 (for example, b 1 1 − b 5 1). The agro-
pastoralist can have some information on these probabilities, such as what
the rainfall was last year, however, there is still uncertainty on what in fact
will happen.
The letters or numbers, in this case X’s at the terminal nodes, repre-
sent the payoffs2 of each decision path. These payoffs will depend on the
utility functions of the players in the game. In this case only the actions
of nature will have an effect on the payoffs for the agro-pastoralist and
not vice versa. Therefore, only one payoff is sketched. In the case that
two or more players have a payoff contingent on the final scenario,
these payoffs will be written at the end of the terminal nodes separated
by a comma, the first letter or number being the payoff for player 1, the
second letter or number the payoff for player 2 and so forth. In the game
considered here, if the agro-pastoralist settles and it rains, she or he will
receive X1, if he or she settles and there is drought, she will receive X2
and so on.
In order to find a solution for this game, we have to compute the
expected payoffs at each node, starting from the right and moving to the
left. This commonly used method is called solving by backward induction.
In this game we thus have to compute the expected payoffs at N.1 and
N.2, that is, the expected payoffs from settling and moving respectively.
To compute the expected payoffs of a decision under uncertainty, one has
to multiply the payoff by the probability of each payoff and add them.3
For example, the expected payoff from settling in this game would be:
Ep(S) 5 b(X1) 1 (1 − b)(X2). The same would have to be done for the
other available strategies for player 1. These two payoffs would have to
be computed and compared in order to derive a preferred strategy.4 To
identify the preferred strategy, an arrow in the branch that represents the
preferred strategy can be drawn. To make this clearer, numerical payoffs
to this example are given below.
Consider the hypothetical payoffs for the agro-pastoralist from each
of her or his available strategies as those depicted in Figure 10.2. These
payoffs would have to include all benefits and costs (for example, mon-
etary costs from moving, opportunity costs, and so on) derived from each
of her or his strategies given the actions of nature. That is, the payoffs
reflect the net benefits of all contingent scenarios.
Given the payoffs from Figure 10.2, the expected payoff for the
agro-pastoralist from settling would be:
Rain (0.5)
6
Settle
N.1 Drought (0.5)

–3
Pastoralist.1
Rain (0.5)
2
Migrate
N.2 Drought (0.5)

–1
Figure 10.2 The ‘migration game’ with numerical payoffs
Ep(S) 5 6(0.5) 2 3(0.5) 5 3 2 1.5 5 1.5
Conversely, the expected payoff from migrating would be:
Ep(M) 5 2(0.5) 2 1(0.5) 5 1 2 0 .5 5 0.5
Given these payoffs and probabilities, the best strategy for the agro-
pastoralist would be to settle since the expected payoff from settling will be
higher than the expected payoff from moving (Ep(S) . Ep(M)). An arrow
on the ‘settle’ branch denotes this.
Now, let us consider another approach to this game. What if the agro-
pastoralist is not sure about the probabilities of drought or rain? How
much would the expected probability have to fall or rise in order for
the chosen strategy to still be preferred? To find a solution, we need to
compute the same expected payoffs and solve for b:
Ep(S) 5 6b 2 3(1 2 b) 5 9b 2 3
And, Ep(M) 5 2b 2 1(1 2 b) 5 3b 2 1
For settling to be a preferred strategy, Ep(S) . Ep(M), thus:
9b 2 3 . 3b 2 1
Solving for the past inequality, b would have to be higher than 2/6 for set-
tling to be the preferred strategy for the agro-pastoralist. This means that
for all probabilities of rain higher than 0.33 it will always be best for the
agro-pastoralist to settle.
Clearly this is a simplified view of the available decisions to the agro-
pastoralist and of the variables that will shape his or her payoffs. However,
once familiar with the way games can be constructed and solved, and the
interactions amongst several players, games illustrating real-life situations
can be constructed. The next section will introduce the other, usually more
often used, form of game in non-cooperative game theory, the ‘normal’ or
strategic form game.
Non-cooperative strategic form games

The other form in which non-cooperative games are depicted is the normal
or strategic form (Figure 10.3). This form is useful to depict interactions
amongst two or more players in which their payoffs are shaped on what
the other player does and for illustrating simultaneous decision-making.
Consider two agro-pastoralists or two groups of agro-pastoralists:
agro-pastoralist 1 (player 1) and agro-pastoralist 2 (player 2). In normal/
strategic form games, the payoffs for each of the players are depicted
within the boxes of the matrix (Figure 10.3). In some normal form repre-
sentations, the payoffs are written in a parenthesis with the first number
being the player 1 payoff and the second number or letter in the parenthe-
sis, the player 2 payoff. In other normal form representations, for example
the game in Figure 10.3, the payoffs are depicted at the corners of each
Past.2
Settle Migrate
0.75 1.5
Settle *
0.75 0.5
Past.1
0.5 –1
Migrate
1.5 –1
Figure 10.3 A normal 2 × 2 form of the revisited ‘migration game’

box within the matrix, the payoff to the left top corner of the box being
the payoff for player 1 (agro-pastoralist 1) and the one located at the right
bottom of the box, being the payoff for player 2 (agro-pastoralist 2). One
of the players plays the columns and the other player plays the rows. The
names of the players and available strategies are written at the top of the
columns and at the side of the rows, respectively.
Drawing on the example presented in the past section, suppose that the
payoffs for the agro-pastoralists are not only dependent on what she or
he thinks nature will do, but on what other agro-pastoralists do. Suppose
then that the payoffs from settling for the agro-pastoralist will be shaped
by what other agro-pastoralists do and vice versa. For example, if land
is scarce then if both agro-pastoralists settle in the same area the benefits
for each agro-pastoralist will be reduced. If both agro-pastoralists decide
to settle in the area (settle, settle), the payoff for each agro-pastoralist will
no longer be 1.5 as derived in the past example. If both agro-pastoralists
decide to stay, each will only receive 0.75. If only one of them decides to
migrate, the one who settles will receive 1.5 while the one who migrates will
receive 0.5. If, on the other hand both players decide to migrate, each of
them will receive –1. That is, both players would carry the costs and risks
of migrating and the fact that both will graze in the same area imposes a
burden on one another.
In order to derive the contingent or best response strategies for each
player given the payoffs for the game and what the other player does, let
us start with player 1. What would the best response for agro-pastoralist
1 be if she or he thinks agro-pastoralist 2 will settle (left column)? Given
that strategy by player 2, the best response strategy for player 1 is to settle
since 0.75 . 0.5 (the two payoffs at the left top corner of the two boxes in
the left column). If, on the other hand, agro-pastoralist 2 migrates (right
column), the best response for agro-pastoralist 1 would be to settle since
1.5 . −1. These best response strategies are illustrated by arrows at the
sides of the matrix. The vertical arrows illustrate agro-pastoralist 1’s best
response strategies for this game. In this game, player 1 (agro-pastoralist
1) has a dominant strategy. A dominant strategy is a strategy that given
the rules and payoffs of the game, in any specific node, is preferred by a
player no matter what the other player does.
The procedure followed with player 1 now needs to be done with player
2. What would the preferred strategy for player 2 be, given the actions
of the other player? If player 1 decides to settle (top row), player 2’s best
response strategy would be to settle given that 0.75 . 0.5 (bottom right
payoffs from boxes in the top row). If, on the other hand, player 1 decides
to migrate (bottom row), the best response strategy for player 2 will still
be to settle, since 1.5 . −1 (bottom right payoffs from boxes in the bottom
row). Thus, the dominant strategy for player 2 will also be to settle. The
horizontal arrows represent the best response strategies for player 2.
When a game has at least one of a player’s arrows pointing to a box and
at least an arrow of the other player pointing to the same box, then the
pair of strategies is said to be in a Nash equilibrium.5 A Nash equilibrium
is thus defined as any pair of strategies with the property that each player
maximizes her or his payoff given the actions of the other player. In this
case, the Nash equilibrium for this game is settle/settle.
There are additional names for specific strategies and equilibria. A pure
strategy is one that does not involve chance (probabilities). And a pure
Nash equilibrium is the equilibrium reached when each player plays a pure
strategy. In the game illustrated in Figure 10.3, there is only one Nash
equilibrium (S,S) and this is a pure Nash equilibrium, since it does not
involve probabilities. Another type of strategy, not present in this game, is
mixed strategies. A mixed strategy requires a player to randomize her or
his pure strategies in order to keep the opponent guessing. Consequently,
a mixed equilibrium is the equilibrium reached when each of the players is
playing a mixed strategy.
Two further equilibrium notations are used for specific purposes in
game theory. Symmetric equilibrium, in which every player chooses the
same strategy and an asymmetric equilibrium in which at least two players
choose different strategies. When there is only one equilibrium in the
game, this is called the unique Nash equilibrium and when there is more
than one equilibrium, it is said that the game has multiple Nash equilibria.
In the game presented in Figure 10.3, there is a unique pure symmetric
Nash equilibrium. This equilibrium is represented with a star in the middle
(note the star in the middle of the box settle/settle), though there are many
other identifiable forms for illustrating where the Nash equilibria are.
Having presented a brief review of game-theoretic language and repre-
sentation forms, next section will review the game-theoretic approach to
CPRs.
Game theory, common pool resources (CPRs) and common pool

institutions (CPIs)6
Game theory has been extensively used as a framework for analysis of
CPR problems (see Ostrom et al., 1994; Baland and Platteau, 1996).
Before presenting the game-theoretic approach to common pool resources
(CPRs), it is important to review the way CPRs have been conceived and
the problems arising in the management of these resources.
Common pool resources have been mainly explained in terms of the phys-
ical attributes of the goods or resources. CPRs are considered to share two
characteristics: (1) the difficulty of excluding individuals from benefiting
from the resource and (2) the subtractability of the benefits consumed by
one individual from those available to others (Ostrom et al., 1994). These
characteristics have been considered to generate two broad problems: that
of appropriation and that of provision. The first relates to extracting more
than the socially optimal level.7 The difficulty of excluding others from the
use of a CPR creates the problem of effectively limiting use (Ostrom et al.,
1994). In the case of renewable natural resources, such as a common prop-
erty forest land, for example, the concern is that the flow extracted will not
exceed its regeneration rate. The second problem refers to the difficulty of
investing for the ‘adequate’ provision of the CPR or in activities related to
its improvement or maintenance. Both these problems can be considered to
generate negative consequences for other appropriators/users.
Commonly used games for depicting CPR problems

Three main games have been used for depicting CPR problems. These
games are: (1) the prisoner’s dilemma game8 (2) the chicken game9 and
(3) the assurance game.10 The prisoner’s dilemma game has been widely
used to represent the appropriation problems encountered in CPRs. When
limits for resource use cannot be established, all users of the CPR will want
to use as much as possible so as to maximize their own profit, resulting in
over-appropriation of the resource. In the case of natural resources this
over-appropriation can lead to over-exploitation.
The prisoner’s dilemma game is probably the most well-known game
with two players each having two strategies (Figure 10.4). The two avail-
able strategies for each player are to cooperate or to defect. The payoffs
considered for this game are such that c . a, d . c and a . d. Thus, given
the payoffs for this game, the best response strategy for each player is to
defect. This game has a unique Nash equilibrium that is: defect, defect.
Here is where the dilemma lies, since the players could both be better off if
both chose to cooperate (C) simultaneously since a . d.
In the case of resource provision, the problem arises with free-riders
benefiting from the resource (as a public good) provided without having
contributed to its provision. For example, time investments for the moni-
toring and enforcement of communitiy forestry or a commonly managed
natural reserve. This problem has been modelled, depending on the provi-
sion technology, with the prisoner’s dilemma or with the assurance game,
the former having no one contributing to the resource provision (Figure
10.4) and the latter having players contributing if, and only if, the others
are to contribute (Figure 10.5). The assurance game for which the payoffs
are such that a . c and d . b (Figure 10.5) thus depicts situations in
which one person’s contribution is not enough to gain a collective benefit,
but the contribution of both players will result in a joint benefit. That
Player 2
C D
a c
C
a c
Player 1
c d
D *
b d
Source: Adapted from Ostrom et al. (1994, p. 53).
Figure 10.4 Adaptation of the prisoner’s dilemma game (c > a, d > c and
a > d)
Player 2
C D
a c
C *
a c
Player 1
c d
D *
b d
Figure 10.5 Adaptation of the assurance game (a > c, c > d and d > b)
is, both players would be willing to cooperate if, and only if, the other
player cooperates. As can be seen in Figure 10.5, this game has multiple
equilibria.
Lastly, the chicken game has been used to illustrate assignment prob-
lems when there are different resource locations with different richness.
Consider, for example, two fishing grounds with different distances to a
protected area. This game has multiple equilibria, as does the assurance
game. If one of the locations is much better than another, both players
will want to use the same location (D,D) and this might not be optimal
Player 2
C D
a c
C *
a c
Player 1
c d
D *
b d
Source: Adapted from Ostrom et al. (1994).
Figure 10.6 Adaptation of the chicken game (c > a, c > d and b > d)
for the group as a whole. If both locations are equally good (Figure 10.6),
players will be indifferent and will play a mixed strategy leading either to
one (C,D) or the other equilibrium (D,C).
Due to the attributes of CPRs, individuals jointly using CPRs are many
times stuck in a dilemma. Without previous communication and the lack
of institutions (norms and conventions), agents pursuing their own inter-
ests will engage in non-rational collective outcomes that result in resource
degradation. Institutional choice theorists have focused on identifying
viable institutional alternatives by which appropriators of the CPR can
choose (1) how much, when, where and with what technology to withdraw
units and (2) how much, when and where to invest in the maintenance
of the CPR (Ostrom et al., 1994). The next section outlines the game-
theoretic logic for such viable institutional solutions.
Institutional solutions to CPR problems

While the use of the prisoner’s dilemma game was for many years mistak-
enly applied to represent all CPR situations (for a critic see Ostrom, 1990;
Ostrom et al., 1994), as mentioned in the previous section, it can be very
useful for illustrating appropriation and provision problems encountered
when managing CPRs.
Consider the prisoner’s dilemma game again, now in the extensive form,
with the payoffs thought out by Dawes (1973). As we have seen, the pris-
oner’s dilemma structure games have a unique Nash equilibium which is
(D,D). The game in Figure 10.7 depicts this equilibrium in a square with
the resulting payoffs of 0,0 respectively.
C 10,10
C
D –1,11
1 2 C 11,–1
D
D 0,0
Source: Dawes (1973).
Figure 10.7 Adaptation of the prisoner’s dilemma game in extensive form

and numerical payoffs
Now, consider that this game is not finite. That is, there is not only one
round to the game, but that the game is played over and over again. This
leads to the alternative game set by Ostrom (1990) following the structure
of the prisoner’s dilemma game but adding two branches for each player
at the beginning of the game. These branches are termed: agree (A) or do
not agree (~A). That is, players can get together and arrange a contract in
order to arrive at a joint payoff (Figure 10.8).
In this game, the issues are: (1) whether the contracts can be binding,
that is, there is unfailing enforcement and (2) whether the payoffs are such
that the transaction costs of enforcing the contract are lower than the
expected gains from the contract. A binding contract is interpreted in non-
cooperative game theory as one that is unfailingly enforced by an external
actor. The cost of enforcing the agreement is denoted in Figure 10.8 by e.
Given the payoffs and structure of the game, the dominant equilibrium
is having both parties agreeing to cooperate. The equilibrium is shown
in the rectangle at the bottom of the tree. If these strategies are followed,
players will receive the joint benefit from cooperation minus the shared
enforcement costs (10 – e/20, 10 − e/20). Note that this equilibrium will
hold only if joint enforcement costs (e) are lower than 20.
Encouraged by this and other parallel work, over the past few decades
theorists have explored the relationships between rules, institutions,
property rights and resource user characteristics for arriving at coopera-
tive outcomes that can enhance collective benefits. The next section will
illustrate the use of game theory for identifying key variables and levels of
D 0,0
D
~A C
11,–1
D
C –1,–11
~A
C
10,10
D
0,0
D
* e < 20 C
A 11,–1
D
C –1,11
C
10,10
1 2 1 2 D
0,0
D
~A C
11,–1
D
C –1,11
C
A 10,10
A
10 – e/2, 10 – e/2
Source: Ostrom (1990, p. 15).
Figure 10.8 Example of a self-financed contract enforcement game
variables that can enhance cooperation and for assessing the viability and
desirability of policy options using examples from common pool resource
management in semi-arid Tanzania.
Worked examples from semi-arid Tanzania

Much of the Tanzanian land area is still under open access or some form
of common property or management arrangement (Quinn and Ockwell,
Chapter 9, this volume).
Two different problems will be taken as examples to illustrate the use
of game theory for understanding decision-making processes and assess-
ing the viability and desirability of policy options in the management of
common pool resources. The first will be the problems encountered by vil-
lages that face soil erosion on communally managed land. The second will
be the problems faced by communities living next to wildlife conservation
reserves and who are prevented from using resources within the reserve.
Erosion prevention/restoration and cooperation in common pool scenarios

Semi-arid ecosystems are characterized by highly variable rainfall pat-
terns, such as short, intense storms, and high evapotranspiration rates.
In several areas of semi-arid Tanzania, there are serious problems of land
Player 2
I ~I
10 –1
I
10 11
Player 1
11 0
~I *
–1 0
Figure 10.9 Adaptation of a prisoner’s dilemma structure game for

erosion restoration/prevention investments in common pool
scenarios
degradation and soil erosion affecting the already low living standards of
people living in those areas.
Consider an area with a high degree of land erosion. Erosion can depend
on natural factors such as rainfall patterns, wind and topography but can
also result from human activities such as deforestation, clearing for cul-
tivation, overgrazing, extensive fuel wood cutting, and so on. Erosion in
turn affects the productivity of the natural resource base. The negative
impacts of erosion can be ameliorated by investments in the common
property land such as afforestation, the construction of diversion ditches,
ridge banking, de-stocking, and so on. The implementation of these activi-
ties would provide joint benefits for all users; however, problems of provi-
sion can arise with free-riders benefiting from the resource without having
contributed efforts towards these activities.
Recall that provision problems in CPRs can be modelled, depending on
provision technology, with the prisoner’s dilemma or with the assurance
game, the former having no one contributing to resource provision and the
latter having players contributing if, and only if, the others are to contrib-
ute. These two games are illustrated in Figures 10.9 and 10.10.
The individual profits from realizing the investments are the individual
benefits derived from the investment, minus the costs incurred in the
investment, including the opportunity costs of time devoted to the activity.
Realization of these investments will depend on having the flow of money
or resources required for realizing the investment, but will also depend on
what an individual thinks other members of the village will do.
Consider the payoffs and strategies of Figure 10.9. Two different
Player 2
I ~I
10 –2
I *
10 9
Player 1
9 0
~I *
–2 0
Figure 10.10 Adaptation of an assurance structure game for erosion

restoration/prevention investments in common pool scenarios
members of the community are represented as player 1 and player 2 respec-

tively, each having two possible strategies: to invest (I) or not to invest (~I)
in restoration/prevention practices. If both players invest in the improve-
ment of the CPR, there will be a joint benefit of 20. However, when maxi-
mizing individual payoffs, the resulting equilibrium is no one contributing
to the investment (~I, ~I), thereby deriving a joint benefit of zero.
Consider now that the technology is such that it is convenient to con-
tribute only if the other player is to contribute. As seen in the past section,
these situations can be modelled by an assurance structure game for which
there are two possible equilibria. That is, when a player takes the initiative
to invest (I), the best the other player can do is go and help him or her (I),
since by doing so her or his payoffs will increase. However, if nobody takes
the initiative, there will be no investment (~I). The result, thus, is either
investment by both parties (I,I) or non-investment by both parties (~I,~I)
as illustrated in Figure 10.10.
As seen in the previous section, there are institutional alternatives by
which resource users can organize and arrange for the collective gain
to be achieved, transaction costs permitting. The question that remains
is the effectiveness of village governance. If communal institutions are
weak, should there be a role for third parties, such as the government or
NGOs entering into co-management arrangements for the fostering of
such agreements? Among the ways in which the fostering of agreements
could be effected are providing the initial money for realizing investments,
improving information to stakeholders on the benefits that can be derived
from agreements and/or reducing the transaction costs of enforcing
agreements.
Protected areas, eviction and encroachment, game theory as an analytical

tool for assessing the viability of policy options
A central policy in Tanzania has been the creation of protected areas as
part of a broader strategy aiming to promote biodiversity conservation
and the meeting of local and national development needs. The viability of
these policies cannot be isolated from strategic interactions and decisions
made by different individuals and institutions at the local level. As will
be illustrated in this section, game theory can also be useful in assessing
the viability and desirability of policy options in the context of strategic
interactions.
Until now, we have considered games made under the assumption that
players are symmetric, facing the same decision alternatives and being
in the same position, for example, both players being appropriators and
facing cooperate or defect strategies. These symmetries, however, can be
relaxed to construct games in which the strategies do not need to be the
same, nor the players to be assigned to the same positions.
Let us first try to map a decision problem faced by communities who are
living adjacent to a reserve and who are completely prevented from the use
of resources within the reserve.
Consider first a hypothetical scenario illustrated by Figure 10.11,
where M.1 represents the first node for a member or members of the
reserve adjacent communities and Ch.1 represents a chance decision
node with a probability distribution equal to 1. M.1 has two available
strategies: to trespass (t) and use the resources within the area or not to
trespass (~t). If the user trespasses and is not detected (~D) he or she
will receive a benefit (B) from the additional resources within the reserve
(for example, grazing land, water, wood, bushmeat, medical plants, and
so on) with a probability of (1 − a). B will be shaped by the availability,
quality and dependence of resources inside the reserve relative to those
D (␣)
–F
t
Ch.1
B
~D (1 – ␣)
M.1
~t
Figure 10.11 The trespassing game

outside the reserve. If, on the other hand, the member is detected (D)
she or he will carry the cost of the fine (−F) with probability a. If the
member does not trespass she or he will remain under status quo (SQ)
conditions.
Let us derive the expected probabilities for the unique decision node for
the members of the reserve adjacent community:
Ep(t) 5 a (−F) 1 (1 − a) B 5 −a F 1 B − aB 5 −a (F 1 B) 1 B
Ep(~t) 5 0
Thus, for not trespassing to be a preferred strategy either:
(1) a . B / (F 1 B) or
(2) F . B (1 − a) / a or
(3) B , aF / (1 − a)
From the previous set of equations, there are four ways of making not
trespassing to be a preferred strategy. These are either to:
● increase the probability of getting caught to the level shown in the

equation; or
● increase the level of the fine to the level shown in the equation; or
● decrease the level of benefits to the level shown in the equation; or
● a combination.
The use of game theory is not a substitute for assessing side-effects from
each of the available strategies. These can be built up in other rounds of
the game or subsequent games. Possible side-effects from each of these
policy options are, for example, increasing the probability of getting
caught would mean having to invest more resources in monitoring activi-
ties, increasing the level of the fine might foster corruption and so on. The
usefulness of this example rests in illustrating a situation to identify the
key variables shaping a decision and their levels.
Let us now consider that complete exclusion is not the only scenario
into which reserve adjacent communities can be put. The state, or the
managers of the reserve, could have a continuum of possibilities ranging
from uncontrolled access to total exclusion (Figure 10.12).
Merging Figure 10.11 and 10.12, we could construct a game with
two players: the state (or managers of the reserve) and reserve-adjacent
Uncontrolled
access
R.1
Total
exclusion
Figure 10.12 Reserves and the continuum of resource use allowances
UA
–L, B
t –m +f, –f
D (␣)
RA ~D (1 – ␣)
R.1 –m –L, B
~t
M.1 –m –l, b
t –M –L +F, –F
D (␣)
~A ~D (1 – ␣)
–M –L, B
M.2 ~t
–M, 0
Figure 10.13 The trespassing game with two players
communities (Figure 10.13). R.1 is the first and only decision node for the
reserve managers (or the state) in this game and M.1 and M.2 are two dif-
ferent nodes for the members of the resource-adjacent communities. For
simplicity consider the three available strategies for the managers of the
reserve to be (1) uncontrolled access denoted by UA, (2) regulated access
denoted by RA and, (3) total exclusion denoted by ~A (no access). For
the adjacent communities, consider the same available strategies at each
decision node as those in Figure 10.11.
If the state decides to impose a policy of unregulated access, the reserve
will bear a cost from the loss of resources (−L) and the adjacent com-
munities will receive a benefit (B) from unregulated access to additional
resources within the reserve.11 If the state decides to regulate access,
allowing uses that do not threaten the ecological stability of the reserve,
the adjacent communities have two options, to trespass and carry out
activities that are not allowed (t) or not to trespass (~t). If they trespass (t),
and are detected (D), the reserve will bear the costs of monitoring illegal
activities (−m) and will receive the level of the fine (f), whilst the adjacent
communities will bear the cost of the fine (−f). If community members
are not detected (~D), the reserve will bear the costs of monitoring (−m)
anyway and the total losses from activities (L). In this case, the members
of the adjacent communities will receive the benefits from the additional
resources (B). If the communities do not trespass (~t), the reserve will bear
the cost of monitoring (−m) plus a smaller loss (−l) due to allowed activi-
ties. The communities in this case will receive a smaller benefit due to the
allowed activities (b). In the third strategy, if no access is allowed (~A), the
reserve will carry the costs of monitoring for all types of activities (−M).12
The adjacent communities, if trespassing and detected, will receive the
level of the fine (−F), minus the loss from additional resources (−L). If
members are not detected (~D), the reserve will bear the no access moni-
toring costs (−M) minus the loss from resources (−L). Communities in this
case will receive the additional benefits (B). If adjacent communities do
not trespass, the reserve will bear the monitoring costs of no access (−M)
and the adjacent communities will remain under status quo conditions
(SQ).
Let us give hypothetical numerical payoffs to solve for this game.
Suppose that:
L 5 180 B 5 150 m 5 20 f 5 20 l 5 20 b 5 80
M 5 30 F 5 40 a 5 0.5
Solving by backward induction, in M.1 the best the member can do is

not to trespass since her or his expected payoff from not trespassing is
higher that that of trespassing:
Ep(~t) 5 80 . Ep(t) 5 −20(0.5) 1 150(0.5) 5 −10 1 75 5 65
If the member were in M.2, the best she or he can do is to trespass since the
expected payoff from trespassing is higher than that of not trespassing:
Ep(t) 5 −40(0.5) 1 150(0.5) 5 −20 1 75 5 55 . Ep(~t) 5 0
Given this preferred strategy for the community member, the best the
reserve can do is to regulate access (RA) since the expected payoffs from
regulating access are higher than those of unregulated access (UA) and
higher than those of no access (~A):
UA
–180, 150
0, –20
t D (0.5)
RA ~D (0.5)
–200, 150
R.1
~t
M.1 –40, 80
–170, –40
t D (0.5)
~A ~D (0.5)
–210, 150
M.2 ~t
–30, 0
Figure 10.14 The trespassing game with two players and numerical values
Ep(RA/~t) 5 −40 . Ep(UA) 5 −180 . Ep(~A/t)

5 −170(0.5) –210(0.5) 5 −85 –105 5 −190
Note that if the state decided to completely restrict access (~A) the best
response strategy for the community would be to trespass (t), the payoffs
for these strategies being (−190, 55), yielding a negative joint payoff of
–135.
Given the rules, available strategies and values for this game, the result
will be to regulate access and not to trespass (RA, ~t) illustrated by a
square. Note that from the way in which the payoffs were given, the result-
ing pair of strategies is also the one that yields the higher social profits, since
the joint profits of regulating access and not trespassing (RA,~t) are equal
to 40, being the highest joint profit that can be achieved in this game.
Notes
1. Current applications of game theory include biology, engineering, political science,
internal relations, computer science and philosophy.
2. In economic jargon, the payoffs represent utilities that comply with the Von Neumann–
Morgenstern axiom.
3. If there are subsequent nodes involving probabilities, starting with the payoffs at the
terminal nodes, we would move from right to left computing the conditional or inde-
pendent probabilities, following the rules for independent or conditional probabilities
respectively.
4. In some cases knowing that one outcome is preferred over another is enough to derive
a preferred strategy.
5. A pair of strategies with these properties is called a Nash equilibrium, after J. Nash
(1953) who showed that all games with a finite number of strategies have at least
one equilibrium, provided that mixed strategies are allowed. Mixed strategies will be
explained later.
6. This section is adaptively drawn from Ostrom et al. (1994).
7. There are some problems in determining the socially optimal point. In addition to its
praxis it will depend on which and how many involved or affected agents are introduced
to the equation. In this case, social optimality is considered the result of equating mar-
ginal social benefits to marginal social costs of appropriators, taking appropriators to
be the individuals sharing the use of the CPR (Ostrom et al., 1994).
8. The ‘prisoner’s dilemma’ was originally framed by Merrill Flood and Melvin Dresher in
1950. Albert W. Tucker formalized the game in an academic presentation in the 1950s
with prison sentence payoffs and gave it the ‘prisoner’s dilemma’ name. Imagine that
two suspects have been arrested by the police. The police have insufficient evidence for
a conviction, and, having separated both prisoners, visit each of them to offer the same
deal. If one testifies (defects from the other) for the prosecution against the other and
the other remains silent (cooperates with the other), the betrayer goes free and the silent
accomplice receives , say, the full ten-year sentence. If both remain silent, both prisoners
may be sentenced to only six months in jail for a minor charge. If each betrays the other,
they each receive a five-year sentence. Each prisoner must choose to betray the other or
to remain silent. Each one is assured that the other would not know about the betrayal
before the end of the investigation. How should the prisoners act? (For more details on
the origin and formalization of the prisoner’s dilemma see Poundstone, 1992).
9. The ‘chicken game’, also known as the hawk–dove or snowdrift game, was first formal-
ized by John Maynard Smith and George Price in 1973. The principle of the game is
that while each player prefers not to yield to the other, the outcome where neither player
yields is the worst possible one for both players. The name ‘chicken’ has its origins in
a game in which two drivers drive towards each other on a collision course: one must
swerve, or both may die in the crash, but if one driver swerves and the other does not,
the one who swerved will be called a ‘chicken’, meaning a coward.
10. The ‘assurance game’ is a generic name after Sen (1967), for the game also known as the
stag hunt game. The French philosopher Jean Jacques Rousseau, in his 1755 writings
on inequality, presented the following situation: two hunters can either jointly hunt a
stag (an adult deer and rather large meal) or individually hunt a rabbit (tasty, but sub-
stantially less filling). Hunting stags is quite challenging and requires mutual coopera-
tion. If either hunts a stag alone, the chance of success is minimal and thus there is the
need of cooperation and trust to achieve a larger joint benefit.
11. Non-priced/non-marketed values could be incorporated to the payoffs of each
alternative.
12. It could be argued that M is very similar to m, as the costs of monitoring for additional
activities, while monitoring anyway may be very small (decreasing marginal costs).
Even if this difference is small, it exists and so, for illustrative purposes, let us denote it
by using the letter M.
References
Baland, J.-M. and J.-P. Platteau (1996), Halting Degradation of Natural Resources. Is There a
Role for Rural Communities?, Rome: FAO and Oxford, UK: Oxford University Press.
Binmore, K. (1994), Just Playing: Game Theory and the Social Contract, Cambridge, MA:
MIT Press.
Dawes, R.M. (1973), ‘The commons dilemma game: an n-person mixed motive game with
a dominating strategy for defection’, Oregon Research Institute Research Bulletin, 13(2),
1–12.
Maynard Smith, J. and G. Price (1973), ‘The logic of animal conflict’, Nature, 246, 15–18.
Nash J. (1953), ‘Non-cooperative games’, Annals of Mathematics, 54(2), 286–95.
Ostrom, E. (1990), Governing the Commons. The Evolution of Institutions for Collective
Action, New York: Cambridge University Press.
Ostrom, E. and R. Gardner (1993), ‘Coping with asymmetries in the commons: self-governing
irrigation systems can work’, Journal of Economic Perspectives, 7(4), 93–112.
Ostrom E., R. Gardner and J. Walker (1994), Rules, Games and Common Pool Resources,
Ann Arbor: The University of Michigan Press
Poundstone, W. (1992), Prisoner’s Dilemma, John Von Neumann, Game Theory and the
Puzzle of the Bomb, New York: Anchor Books, Doubleday.
Sen, A. (1967), ‘Isolation, assurance and the social rate of discount’, Quarterly Journal of
Economics, 81, 112–24.
Snidal, D. (1985), ‘The Game Theory of International Politics’, World Politics, 36(1),
25–57.
Von Neumann, J. and O. Morgenstern (1944), Theory of Games and Economic Behaviour,
Princeton, NJ: Princeton University Press.
Weissing, F. and E. Ostrom (1991), ‘Irrigation institutions and the game irrigators play: rule
enforcement without guards’, in R. Shelten (ed.) Game Equilibrium Models II: Methods,
Morals and Markets, Berlin, New York: Springer.
11. Economic valuation of different forms
of land-use in semi-arid Tanzania
Deborah Kirby
Introduction
In many ecosystems there is a conflict of interest between land-users who
wish to utilize the land for different purposes. In most of these situations,
the overriding factor influencing the use of the land is its perceived economic
value. A forest or savanna region may produce many different types of goods
for human use. Although these goods have a value to humankind, many
have traditionally been excluded from estimates of valuation. This has arisen
because policy-makers have generally assumed that natural ecosystems such
as forests or savannas have no economic value other than their direct use of
producing saleable products, and thus non-monetary values from their land-
use have been excluded from any economic analysis. Thus, forested areas
have only been valued in terms of timber production or land area available
for farming (Barreto et al., 1998), and savanna areas for the volume of sale-
able livestock (for example, Bembridge and Steenkamp, 1976).
A comparison of the relative values of the different forms of land-use,
which includes all values of the resource, monetary and non-monetary,
can help identify a socially equitable use of land. This chapter considers
how valuations of different forms of land-use can be made and emphasizes
the importance of establishing linkages between ecological and economic
systems to facilitate valuation of biological goods and services in a devel-
oping country, Tanzania. In this context, the use of a particular tool in
economic analysis, the production function, is described.
The importance of valuation of ecosystems

Many resources provided by ecosystems do not have a direct market
value. For example, the indirect values of ecosystem services provided
under traditional pastoral systems of land-use are harder to quantify com-
pared with the direct market prices of commercial ranching. Consequently,
evaluation of the economic worth of one system of land-use against
another will not be equitable if only those products that have a direct
market value are compared. One of the principal pressures on natural eco-
systems is their conversion into more productive systems in commercial
terms (Wilson, 1994). This occurs when decisions over land-use are taken
331
by individuals or groups of individuals, who are interested in maximiz-

ing their private benefits through the production of particular products
that have a specific market value to them (for example, Kirby, 2000).
However, creating a commercially increased output through a change in
land-use may not in reality result in the most productive ecosystem if as
a consequence the non-marketed values of the ecosystem are reduced or
eliminated. Policy-makers and governments have frequently overlooked
the non-market values of ecosystems with the consequence that a change
in land-use has occurred in favour of the commercial system of produc-
tion, thus leading to an overall reduction in the net benefits produced by
the ecosystem (Godoy, 1992). This has resulted in policies that have mis-
directed resource use away from the most economically valuable form of
production and towards maximizing financial outputs.
In order to make rational decisions on issues of natural resource man-
agement it is therefore important that consideration is given by policy
analysts to all aspects of an ecosystem’s value. Only if this is undertaken
can the total economic worth of the ecosystem be estimated and an effec-
tive comparison of its value under different uses made. Valuation of
ecosystem productivity can be undertaken by numerous methodologies
(see, for example, Dixon and Sherman, 1990; Heywood, 1995). However,
this chapter concentrates on one type of methodology that is particularly
useful for the valuation of non-marketed and marketed values associated
with biological resources that support economic activity. This is the pro-
duction function approach.
Tanzanian land-use issues

As more detailed descriptions of the land-use problems of Tanzania are
given elsewhere (Quinn and Ockwell, Chapter 9, this volume), this chapter
gives only a brief overview of the relevant issues. The Tanzanian semi-arid
to arid climate has a rainfall ranging from 400 to 1200 mm. The result-
ing vegetation varies from grasslands to savanna thornbush and wooded
savanna in higher rainfall areas. The population density is relatively low
and at least 11 per cent of the population are below the poverty line. Much
of the land area is still under open access or some form of common prop-
erty tenure.
The traditional form of land-use in semi-arid Tanzania has been pas-
toralism, defined as a production system where at least 50 per cent of
production (subsistence and marketed) comes from livestock or livestock-
related activities. This is predominantly an extensive production system,
ranging from nomadism, where no form of cultivation takes place, to
semi-transhumance, where part of the family is sedentary and practises
cultivation (Niamir-Fuller, 1998).
Economic valuation of land-use in semi-arid Tanzania 333
In recent decades, pastoralists have faced increasing problems of

resource degradation as a result of immigration of other populations and
changes in land-use. In many parts of the country this has led to appar-
ently unsustainable forms of land-use and the exclusion of many of the
indigenous population from their homelands, with a resulting increase in
poverty and deprivation. The reasons and consequences of such changes
in land-use are discussed below.
The causes and consequences of change in the pastoral lifestyle
● Settlement policies. Pastoral settlement leading to agricultural farm-

ing is a common government policy in many African countries.
● Security of land tenure. Many pastoralists are also settling of their
own accord in response to decreasing rangeland areas and also in
order to guarantee land tenure.
● Tourism and wildlife conservation. Government policies have encour-
aged the takeover of land for private hunting enterprises. This has
resulted in the progressive reduction of rangeland area for grazing
and in the loss of wildlife resources important for the pastoral
lifestyle.
● Forestry. Clearance of forested land has occurred for charcoal and
in order to increase grazing area.
● Land privatization. A policy of agricultural promotion as the pre-
dominant land-use has encouraged the migration of agriculturalists
into ecologically fragile areas. With land privatization has come the
fencing of areas of rangeland for private ranching. This has resulted
in the expropriation of the land for the rich and a system of farming
incompatible with the ecological conditions.
In conjunction, these policies have resulted in the deterioration of the

dryland ecosystems of Tanzania through habitat loss, wildlife decline,
deforestation and overgrazing. Settled agriculture results in active clear-
ance of vegetation to make space for livestock and has thus reduced the
tree and plant diversity that has traditionally provided food and medicines
for the local population. In addition, lower grazing pressure in distant
pastures often results in the invasion of the range in these areas by unpal-
atable plants.
Settlement of pastoralists has also resulted in the breakdown of tra-
ditional systems for managing natural resources, resulting in inefficient
management of the habitat and a consequential decrease in livestock prod-
uctivity and increased impoverishment of this sector of the population.
The predominance of a social and cultural emphasis towards agriculture
and against other forms of land-use in Tanzania has enabled changes to

occur. Political discrimination against nomadic pastoralism and hunter-
gathering has led to the belief by many that this form of lifestyle is
backward, primitive and economically unproductive. Pastoralism and
hunter-gathering is predominantly seen as an inferior and unproductive
use of the land. That much of the land is common land enables those
engaged in formal agricultural practices that require privatized land,
greater economic status and political control over people involved in more
traditional land management practices. This gives the latter groups little or
no control over the land-use. In this context, and with the gathering body
of evidence that settled agriculture is an ecologically, and thus economi-
cally, unsustainable land-use option in semi-arid land, there is a clear need
for an accurate assessment of the full values of each form of production.
More precise estimation in monetary terms of the value of each system
in supporting production will allow a direct and fair comparison of each
form of land-use to be made.
Ecosystem values
In the past few decades, attention has been increasingly turning towards
recognizing the fact that natural ecosystems such as forest and savanna
provide a wide range of products that are highly valuable but may not
have an immediately apparent market value (Scoones et al., 1992). It
is realized that failure in the past to include these types of ecosystem
values in analyses of land-use has led to the undervaluing of ecosystems
(Swanson and Barbier, 1991). Scientists and to a certain extent policy-
makers are therefore now recognizing that if the total economic value of
an ecosystem is to be realized, a monetary value must be assigned to the
resources in question, enabling them to be ‘counted’ in an estimation of
the ecosystem’s worth.
The lack of a market value, and thus underestimation or complete
absence of the resource’s value, may be for one of two reasons. First,
the resource may not be one that is exchanged in the marketplace. For
example, hunting and gathering of wild products for subsistence living will
directly use many natural resources but these may be consumed directly
rather than sold (Campbell et al., 1997). In the case of livestock, animals
may be used for products other than meat that do not have a market value,
such as transport or labour (Scoones, 1992). Second, the value of the
resource may be in supporting the production of other goods rather than
through its own direct utilization.
The remainder of this section describes in detail the differences between
different forms of ecosystem products. These products can be separated
into three categories, according to their type of use.
Commercial values
Extraction of timber or production of livestock from an ecosystem will
directly produce revenue for the land-users. If they are marketed, their value
will be equal to the revenue from selling the timber or livestock minus the
costs of production and transport to market (Godoy, 1992). If they are not
marketed but kept for home consumption they will also have a value although
this will be considerably harder to impute, and may not be equal to the value
that they would realize if they were sold on the open market (Godoy, 1992).
Non-timber and non-meat values

A savanna or forest ecosystem can potentially provide a great number of
non-timber or non-meat goods and services. Studies have shown that in
the case of a forest ecosystem, local people living in and around it regularly
use a large number of forest products. Many of these products may be
marketed and thus have market values, for example, wild fruit and insects,
wood for construction purposes, herbs and medicinal plants (Fuentes,
1980; Oommachan and Masih, 1988). However, many goods and services
may be extracted and only used for home consumption, in which case they
will not have an immediately identifiable market value (Peters et al., 1989).
In the case of savanna ecosystems, a large number of products including
palms, grasses, reeds, fruits and wild mushrooms have all been shown to be
used by the local inhabitants (Campbell et al., 2000a; IIED and HNWCP,
1997). In addition, livestock production may produce a number of benefits
besides those of meat, for example, transport, which must be included in
an evaluation of the productivity of the system. As with forest products,
these products may or may not have a direct market value, depending on
the use to which they are put.
Regardless of whether either of these products are marketed or non-
marketed, they can be described as direct-use goods. As this name implies,
these are goods that are supplied directly from the ecosystem and that
may be either domesticated or wild. In this sense, forestry and livestock
production as well as products derived from fishing and hunting are
included.
External benefits
The third category of goods and services that can be derived from ecosys-
tems are described as indirect-use resources. Many ecosystems will provide
invisible benefits to other ecosystems or people, both within the geographi-
cal area of the ecosystem, as well as outside it. In the case of both forest and
savanna ecosystems, these external benefits can be considered to exist at two
separate levels. A number of empirical studies have shown that forest ecosys-
tems provide protection for other ecosystems located downstream from the
forest. In particular, forests have been shown to provide soil protection for
downstream agricultural land (Anderson, 1987), protect downstream water
courses from increased levels of sedimentation caused by surface run-off
and increased erosion and leading to high costs of de-siltation (Panayotou,
1990; Lal, 1997), to impact on water yield and groundwater recharge, and
to control watershed hydrology and provide protection against erosion
(Rawat and Rawat, 1994; Ataroff and Rada, 2000; Putuhena and Cordery,
2000). For dry savanna ecosystems, the invisible benefits may be as signifi-
cant as those of forest ecosystems. Hydrological functions of dry savannas
have been shown to have a direct impact on the productivity of surrounding
areas. The vegetation cover of woodlands is thought to stabilize the local
climate and maintain rainfall patterns (Myers, 1995) as well as prevent soil
erosion and flooding, and control the recharging of groundwater reservoirs
(Huntoon, 1992). These positive external effects, although they may occur
outwith the boundary of the ecosystem, are nonetheless specific in their area
of impact. As such, there is a limit to the number of individuals upon which
they will have an effect.
A second type of externality, however, is also provided by forest and
savanna ecosystems, and has a greater range of impact. It is now known
that areas of vegetation cover, in particular forested areas, act as import
sinks for atmospheric carbon (Harmon, 2001; Pautsch et al., 2001; Uri,
2001). This function is globally important in terms of regulating atmos-
pheric composition and controlling global temperature. Many savanna
ecosystems also form the watershed zones of major international rivers,
for example, the Zambezi (Kundhlande et al., 2000), which are interna-
tionally important for domestic water consumption and power supply.
This type of ecosystem service differs from all others in that its use is
‘public’ in nature, that is, it is used by society as a whole, and individuals
do not have a choice over whether or not they wish to use it.
Both of these invisible benefits are termed indirect-use values, since
although not consumed directly they are essential for the production of
direct-use values. However, their very nature implies that they will have
no direct market value and thus are difficult to quantify in a cost–benefit
analysis.
Past valuation studies

A number of research studies have focused on attempting to value forest
and savanna resources (Hanemann, 1988). In all of these studies, valua-
tion of various aspects of the complete ecosystem has been undertaken.
The relative ease of valuing direct-use resources rather than indirect-use
resources is immediately obvious and for this reason, although some
studies have attempted to value indirect-use resources, the majority of
research has concentrated on the valuation of direct-use resources. Above

all, attention has also been placed on valuing tropical forest resources, and
although studies have been made across the globe, there has been a dispro-
portionate interest in Latin American forests (Godoy et al., 1993).
Phillips et al. (1994) compared the usefulness to the indigenous popu-
lation of six floristically distinct tropical forest types in south-east Peru.
They attempted to estimate the use values of each forest type through a
series of questionnaires to the local people about the relative importance
of the different forest plant species. However, the results were produced as
a rating of different values of the forest products and no monetary value
was placed on the products, thus making quantitative comparisons with
other forest values or uses impossible. In a very similar study, Toledo et
al. (1995) evaluated the usefulness of plant species in a Mexican tropical
forest. Through questioning and working with the local population, they
evaluated the number of non-timber products that were used from the
forest. Over 3000 products were identified, but as with Phillips et al. (1994),
no attempt at placing an economic value on these products was made.
A number of studies have, however, attempted to put a monetary value
on the non-timber forest resources. Peters et al. (1989) estimated the
economic value of non-timber plant products of a small (1 hectare) area
of Peruvian tropical forest. Their calculations were based on empirical
fieldwork estimating the amount of standing vegetation in the forest, and
on the prices of the products in the local market. To arrive at a valua-
tion figure, they multiplied the standing inventory of certain non-timber
forest products by the local market price, thus carrying out a very simple
cost–benefit analysis. A value of US$420 ha 1 year 1 for the products was
estimated, however, as was pointed out by Godoy et al. (1993), Peters et
al. failed to take into account that the local market prices are a reflection
of the supply of the produce. Supplying the market with all the forest
resources would result in a decrease in market prices. In addition, the
study did not provide an accurate value of the entire forest resources, since
only plant products were valued, with faunal products being completely
disregarded. Pinedo-Vasquez et al. (1992) estimated the potential value
of six fruits and derived a total value of the forest products of US$20 ha 1
year 1. Hecht (1992) estimated the opportunity cost (that is, the value of
the other products lost from using the forest in this way) of using forests
for livestock-rearing in western Amazonia. The final value derived was
dependent on the size of the area over which the extraction was carried
out, and ranged from US$5–16 ha 1 year 1. Nations (1992) considered the
value of only three plant products from a Guatemalan forest and derived
a value of US$10 ha 1 year 1 and Anderson and Ioris (1992) estimated a
value of US$79 ha 1 year 1 for only three plant species in tropical Brazil.
From the above-cited studies it can be seen that any comparison of the
relative values of forest resources encounters serious difficulties. First,
several of the studies do not consider the full economic value of non-
timber forest products but only of a few selected species. In particular,
none of the studies includes the value of faunal products from the forest.
Second, largely differing values were derived for studies that occurred
over the same areas (for example, Peters et al., 1989; Pinedo-Vasquez et
al., 1992) bringing into question the validity of these studies. In addition,
Nations (1992) and Hecht (1992) consider only the gross value of the
resources they estimated, making comparisons with studies that estimated
net values impossible.
Fewer studies have attempted to value wildlife resources in forest or
savanna ecosystems in Africa. Those early studies that attempted to do so
concentrated mainly on animal rather than plant resources, for example,
Martin (1983), Redford and Robinson (1985) and Barbier et al. (1990).
Scoones et al. (1992) reviewed much of the literature on the use and
value of wild resources in agricultural systems. Although much literature
existed about the use of wild resources, it was shown that almost nothing
was known about the value of these products. One of the few studies to
undertake the economic assessment of wildlife resources is that of the
International Institute for Environment and Development (IIED) and
Hadejia-Nguru Wetlands Conservation Programme (HNWCP) (1997).
The principal method of data collection was Participatory Rural Analysis
(PRA), and the extensive report gives a detailed description of the
methods used to obtain the data; which products were used, how they
were harvested, the quantities taken and their value. An overall economic
value is derived for a number of the wildlife resources based on a simple
cost–benefit analysis of value equals quantity sold times price, minus costs
of production. However, no assessment of the value of crops and livestock
is made in this study.
In addition to wildlife resources, livestock resources are also of high
value to the people living in and around many savanna ecosystems. As was
discussed earlier in this report, the value of livestock has frequently only
been seen in their commercial meat production. This has frequently led to
the belief by policy-makers that traditional production systems of livestock
are backward and inefficient compared with those of commercial systems
such as cattle ranching. In recognition of this shortcoming, research was
initiated to attempt to assess the true productive value of traditional forms
of livestock production systems (for example, Cossins, 1985). Scoones
(1992) recognized that to understand the full value of livestock within
agropastoral systems required a detailed study of all of the useful outputs
from the livestock in the system. He defined productivity not simply in
terms of meat production, which had a marketed value, but also included
non-marketed products such as milk and manure. In addition, services
provided by the livestock were also valued, such as the provision of labour
and transport. Finally, Scoones also included the costs of production of
livestock in the study area, and made a comparison between the relative
benefits of different types of livestock production. This study was carried
out through a simple cost–benefit analysis, that is, the total value of any
animal was calculated as equal to the value or price of it and its products
minus the costs of production. Data on the production parameters of
livestock (for example, births, deaths, milk production, and so on) were
obtained though household surveys. A comparison of livestock productiv-
ity on different soil types was also made to assess the effect of a change in
this ecological parameter. He found that the productivity of communal
cattle was higher than that of beef cattle in the same region and that this
was at least partly due to the higher stocking rate of the communal cattle.
Campbell et al. (1997) estimated values for both plant and animal wild-
life resources in a Zimbabwean savanna. A large number of products that
were used by the local population were identified, for example, firewood,
wood for construction, fruit, birds and thatch. As with Scoones et al.
(1992), the total value of the ecosystem was calculated by multiplying the
amount of each good produced by its value. For many of these resources,
an economic value was imputed by subtracting the costs of production (for
example, labour and transport costs) from their market value. Two differ-
ent methods were used to estimate how much of a good was produced by
each area of land. For products that were regularly used by households,
such as firewood and some wild foods, the amount that could be sustain-
ably produced by the forest was taken to be the amount of the product used
(an assessment of sustainability was also made). However, for other goods,
such as wild fruits, the amount used was judged to be a proportion of the
total annual biological production. Thus, in these cases, an ecological analy-
sis was made of the productivity of each plant species each year. Although
not explicitly stated, in this latter case a simple production function was
therefore set up, where the quantity consumed by households equalled
a proportion of the quantity produced by the vegetation. Although the
analysis made by Campbell et al. was extremely precise and detailed it did
not include a valuation of the livestock and crops produced in the region.
Campbell et al. (2000b) compared the economics of four different cattle
production scenarios in communal grazing lands in Zimbabwe. In this
study, data on inputs to, and outputs from, livestock production, and data
on prices, were obtained through surveys of households. This allowed the
researchers to identify relationships between the inputs into production
of livestock (for example, feeding, herding labour, dipping) and various
outputs (for example, manure, milk, transport). Although not explicitly

stated, this approach again followed a production function approach
where the outputs from the livestock production were directly related to
the inputs. By multiplying the outputs of livestock production by their
price, and subtracting the costs of production, the total profitability of
each management system was calculated.
In addition to the valuation of direct-use resources, some research has
also been directed towards valuing the non-use values of forested and
savanna ecosystems. Since in this case, the value of an ecosystem is meas-
ured in terms of its contribution to the production of other goods or serv-
ices, a relationship (or production function) must be identified between the
two factors. For example, if the loss of watershed protection leads to an
annual decrease in agricultural productivity worth $50, this can be imputed
as the value of the watershed for agricultural production. Kundhlande et
al. (2000) estimated the value of carbon sequestration and of water supply
for a tropical savanna-woodland ecosystem in Zimbabwe. By measuring
the standing biomass of vegetation in the region they calculated the volume
of carbon that could be sequestered by the land when forested. This value
was then compared with an estimate of the decrease in sequestration
capacity that would occur if the land-use were changed to agriculture. The
difference between the two amounts was then multiplied by the estimated
price of carbon to give an overall change in value of carbon sequestration
services. In this sense, the value of carbon sequestration was measured as a
function of the available biomass. In estimating the value of water in crop
production, a crop production function was used that related the amount
of crop produced to the amount of rainfall available. The value of water
was then measured as the change in crop production with changing levels
of rainfall, multiplied by the price of the crop. In addition, although not
explicitly mentioned, similar production functions were also used to define
the relationships between wild foods and grass production and rainfall.
These production functions were derived by a group of biological and agri-
cultural scientists working on the project. In some cases there were docu-
mented relationships, whilst in other cases the relationships were based on
the best judgements of researchers who were familiar with the local agricul-
tural practices (W.L. Adamowicz, personal communication).
This review has shown that many studies have been carried out that
have valued a certain aspect or layer of an ecosystem’s resources. It is
likely that the relative recentness of this field’s development along with
the large amount of resources required to carry out such valuation studies
have so far precluded evaluation of entire ecosystems. However, although
no study has attempted to estimate the total economic value of an eco-
system, comparisons of the values of ecosystems under different forms
of land-use have been made. With the progressive increase in data avail-
ability and a growing understanding of the requirements for making such
studies, it is anticipated that total valuation of different forms of land-use
will become progressively easier.
A method for comparing different ecosystem values: the production function

approach
One of the key shortcomings of many of the current studies has been that
the relative values derived for different forms of land-use have been dif-
ficult, if not impossible, to compare because of the different type of valu-
ations that have been made. However, in cases where a comparison may
be possible, the valuation of each type of resource is made in terms of the
static state of land-use, that is, a forested ecosystem provides a certain
value of resources, whilst a deforested agricultural ecosystem provides
another value. Thus, the value of the ecosystem is dependent on the area of
cover of a particular vegetation type. At first sight, the equitable method
of choosing the use of the ecosystem, based on economic value, may be
the one that produces the highest value of resources. However, economic
theory shows that this is not the case, but that the economic value of the
resource may be maximized through a combination of land-uses. Further
detailed information on the theory behind profit maximization is given in
Common (1996) but a brief outline is given below.
If the value and thus profit of one use of the ecosystem increases with
an increasing area of that type of land-use, profit can be expected to be
maximized when the area of that cover of land type is maximized. The
reverse is also true, in that the profit of another type of land-use will be
maximized when its area is maximized. This can be explained graphically
as is shown in Figure 11.1. The profit of any enterprise is maximized when
the marginal profit (the rate in change of profit with a unit increase in
input) is equal to zero.1 If we take the example of two types of land cover,
forested or deforested, the graph in Figure 11.1 shows that individual
profits for each enterprise will be maximized at C3 and C1 respectively,
where marginal profits are zero. However, if an area of forest cover
shown at C2 is chosen, the total value of the area will be greater than at
either at C3 or C1. This theory is known as a Pareto improvement but
will not be discussed in further detail here. In order to identify this level
of forest cover, a relationship between the total value of each ecosystem
use and levels of vegetation cover must be identified, that is to say that
a common denominator must be identified. The technique behind iden-
tifying a common denominator driving productivity is known as the
production function approach, which has been described in literature on
valuation (see Barbier, 1994).
50
Agricultural land-use
Forested land-use
40
Marginal profit/hectare
30
20
10
0
0 20 C1 40 60 C2 80 100 C3 120
–10
Percentage area cover of forest (ha)
Figure 11.1 Showing the level of marginal profits for different types of
vegetation cover
The following section describes the theory of the production function

approach, and is followed by a worked example of applying this approach
to the valuation of savanna and forest resources.
The production function approach to valuing ecosystems

A production function is a mathematical description of the relationship
between the production of a good (for example, sheep, maize), and the
amount of inputs (for example, environmental or human-made) required for
its production. For example, a production function for clay pots (P) might
be determined by the amount of clay (C), water (W) and people (L) avail-
able to make them, where the way in which these factors interact defines the
mathematical form of the function. The development of a production func-
tion provides a quantitative description of a relationship between several
factors where previously the specific relationship was unknown.
Although initially used in economic analysis to describe the production
of goods by firms, the use of this methodology has subsequently been
expanded to include measures of environmental factors, and can be used
to describe the production of any good where the inputs into its produc-
tion can be quantified.
Direct-use values
Ecosystems provide many important functions for humankind, which, as
has already been described, can be grouped according to their direct-use
or indirect-use values.2 The valuation of an ecosystem is frequently made
according to its tangible outputs (direct-use values), for example, the value
of the crop harvested, and where no indirect-use value of the ecosystem
exists, this valuation is perfectly acceptable. A production function can
then be developed to quantify the components of an ecosystem that con-
tribute towards the production of a good. In doing this, an estimate can be
made of the relative values of each of the inputs into production.
If a comparison is being made of the relative values of different forms of
ecosystem use, it is imperative that all components of the production func-
tion are included in the analysis. For example, the production of livestock
and/or game in a particular area (Q), will require inputs of labour (L) and
forage (F ). However, as was shown by Scoones (1992), the productivity of
livestock may also be dependent on factors such as the physical conditions
of the environment. Thus, in the case of Scoones’ research, livestock pro-
ductivity was also affected by soil type. If this latter factor is overlooked and
excluded from the livestock production function, the result will be an under-
valuation of the soil value (S), giving it the incorrect production function:
Q 5 f (L, F)
rather than:
Q 5 f (L, F, S) .
If this production function were defined incorrectly in this way, no value

would be attributed to the presence of a specific type of soil cover.
In addition, Scoones (1992) showed that the output of livestock
amounted to much more than simply its meat production. Included in the
values of livestock in his study were milk production, manure production
and the provision of labour. By including these output variables the value
of livestock significantly increased. It can therefore be seen that care must
be taken to ensure that all inputs into production are incorporated in a
production function, and that all the outputs from an activity are also
included, if an accurate valuation of the input resources is to be obtained.
Indirect-use values
In agriculture and forestry, which involve the primary production of
goods, in addition to providing direct-use goods, in many cases the eco-
system may provide an ecological function that has an indirect value in
supporting the production of marketable goods. The ecological function
is then a ‘factor input’ in the production process. As already discussed, the
ecological function may have an important input for activities that are far
removed, either spatially or temporally, from the production process, and
thus the benefits derived may not be easily ascribed to it. For example,
deforestation may result in increased erosion, leading to sediment deposi-
tion on downstream farmland, potentially affecting the nutritional content
of the deforested soil and crop production on the downstream farmland.
Thus, in addition to its value for timber production, the forested land may
also have a direct influence on crop and livestock/game production in the
immediate area and crop production in a different area. Including forest
area as a determinant of livestock/game or crop production may therefore
capture some element of the economic contribution of this ecological
function. In mathematical terms this can be described as:
Q 5 f (xi. . . . .xn, A)
where Q can be considered as the output or marketed good, for example,

crop production, xi. . . . xn are marketable inputs required to produce that
good, for example, labour, seed and so on, and A is the environmental
input of interest required for the production of the good, for example, the
area of forested land.
In attempting to value these environmental benefits, two underlying
relationships must be elucidated. First, the physical effects of changes in
a biological resource or ecological function on an economic activity have
to be ascertained. That is, an accurate understanding and quantification
of the change in the output of a good with a change in land-use patterns
must be known. To use our previous example, which other ecosystems
are affected by the deforestation of an area and what are these effects?
Knowledge of these processes requires detailed ecological understanding
of the systems involved. Second, the effect of these environmental changes
must be valued by considering the corresponding change in the output
of the marketed good, for example, what is the quantitative change in
downstream crop productivity and thus farmers’ incomes in response
to the deforestation of an upstream forest? Again this requires detailed
ecological and sociological understanding of the processes in place. In this
way, the biological resource is considered as an input into an economic
activity, and can be valued in terms of the impact it has on the output of
a marketed good.
Limitations to the use of the production function approach: the absence of a

monetary value of the output good
When a product, Q, such as crop production, is measurable and a market
price is available for it or one can be imputed, determining the values of
the environmental inputs (A) into its production is relatively simple.3 If
however, Q can not be measured directly, then a measure of the change in
value of Q with a change in the environmental input, A, cannot be directly

obtained. This is often the case in developing countries where the use of
the environment is to support a subsistence lifestyle. Since the products do
not pass through a market, the benefits accrued do not have a monetary
value. In these cases, an accurate comparison in economic terms between
different forms of the land-use is difficult to make, given the absence of
comparable systems. To overcome this problem, the output of the ecosys-
tem under different uses must be considered in common terms. The stand-
ard approach to the valuation of products is in monetary terms and so if
goods produced by an ecosystem do not have a monetary value, a proxy
must be found for this value. In the case of land-use in Tanzania, land-
users include nomads who utilize the natural resources (Q), such as vegeta-
tion and game, for their own consumption to sustain a subsistence-level
lifestyle, and who do not generally trade goods in the marketplace. Many
of these products may have no market value, so to obtain an approxima-
tion of their ‘worth’, an estimate can be made of the market value of an
‘equivalent’ good.
In general terms, if this solution is unfeasible, the alternative is to find a
substitute or complementary resource between S and one of the marketed
inputs x1. . .xn. Both this and the previous solution require a detailed
knowledge of the ecological interactions between inputs and outputs of
goods. It can be realized however, that in a system such as nomadism, this
latter solution is implausible since there are no marketed inputs into the
production of the goods.
The following section will illustrate, through the use of examples, the
employment of the production function approach in estimating resource
values for forest and/or savanna ecosystems.
Worked examples
Direct-use values: calculating the direct-use value of agricultural cover

The direct-use value assessed here is the amount of land available for live-
stock production and for wild fruit production. A hypothetical situation is
set up whereby a fixed area of land can either be used to produce livestock
(if it is deforested) or wild fruit (if it is forested). A production function is
estimated for both products based partially on the area of land available
for their production, and the optimal level of land cover of both types of
land is identified, in terms of profit maximization.
Step 1: Identifying the production function for livestock The amount of

livestock produced is (partially) dependent on the amount of land avail-
able for grazing. It will also be dependent on other factors such as the
amount of labour used in herding and veterinary costs. This could, for
example, be represented as:
Ql 5 Aas Lbl (11.1)
Where:
Ql 5 the amount of livestock produced (output in kg per year);

As 5 the area of land available for livestock production (hectares);
Ll 5 labour (person hours per year);
a, b 5 unknown parameter values to be estimated.
In this example, primary data would be required for the input and output
variables given in Equation 11.1. These data could be obtained through a
questionnaire survey of the local population answering the following type
of questions:
Livestock production data:
1. Do you keep livestock?

2. What type of livestock do you rear: sheep, cattle, goats, poultry?
3. What was the calving/lambing, and so on, rate of this livestock in the
past year?
4. What was the mortality rate of your livestock in the past year?
5. What are the uses of your livestock other than for meat production,
for example, milk, manure?
6. How much of each of these products was produced?
7. How many people were involved in caring for your livestock and for
how many hours did each work?
It should be noted that this illustration gives an example of the kinds of

questions that should be asked in a questionnaire survey but not of the
form the questions should take. To elucidate the correct information,
questionnaire surveys must be presented in an objective manner and in a
way that can be understood by the recipients of the questionnaire.
A simplified example of the results of such a study is given in Table
11.1 using only two inputs, labour and land area, and three study areas.
To make a comparison over time, two different approaches can be taken:
either the survey can be longitudinal, that is, carried out over a number of
years, or cross-sectional, that is, carried out at the same time but over a
number of different study sites. This will then provide variation in the data
of the inputs into production.
Table 11.1 Hypothetical results from agricultural questionnaire survey on

livestock production
Input into Livestock Amount of Inputs Amount of Livestock

Production Per Year Produced (kg)
Labour (hours) 200 63
Area (ha) 20
Area (ha) 20
Area (ha) 50
To identify the quantitative relationship between these variables, a

multiple regression analysis is carried out using the quantity of livestock
produced as the dependent variable and the input data as the independent
variables. The results of the multiple regression for this example are given
in Equation 11.2.
Ql 5 L0.5
l As
0.5
R2 5 1 (11.2)
The results could, of course, take an alternative form of production such

as a linear relationship shown below if the data fitted the multiple regres-
sion more accurately in this form.
Ql 5 4.5Ll 1 2As
Clear and accessible information on the choice and understanding of dif-

ferent functional forms can be found in Hill et al. (1997).
Step 2: Calculating the value of livestock production Once the production

function has been identified, the value of the ecosystem can be calculated
by estimating the value of the livestock products. Data are therefore
required on the price of livestock products and on the costs of production.
Again these data would be collected through a questionnaire survey, either
of the householders involved in production or through other individuals
who have knowledge of the markets.
Livestock costs data:
1. What was the price of the livestock that was sold?

2. How much of the livestock was retained for home consumption?
3. How many hours were spent herding the livestock?

4. What was the cost of this labour?
5. Were there any other costs in involved in rearing the livestock? What
were they?
Once again, it should be noted that this illustration gives an example of the
kinds of questions that should be asked in a questionnaire survey, but does
not give the form that the questions should take.
With this information, a profit function can be written:
Pl 5 PlQl 2 Cl (11.3)
Where:
Pl 5 total profit of livestock($);

Pl 5 price of livestock ($/kg);
Ql 5 quantity of livestock produced (kg);
Cl 5 costs of production of livestock ($).
Substituting Equation 11.2 into Equation 11.3 gives:
Pl 5 Rl (L0.5
l As ) 2 Cl
0.5
(11.4)
The total profit from livestock production can then be identified for any
given levels of land and labour.
Thus, if Pl 5 10, Ll 5 200, As 5 20 and Cl 5 500:
Pl 5 (10 * 2000.5 * 200.5) 2 500

5 $126 annum21 (11.5)
Step 3: Identifying the production function for forest goods In exactly the
same manner as described for the production function of livestock, a pro-
duction function of forest products can also be identified.
The amount of a good, for example, wild fruit, produced by the forest
will be partially dependent on the area of forest. It will also be dependent on
other factors such as the amount of time spent in collecting it. The methodol-
ogy required would be the same as that for identifying the livestock produc-
tion function in that data would be obtained through a questionnaire survey
of the local population. Therefore further details are not given here. In
Equation 11.6, an example is given of a possible fruit production function:
Qf 5 2 (At 2 A0.2
s ) 1 3Lf
0.7
(11.6)
Where:
Qf 5 the amount of fruit collected (kg per year);

At 5 total area of study region (hectares);
Lf 5 labour for collecting fruit (person hours per year).
Note that the area of land is given as the total area of the study area minus
the area of land used for livestock production, that is, the area of forested
land.
Step 4: Calculating the value of the fruit production As with the livestock
profit function, a fruit profit function can be estimated through identi-
fying the price of the fruit and the costs of its production. This would
give a profit function such as that shown for livestock in Equation 11.3.
Substituting Equation 11.6 into this will then give the total profit func-
tion, as shown in Equation 11.7. An example of the fruit profit function
is given in Equation 11.8 for values of Pf 5 5, Lf 5 200, At 5 50 and Cf
5 25.
Pf 5 Pf (2 (At 2 As) 0.2 1 3L0.7

f ) 2 Cf (11.7)
Where:
Pf 5 total profit of fruit production ($);

Pf 5 price of fruit ($/kg);
Cf 5 costs of production of fruit ($).
Pf 5 (5 * 2 ((50 2 20) 0.2) 1 (3 * 2000.7) 2 25

5 $116 annum21 (11.8)
Step 5: Identifying the total value of produce from the region and the optimal
level of different vegetation covers To identify the total value of the area,
the values of all the forms of production must be summed together. In this
worked example, these values are simply the value from livestock produc-
tion and the value from wild fruit harvesting. Thus, the total value of the
region would equal the sum of Equations 11.5 and 11.8, given algebrai-
cally in Equation 11.9 and numerically in Equation 11.10.
Ptotal 5 Pf (2 (At 2 As) 0.2 1 3L0.7

f ) 2 Cf 1 Pl (Ll As ) 2 Cl (11.9)
0.5 0.5
Ptotal 5 (5*2* (50 2 20) 0.2) 1 (3*2000.7) 2 25 1 (10*2000.5 * 200.5) 2 500

5 $242 annum21 (11.10)
Following conventional theory on optimization (see, for example,

Nicholson, 1992), the optimal level of agricultural land area (As) and thus
also forested land area, to give the maximum total value of the region
(Ptotal) can be found by taking the partial derivative of the total profit
function with respect to land area and setting the value to zero.4 This is
shown algebraically and mathematically in Equations 11.11 and 11.12
respectively.
0Pt
5 Pf (2* 2 0.2 * (At 2 A20.8 ) 1 0.5PlL0.5
l As
20.5
50 (11.11)
0A s
0Pt
5 5 (20.4 (50 2 A20.8 ) 1 0.5 (10 * 2000.5A20.5 ) 50
0A s s
As 5 1.0 (11.12)
Thus, for a total study area of 50 hectares, the optimal area of agricul-
tural land is one hectare and hence the optimal area of forested land is 49
hectares.
It is obvious that for reasons of clarity and simplicity, this worked
example has been chosen with only a few variables. However, in carry-
ing out the complete research, two critical points must be noted. (1) To
identify the total value of the land, all of the products from it must be
identified and assessed. Summing the total profit from all of these prod-
ucts would then give the total value of the region. For the sake of brevity,
in this worked example only one product from each type of land-use is
given, however the method is equally applicable with numerous products
from each type of land-use. (2) Acquiring the data outlined here requires
a detailed socioeconomic knowledge of the population living within the
region. It must be noted that although the questions outlined here for
obtaining information on the production of goods give a guideline as to
the type of information required, they do not precisely illustrate the meth-
odology for obtaining that information. Much of this information may
not be directly obvious but may have to be imputed through other means.
For example, to obtain an estimation of the price of wild foods that are
not marketed will require more than a simple question. The value may be
able to be imputed, for example, by comparison with the value of other
goods that are marketed. The reader is referred to the other valuation
studies referenced in this report for more detailed descriptions of survey
techniques (for example, Scoones, 1992; Campbell et al., 1997; IIED and
HNWCP, 1997).
Indirect-use values: calculating the indirect-use value of rainfall

To measure the value of an indirect-use good or resource, economists use
changes in the productivity of a direct-use good as an indicator. A produc-
tivity change simply examines the physical relationship between the eco-
system and private goods for which values are known or can be imputed.
For example, in the case of agricultural production, the economic value of
a change in the supply of an environmental resource will be the difference
between the profit made from agricultural production before and after the
environmental change. It is important to note that this method does not
provide the total value of the indirect-use resource, but rather the marginal
value. That is, the change in the value of the good produced, for example,
a crop, for a change in the amount of input of the indirect-use resource.
The following example shows how this evaluation can be carried out.
By a very similar method as described for estimating direct-use values
of ecosystems, indirect-use values can also be measured. The indirect-
use value assessed here is the amount of water present in the ecosystem,
measured as annual rainfall. It is known that the presence of woodland
and forest vegetation in a region maintains rainfall patterns and the local
climate (for example, Myers, 1995). Thus it can be assumed that changing
the area of vegetation cover in a region will have an impact on the rainfall
patterns in that area.
Step 1: Identifying the production function The amount of a crop pro-

duced is (partially) dependent on the amount of water available through
rainfall. It will also be dependent on other factors such as the amount of
labour and fertilizer used.
This could, for example, be represented as:
Q 5 Aa WbLcFdSe (11.13)
Where:
Q 5 the amount of crop produced (output in kg per year);

A 5 the area of land used (hectares);
W 5 average annual rainfall (millimetres of rainfall per year);
L 5 labour (person hours per year);
F 5 fertilizer (kg applied per year);
S 5 amount of seed sown (kg per year);
a, b, c, d, e 5 unknown parameter values to be estimated.
Estimates have already been made of the quantitative relationship

between crop production and water for certain African dry tropical
ecosystems (Coe et al., 1976; Kundhlande, 2000). However, if the analysis

is to be accurate, it is essential that this relationship is estimated as pre-
cisely as possible. Data may therefore be required to estimate this relation-
ship for a particular ecosystem or study area.
In this example, primary data would therefore be required of the input
and output variables given in Equation 11.13. With the exception of the
data on rainfall, these data could be obtained through a questionnaire
survey of the local population answering the following type of questions:
Crop production data:
1. Do you own or rent land?

2. What is the total size of your land holding?
3. What do you produce on your land?
4. What was the total amount you produced of each crop last year?
5. What is the total amount of seed you sowed last year for each crop?
6. How many people were involved in caring for that crop and for how
many hours did each work?
7. Did you apply fertilizer to any of your land? If so, what type of ferti-
lizer, to which crops, and how much?
8. Did you irrigate your crops with groundwater? If so, how much water
was used?
It should be noted that again this illustration gives an example of the kinds
of questions that should be asked in a questionnaire survey, but not of the
form the questions should take.
In addition to this data, the level of rainfall must be known for the study
area. To make a comparison over time, two different approaches can be
taken: either the survey can be longitudinal, that is, carried out over a
number of years, or cross-sectional, that is, carried out at the same time
but over a number of different study sites. This will then provide variation
in the data of rainfall and the other inputs into production. A simplified
example of the results of such a study is given in Table 11.2 using only
three inputs, rainfall, labour and land area, and three study areas.
To identify the quantitative relationship between these variables, a
multiple regression analysis can be carried out using the quantity of crop
produced as the dependent variable and the input data as the independent
variables. The results of the multiple regression for this example are given
in Equation 11.14.
Q 5 W0.7L0.2A0.1 R2 5 1 (11.14)
Table 11.2 Hypothetical results from agricultural questionnaire survey
Input into Crop Amount of Input Amount of Crop

Production Used Per Year Output (kg)
Rainfall (mm) 1000 898
Labour (hours) 1500
Area (ha) 20

Labour (hours) 800
Area (ha) 20

Labour (hours) 1000
Area (ha) 50
Step 2: Calculating the value of the ecosystem in terms of water produc-

tion Once the production function has been identified, the value of the
ecosystem can be calculated by estimating the value of the agricultural
products. Data are therefore required on the price of agricultural prod-
ucts, and on the costs of production. Again these data would be collected
through a questionnaire survey, either of the householders involved in pro-
duction or through other individuals who have knowledge of the markets.
Crop costs data:
1. What was the price of the crops that were sold?

2. How much of the crops were retained for home consumption?
3. What was the price of labour and other inputs into production, for
example, fertilizer?
With this information, a profit function can be written:
P 5 PCQ 2 C (11.15)
Where:
P 5 total profit ($);
PC 5 price of crop ($/kg);
Q 5 quantity of crop produced (kg);
C 5 costs of production of crop ($).
Substituting Equation 11.14 into Equation 11.15 gives:
P 5 RC (W 0.7L0.2A0.1) 2 C (11.16)
The change in the value of crop production for changing levels of rainfall
can be identified for any given levels of rainfall, labour, area and costs. It
is important to note that these other variables must be held constant and
can be chosen to be at any level.
Thus, if P 5 10, W 5 1000, L 5 800, A 5 50 and C 5 5000:
P 5 (10 * 10000.7 * 8000.2 * 500.1) 2 5000

5 $2087 annum21
But reducing the rainfall level by half to W 5 500 mm per annum gives:
P 5 (10 * 5000.7 * 8000.2 * 500.1) 2 5000

5 $ 2 636 annum21
Thus, the value of rainfall changing from 1000 mm per annum to 500 mm
per annum would be estimated as $−2723 ($−636 − $2087).
This example does not quantify a relationship between the amount of
forest cover in a region and the changes in crop productivity but simply
quantifies the relationship between rainfall levels and crop production.
Including the valuation of the area of forest cover in the region would
require a second production defining the relationship between forest cover
and average annual rainfall. If this relationship is known, it can be simply
incorporated into the original production function. For example, if:
W 5 F 0.5 (11.17)
Where:
W 5 average annual rainfall (millimetres of rainfall per year);

F 5 area of region with forest cover (hectares).
Substituting Equation 11.17 into Equation 11.16 will give Equation 11.18.
Exactly the same calculations can then be carried out as before but with
changing forest cover values rather than rainfall levels to provide a value of
the change in crop production for a change in forest levels in the region.
P 5 PC [ (F 0.5) 0.7L0.2A0.1 ] 2 C (11.18)
The above example demonstrates the technique of using a production

function approach to value indirect-use resources of an ecosystem. In this
example, the value of water on crop production is used; however, crop

production is only one of a number of factors that will be affected by the
level of rainfall in a region. To make a complete assessment of rainfall
value or forest cover value in the region, all factors affected by water avail-
ability should be assessed in the same manner. These could include live-
stock production, wild food production, fuel wood production and human
health (Kundhlande et al., 2000).
Conclusions
Any research project that aims to undertake work in evaluating different
land-use systems would be required to consider several different areas
of research. The research would require the integration of ecological,
sociological and economic skills in order to gain a full understanding of
the problem. To assess the total economic productivity of the ecosystems
of Tanzania under different forms of land-use, for example, nomadic
systems, sedentary agricultural systems, ranching and commercial hunting/
tourism, each system of land-use must be considered in turn, and an esti-
mate made of the total economic value of that system. As described in
previous sections of this report, total economic value comprises the direct-
use and indirect-use values of the ecosystem. Therefore an analysis must
incorporate an estimate of both types of value.
To assess the primary products utilized by individuals from the eco-
system, that is, the direct-use value of the system, studies will be required
to estimate the off-take from the environment of natural products, both
vegetation and animal, as well as the number of people sustained by this
system (population density). This can be achieved through, for example,
sociological studies that directly measure the quantity and type of natural
resources used. The productivity of agricultural goods can also be meas-
ured in this manner.
To estimate the value of products that are not marketed, a market value
must be assigned to them. This can be through comparison with values of
similar resources that have a market value or, alternatively, an estimate
may be made of the compensation that individuals would have to receive
in order to provide themselves with similar products to the ones pro-
vided by their ecosystem. For systems of land-use where the products are
marketed a direct estimate can be made of their economic value.
To estimate the indirect-use values of the ecosystem, an understanding
of the role of ecological support functions of an ecosystem must be gained
through, for example, ecological, geographical or hydrological studies.
With this knowledge, a production function approach, as outlined earlier,
can then be used to translate these ecological relationships into economic
values.
Achieving a complete economic valuation of an ecosystem requires a

substantial amount of research. However, even if complete valuation is
not initially achievable, an estimation of some of the indirect values of
ecosystems may be sufficient to allow a comparison of different land-uses.
Optimization techniques can then be used to estimate the socially optimal
level of various natural resources required to maximize social welfare.
Notes
1. For a detailed explanation of the mathematics of profit maximization, see Dowling
(1992).
2. Non-use values have also been identified but are not considered in this case. Estimation
of their values is extremely difficult and not deemed relevant here. See Heywood (1995)
for further information.
3. The mathematics of the problem-solving are not given here but are relatively simple to
carry out.
4. Further details of how to carry out partial differentiation can be found in Dowling
(1992).
References
Anderson, A.B. and E.M. Ioris (1992), ‘The logic of extraction: resource management and
income generation by extractive producers in the Amazonian estuary’, in K.H. Redford
and C. Padoch (eds), Conservation of Neotropical Forests: Working from Traditional
Resource Use, New York: Colombia University Press, pp. 175–99.
Anderson, D. (1987), Economic Aspects of Deforestation: A Case Study in Africa, Washington,
DC: The Johns Hopkins University Press.
Ataroff, V. and F. Rada (2000), ‘Deforestation impact on water dynamics in a Venezuelan
Andean cloud forest’, Ambio, 29(7), 440–44.
Barbier, E.B. (1994), ‘Valuing environmental functions: tropical wetlands’, Land Economics,
70, 155–73.
Barbier, E., J. Burgess, T. Swanson and D. Pearce (1990), Elephants, Economics and Ivory,
London: Earthscan Publications.
Barreto, P., P. Amaral, E. Vidal and C. Uhl (1998), ‘Costs and benefits of forest management
for timber production in eastern Amazonia’, Forest Ecology and Management, 108(1/2),
9–26.
Bembridge, T. and J. Steenkamp (1976), ‘An agroeconomic investigation of beef produc-
tion in the Matabeleland Midlands provinces of Rhodesia’, Rhodesian Agriculture, 73(2),
100–112.
Campbell, B.M., M. Luckert and I. Scoones (1997), ‘Local-level valuation of Savanna
resources: a case study from Zimbabwe’, Economic Botany, 51(1), 59–77.
Campbell, B.M., R. Costanza and M. van den Belt (2000a), ‘Special section: Land-use
options in dry tropical woodland ecosystems in Zimbabwe: Introduction, overview and
synthesis’, Ecological Economics, 33(3), 341–51.
Campbell, B.M., D. Dore, M. Luckert, B. Mukamuri and J. Gambiza (2000b), ‘Economic
comparisons of livestock production in communal grazing lands in Zimbabwe’, Ecological
Economics, 33(3), 413–38.
Coe, M.J., D.M. Cumming and J. Phillipson (1976), ‘Biomass and production of large
African herbivores in relation to rainfall and primary production’ [cited in Campbell et al.,
2000a], Ecology, 22(4), 341–54.
Common, M.S. (1996), Environmental and Resource Economics: An Introduction, London:
Longman.
Cossins, N. (1985), ‘The productivity and potential of pastoral systems’, ILCA Bulletin, No.
21, 10–15.
Dixon, J.A. and P.B. Sherman (1990), Economics of Protected Areas. A New Look at Benefits
and Costs, London: Earthscan Publications Ltd.
Dowling, E.T. (1992), Introduction to Mathematical Economics, New York: McGraw-Hill.
Fuentes, E. (1980), ‘Los Yanomami y las plantas silvestres’ (The Yanomami and wild plants),
Antropologica, 54(3), 3–138.
Godoy, R. (1992), ‘Some organizing principles in the valuation of tropical forests’, Forest
Ecology and Management, 50(1/2), 171–80.
Godoy, R., R. Lubowski and A. Markandya (1993), ‘A method for the economic valuation
of non-timber tropical forest products’, Economic Botany, 47(3), 220–33.
Hanemann, W. (1988), ‘Economics and the preservation of biodiversity’, in E.O. Wilson and
F.M. Peters (eds), Biodiversity, Washington, DC: National Academy Press, pp. 193–9.
Harmon, M.E. (2001), ‘Carbon sequestration in forests – addressing the scale question’,
Journal of Forestry, 99(4), 24–9.
Hecht, S. (1992), ‘Valuing land-use in Amazonia: colonist agriculture, cattle and petty extrac-
tion in comparative perspective’, in K.H. Redford and C. Padoch (eds), Conservation
of Neotropical Forests: Working from Traditional Resource Use, New York: Colombia
University Press, pp. 379–99.
Heywood, V.H. (1995), Global Biodiversity Assessment, New York: Cambridge University
Press.
Hill, C., W. Griffiths and G. Judge (1997), Undergraduate Econometrics, New York: John
Wiley and Sons.
Huntoon, P.W. (1992), ‘Hydrogeologic characteristics and deforestation of the stone forest
karst aquifers of South China’, Ground Water, 30(2), 167–76.
IIED and HNWCP (1997), ‘Local-level assessment of the economic importance of wild
resources in the Hadejia-Nguru wetlands, Nigeria’, Sustainable Agricultural Programme
Research Series, 3(3).
Kirby, D.K. (2000), ‘An ecological economic approach to upland heather moorland manage-
ment’, DPhil thesis, York: University of York.
Kundhlande, G., W.L. Adamowicz and I. Mapaure (2000), ‘Valuing ecological services
in a savanna ecosystem: a case study from Zimbabwe’, Ecological Economics, 33(3),
401–12.
Lal, R. (1997), ‘Deforestation effects on soil degradation and rehabilitation in western
Nigeria. IV. Hydrology and water quality’, Land Degradation and Development, 8(2),
95–126.
Martin, G. (1983), ‘Bushmeat in Nigeria as a natural resource with environmental implica-
tions’, Environmental Conservation, 10(2), 125–32.
Myers, N. (1995), ‘The world’s forests – need for a policy appraisal’, Science, 268(5212),
823–4.
Nations, J.D. (1992), ‘Xateros, chicleros and pimenteros: harvesting renewable tropical forest
resources in the Guatemalan Petén’, in K.H. Redford and C. Padoch (eds), Conservation
of Neotropical Forests: Working from Traditional Resource Use, New York: Colombia
University Press, pp. 208–19.
Niamir-Fuller, M. (1998), ‘The resilience of pastoral herding in Sahelian Africa’, in F. Berkes,
C. Folke and J. Colding (eds), Linking Social and Ecological Systems, Cambridge, UK and
New York: Cambridge University Press.
Nicholson, W. (1992), Microeconomic Theory, Orlando: Dryden Press.
Oommachan, M. and S.K. Masih (1988), ‘Multifarious uses of plants by the forest tribals of
Madhya Pradesh: wild edible plants’, Journal of Tropical Forestry, 4(2), 163–9.
Panayotou, T. (1990), ‘The economics of environmental degradation. Problems, causes
and responses’, Development and Discussion Paper No. 335, Harvard Institute for
International Development, Cambridge, MA.
Pautsch, G.R., L.A. Kurkalova, B.A. Babcock and C.L. Kling (2001), ‘The efficiency of
sequestering carbon in agricultural soils’, Contemporary Economic Policy, 19(2), 123–34.
Peters, C.M., A.H. Gentry and R.O. Mendelsohn (1989), ‘Valuation of an Amazonian
rainforest’, Nature, 339(6227), 655–6.
Phillips, O., A.H. Gentry, C. Reynel, P. Wilkin and C. Galvezdurand (1994), ‘Quantitative
ethnobotany and Amazonian conservation’, Conservation Biology, 8(1), 225–48.
Pinedo-Vasquez, M., D. Zarin and P. Jipp (1992), ‘Economic returns from forest conversion
in the Peruvian Amazon’, Ecological Economics, 6(2), 163–73.
Putuhena, W.M. and I. Cordery (2000), ‘Some hydrological effects of changing forest cover
from eucalypts to Pinus radiata’, Agricultural and Forest Meteorology, 100(1), 59–72.
Rawat, J.S. and M.S. Rawat (1994), ‘Accelerated erosion and denudation in the Nana-
Kosi watershed, Central Himalaya, India. 1. Sediment load’, Mountain Research and
Development, 14(1), 25–38.
Redford, K.H. and J. Robinson (1985), ‘Hunting by indigenous peoples and conservation of
their game species’, Cultural Survival, 8(4), 41–4.
Scoones, I. (1992), ‘The economic value of livestock in the communal areas of southern
Zimbabwe’, Agricultural Systems, 39(4), 339–59.
Scoones, I., M. Melnyk and J.N. Pretty (1992), The Hidden Harvest: Wild Foods and
Agricultural Systems, Sustainable Agricultural Programme, London: International
Institute for Environment and Development.
Swanson, T. and E. Barbier (1991), Economics for the Wilds. Wildlife, Wildlands, Diversity
and Development, London, Earthscan Publications Ltd.
Toledo, V.M., A.I. Batis, R. Becerra, E. Martinez and C.H. Ramos (1995), ‘The useful forest
– quantitative ethnobotany of the indigenous groups of the humid tropics of Mexico’,
Interciencia, 20(4), 177–87.
Uri, N.D. (2001), ‘The potential impact of conservation practices in US agriculture on global
climate change’, Journal of Sustainable Agriculture, 18(1), 109–31.
Wilson, E.O. (1994), The Diversity of Life, London: Penguin.
12. Economic growth and the environment
Dalia El-Demellawy
Introduction
This chapter critically reviews empirical literature on the relationship
between economic growth and the environment. It starts with outlining
the debate around the relationship and then presents basic models used in
the empirical studies. Finally, the various empirical studies are discussed.
Study of the relationship between economic growth and the environ-
ment goes back to the 1960s. It started in developed countries when
considerable concern arose regarding the impact that economic develop-
ment was having on their environment. This growing concern led to the
1972 United Nations World Conference on the Human Environment in
Stockholm. This conference highlighted the conflict of interest between
developed and developing countries regarding the relationship between
economic growth and the environment. The developing countries were
more concerned with development at the expense of the environment.
In other words, slowing economic development to protect the environ-
ment was not appreciated. Even in developed countries there was con-
flict of interest among different groups in society, with workers ranking
improvements of their standard of living associated with technological
advancements above environmental concerns (Beckerman, 1992). Despite
conflict of interest, concerns about the natural environment and its link to
problems of economic development were explicitly stated. It was argued
that exhaustion of the environmental resource base, in terms of minerals
and food production, would provide limits for future economic growth.
However, it was not thought that economic collapse was inevitable. On
the contrary, it was concluded that the world economic system could be
sustainable into the future provided that radical changes in the way it was
being run were adopted (Meadows et al., 1972).
Awareness of the relationship between the environment and economic
growth gained prominence in 1987 for a number of reasons. First, the
widespread environmental and economic problems in the world; particu-
larly high levels of poverty and famines in various countries. Second, the
urgent and complex problem of the national and global scale of environ-
mental pollution; especially that associated with climate change due to
increasing emissions of greenhouse gases (GHGs) causing global warming.
This, in turn, was considered to lead to serious unfavourable effects on (1)
359
agriculture, with output falling as the interior of most continents became

drier, (2) sea levels rising and flooding coastal communities, (3) other
effects, such as greater need for the use of air-conditioning, loss of forests
and possible increase in the frequency of storms. Third, the serious local
environmental problems in developing countries; these problems included
supply of safe drinking water, access to decent sanitation, deterioration of
air quality and urban degradation, which had damaging effects on health
and human welfare (Brundtland, 1987 and IBRD, 1992).
It was recognized that alleviation of poverty would not be accomplished
by ceasing economic development or by redistribution of wealth from
rich to poor countries, but by reviving economic growth in developing
countries. Also, due to the interdependence of the world economy, the
prospects of the developed countries would depend on the economic
growth of the developing ones. Finally, economic growth had to be based
on policies that would sustain the resource base of the biosphere and such
growth would alleviate the problems of poverty and underdevelopment
(Brundtland, 1987).
The economic-growth–environment relationship is one of the promi-
nent debates of the international community at the present time. At one
extreme, some social and physical scientists argue that, despite the rise
in income, higher economic growth exhausts natural resources, causes
higher accumulation of waste and increases pollution (Meadows et al.,
1972). This, in turn, results in environmental degradation and thereby,
decline in a human welfare. At the other extreme, some social and eco-
nomic environmentalists argue that higher income increases the demand
for better environmental quality (Beckerman, 1992). In other words, eco-
nomic growth and environmental improvements follow the same path. In
between these two views, some economists and environmentalists argue
that the relationship between economic growth and the environment is not
fixed over time. They argue that environmental quality deteriorates with
economic growth until it reaches a certain threshold, after which improve-
ments in environmental quality take place. In other words, the relationship
follows an inverted U-shaped path, known as the Environmental Kuznets
Curve – EKC (Kuznets, 1955; Grossman and Krueger, 1991; Shafik and
Bandyopadhyay, 1992; Panayotou, 1993; Selden and Song, 1994).
Determining the relationship between economic growth and the envi-
ronment is crucial as it has serious policy implications. A positive
relationship between economic growth and environmental degradation
requires strict environmental policies in order to limit economic growth
to a level within the carrying capacity of the environmental base (Arrow
et al., 1995). On the other hand, a negative relationship requires policies
that accelerate economic growth, rather than restrictive environmental
Economic growth and the environment 361
policies, in order to achieve environmental improvements (Beckerman,

1992). However, if the relationship is positive at lower levels of income and
then turns negative at higher levels of income, then policies that acceler-
ate economic development, although improving environmental quality in
the long run, will have serious damaging effects on the environment in the
short and medium run. Therefore, caution is necessary in introducing the
right policies at the different stages of economic development without sac-
rificing environmental improvement at any level. Economic development
policies should be set and viewed in relation to their effects on the environ-
ment. Both economic and environmental policies should complement each
other. In fact, good environmental protection policies will help and sustain
economic development. However, promoting economic growth without
taking into account the impact on the environment can bring economic
development to a halt (IBRD, 1992).
In that respect, the enactment and enforcement of international legisla-
tion by the international community is vital in controlling and minimizing
the problems resulting from economic development and its damaging
effect on the environment for a number of reasons. First, they tend to
moderate the growth rates of pollution associated with economic growth
by encouraging firms and/or consumers to use less polluting technologies.
This is essential because individuals and firms are interested in maximizing
profits, and therefore have few incentives in curbing pollution, such as air
emissions in urban centres, dumping wastes in public waters or the severe
overuse of land (Congleton, 1992). Moreover, the gains from a better
environment, such as improved human health, higher economic produc-
tivity and enhanced amenities, are sometimes difficult to measure com-
pounded with the lengthy time lag between taking the proper actions and
realizing the effects of environmental improvements (Goodstein, 1999).
Furthermore, individuals and firms can make serious trade-offs in their
decision-making due to the lack of consideration of the earth’s limited
resource capacity and pollution to the environment, which, in turn, will
impede sustainable development. This has an intergenerational effect as it
results in compromising the average standard of living of future generations
in comparison with the present one (Beckerman, 1992). This is also known
as the intergenerational displacement of environmental costs, where costs
are transferred across a great distance or to a remote future (Roca, 2003).
Second, the costs of environmental protection are high and are often
unaffordable by poorer countries. Accordingly, the international com-
munity sets policies whereby rich countries bear a reasonable share of the
costs of improving the environment in poor countries. This is due to the
fact that the benefits from a cleaner environment, such as the protection of
tropical forests and biodiversity, will accrue not only to the poor countries
but to the rich ones as well. Also, some of the problems facing the world,
such as global warming and ozone depletion, stem from high consumption
in the richer countries. Moreover, the international community allows
and facilitates the access to cleaner technologies and lessons learned from
developed countries by less-developed ones (IBRD, 1992).
Third, the nature of environmental protection, being a public good,
makes it difficult to control due to transboundary and free-rider effects. A
public good is defined as ‘non-rival and non-excludable in use’ (Common,
1995, p. 129). The transboundary effect results when environmental deg-
radation crosses national boundaries. This can take a number of forms.
The first is where neighbouring countries share a common resource and
the action of one country affects the others. Examples that fall in this
category are acid rain and the management of regional seas and interna-
tional rivers. The second is when one country’s action affects the world’s
global environmental resources, such as the atmosphere, and thereby has
an effect on all of the other countries. Examples that fall in this category
are ozone layer depletion and global warming. The third form is when
resources belonging to one country have high values for the international
community. Examples that fall in this category are tropical rainforests,
individual species and special ecological habitats. The free-rider effect,
where countries have the incentives to take advantage of the efforts of
others to provide the public good without paying their share or contribut-
ing to this step, also necessitates the existence of institutions and policies
(IBRD, 1992; Goodstein, 1999; Stavins, 2000).
Fourth, the international community negotiates international environ-
mental standards and sets them based on common principles and rules of
collaboration among the various countries. It gives proper weight to the
interests of all countries, including the poor and politically weak ones. In
addition, it monitors, enforces and abides by multilateral agreements on
environmental matters (Congleton, 1992; IBRD, 1992).
The relationship between economic development and environmental
pollution is complex as it involves a lot of different interrelated factors.
Among these factors are: the size of the economy, the sectoral structure,
the age of technology, the public demand for environmental quality and
the level and quality of environmental protection expenditure (Goodstein,
1999). However, the relationship mainly centres around five questions.
First, does the EKC relationship between income and environmental deg-
radation exist, and if yes, what is the income turning point? Second, what
are the other factors affecting this relationship, such as population growth/
density, international trade, policy, geography and income distribution and
what roles do they play? Third, how reliable are the results derived from
cross-sectional country or panel data in forecasting the environmental
path for an individual country? Fourth, what is the irreversible damage

to the ecological thresholds as a result of economic growth in terms of the
carrying capacity of the biosphere, ecosystem resilience and sustainability?
Fifth, what is the role played by institutions and policies in (1) explaining
the shape of the environment–income-growth relationship, (2) reducing
environmental damage and (3) ensuring sustainability of outcomes in the
future (Panayotou, 2000)?
Empirical models on economic growth and the environment

Most of the empirical models used in the studies on the economic-
growth–environment relationship consisted of reduced-form equations
relating environmental indicator(s) as the dependent variable(s) to a
measure of income as the independent variable. They basically captured
the net impact of income on the environment, without understanding the
various influences underlying this relationship. Some studies also control-
led for other variables, such as population density/growth, international
trade, income distribution, institutional policies and geography. The
environmental indicators were either in the form of pollutant emissions
(such as sulphur dioxide (SO2)), particulates, nitrogen oxides (NOx),
carbon dioxide (CO2), carbon monoxide (CO) or ambient concentrations
of various pollutants as recorded by monitoring stations. The measure
of income that was commonly used was income per capita, however,
some studies used income data converted into purchasing power parity
(PPP) or income at market exchange rates. The shape of the relationship
between income and environmental degradation was econometrically
estimated using cross-sectional or panel data and the functional equation
used was usually quadratic, log-quadratic or cubic form. The reduced-
form approach turned the income–environment relationship into a ‘black
box’ as it did not consider the underlying determinants of environmental
quality and which, in turn, limited its use in policy formulation. This
led to studies using decomposition analysis in order to disentangle other
effects, such as scale or level of production, composition and structure,
abatement technology, trade, preferences, the role of institutions and
environmental policies to understand the controlling effects behind the
relationship between income and environmental degradation (Panayotou,
2000).
The types of models used in these empirical studies differed. As seg-
regated by Panayotou (2000), the simplest model was the one examin-
ing the relationship between per capita income and an environmental
indicator(s). In some cases, these models included a time trend and were
usually in linear, quadratic, log-linear or log-quadratic forms (Shafik and
Bandyopadhyay, 1992; Hettige et al., 1992; Rothman, 1998; Kahn, 1998):
Linear Eit 5 b0 1 b1 yit 1 eit;

Quadratic Eit 5 b0 1 b1 yit 1 b2 (yit)2 1 eit;
Log-linear Eit 5 b0 1 b1 ln(yit) 1 eit;
Log-quadratic Eit 5 b0 1 b1 ln(yit) 1 b2 (ln yit)2 1 eit;
where: E 5 environmental indicator;

y 5 per capita income;
b 5 parameter to be estimated;
t 5 time trend;
e 5 error term.
Other models included population density as another independent

variable, commonly in a log-quadratic form equation (Panayotou, 1993;
Selden and Song, 1994; Cropper and Griffiths, 1994; Roberts and Grimes,
1997; Vincent, 1997; Carson et al., 1997):
Log-quadratic Eit 5 b0 1 b1 ln(yit) 1 b2 ln(Pit) + b3 (ln yit)2

1 b4 (ln Pit)2 1 eit;

P 5 population density;
t 5 time trend;
e 5 error term.
A third specification of models was used that included geographical char-

acteristics as independent variables, in addition to per capita income and
population density, usually using a quadratic form equation (Grossman
and Krueger, 1991, 1995):
Quadratic Eit 5 b0 1 b1 (yit) 1 b2 (Pit) + b3 (Git) 1 b4 (yit)2

1 b5 (Pit)2 1 b6 (Git)2 1 eit;

G 5 geographic characteristic;
t 5 time trend;
e 5 error term.
A fourth type of model included trade variables and per capita income
as independent variables, commonly using a quadratic form equation.
Examples of trade variables are export or import manufacturing ratio,
commerce intensity and prices for goods, such as wood or steel (Cropper
and Griffiths, 1994; Cole et al., 1997; Suri and Chapman, 1998; Kaufmann
et al., 1998):
Quadratic Eit 5 b0 1 b1 yit 1 b2 (yit)2 + b3 Tit 1 eit;

T 5 trade variable;
t 5 time trend;
e 5 error term.
A fifth type of model used institution variables in addition to per capita

income, mostly using a linear form equation. Examples of institution vari-
ables were political rights and civil freedom and macro-policy variables,
such as proportion of indebtedness to GDP and black market premium
on the exchange rate (Torras and Boyce, 1998; Bhattarai and Hammig,
2001):
Linear Eit 5 b0 1 b1 yit 1 b2 Iit + b3 Mit 1 eit;

I 5 institutions related variable;
M 5 macro-policy related variable;
t 5 time trend;
e 5 error term.
The last and more comprehensive type of models included policy vari-
ables in addition to population density, growth and per capita income as
independent variables, applying a cubic form equation (Panayotou,
1997):
Cubic Eit 5 b0 1 b1 (yit) 1 b2 (yit)2 + b3 (yit)3 1 b4 (Pit)

1 b5 (Pit)2 1 b6 (Pit)3 1 b7 (git) 1 b8 (git) (yit)
1 b9 (pit) 1 b10 (pit) (yit) 1 eit;

g 5 economic growth rate;
p 5 policy variable;
t 5 time trend;
e 5 error term.
Empirical studies on economic growth and the environment

The shape of the relationship between economic growth and the environ-
ment centres around four main explanations. It does not, however, imply
that there is a causal relationship between economic growth and environ-
mental quality (Perrings and Ansuategi, 2000). The first explanation is the
sectoral composition of the economy. The theory is that as the economy
moves along the production structure from agriculture to industry, local
air and water quality is bound to deteriorate due to the nature of industri-
alization. Industry in the low-income countries is mainly based on highly
polluting industries, such as chemicals, textiles, food processing, and so
on. As income rises, the economy moves towards expanding the services
sector, and thereby, environmental quality improves (Grossman and
Krueger, 1991; Panayotou, 1993, 1997; De Bruyn, 1997).
The second explanation is the technology shifts that accompany eco-
nomic growth. The interpretation is that as the economy becomes richer,
companies tend to use cleaner technology, which, in turn, improves envi-
ronmental quality (Seldon and Song, 1995; De Bruyn, 1997; Andreoni and
Levinson, 1998).
The third explanation is the income elasticity of demand for envi-
ronmental quality. It is interpreted that, since environmental quality is
a luxury good, as per capita income rises, the demand or preferences
for environmental amenity increase, and therefore new institutions and
regulations are created to protect the environment. However, given the
nature of environmental quality being a non-marketable good, changes
in its demand can be captured indirectly through changes in technology,
consumption, policy and regulations as well as their direct impact on the
environment (Antle and Heidebrink, 1995; McConnell, 1997).
The fourth explanation is the environmental constraints or the carrying
capacity of the biosphere. It is argued that economic growth is constrained
by the carrying capacity of the environment. In other words, there is a
limit to pollution or depletion resulting from growth that can be absorbed
by the environment, beyond which economic growth will come to a halt,
especially if the effects of growth on the environment are irreversible. This,
in turn, stimulates policy responses to improve the environmental quality.

The argument here is that, at low levels of income, the impact of economic
growth is within the carrying capacity of the biosphere. As income rises,
the environmental constraints tighten, and therefore induce regulation
and policy responses to reduce environmental degradation (Arrow et al.,
1995).
Basic studies of Environmental Kuznets Curve

The relationship between economic development and environmental
quality is a recent controversial debate, which started at the beginning of
the 1990s. It centres around the hypothesis that there is an empirical rela-
tionship between income per capita and environmental quality in terms
of pollution or resource depletion. There have been several observations
by a number of economists and ecologists that, as income increases, envi-
ronmental degradation increases up to a point, after which environmental
quality improves. In other words, the relationship has an ‘inverted-U’
shape. This relationship is referred to as the ‘Environmental Kuznets
Curve’.
The name of the relationship is derived from the similarity of the curve
to the relationship between income inequality and income growth in the
income distribution theory by Simon Kuznets in 1955, and which is termed
the ‘Kuznets Curve’. Given the obvious analogy, the bell-shaped relation-
ship between per capita income and pollution is named the Environmental
Kuznets Curve (EKC, hereafter).
Empirical literature on the EKC goes back to a paper prepared by
Grossman and Krueger (1991) for a conference on the US–Mexico Free
Trade Agreement. In this paper, which was later published in 1993,
Grossman and Krueger investigated the implications of economic growth
as a result of the North American Free Trade Agreement (NAFTA) on
environmental degradation as there were claims that the agreement might
increase environmental degradation. They studied the effect of reduction
in trade barriers on the environment in Mexico and assessed the relative
magnitudes of change in the level of pollution resulting from further trade
liberalization of three effects: expanding the scale of economic activity,
altering the composition of economic activity and changing the produc-
tion techniques. They carried out their study using a cross-country panel
of data on urban air pollution in two or three locations in each group of
cities in 42 countries from 1977 to 1988. The three pollutants were sus-
pended particulate matter (SPM), SO2 and dark matter (smoke). The data
were taken from the Global Environmental Monitoring System (GEMS)
published by the World Health Organization (WHO). The income data
used were real per capita GDP measured in 1985 US PPP from Summers
and Heston (1991) (Penn World Tables). They found that concentrations
for sulphur dioxide and smoke increased with higher per capita GDP at
low levels of national income and decreased with higher levels of income,
with a turning point between $4000 and $5000, measured in 1985 US
dollars of GDP. For the mass of suspended particles, in a given volume of
air, they found that the relationship was monotonically decreasing. They
also found that all three pollutants increased at an income level of $10 000
to $15 000.
In another study, Grossman and Krueger (1995) examined the relation-
ship between per capita income and four types of environmental indica-
tors: urban air pollution (SO2, smoke and heavy particles), the state of the
oxygen regime in river basins (dissolved oxygen, biological oxygen demand
(BOD), chemical oxygen demand (COD) and concentration of nitrates),
faecal contamination of river basins (faecal coliform and total coliform)
and contamination of river basins by heavy metals (concentration of lead,
cadmium, arsenic, mercury and nickel). They used panel data across many
countries for the 14 environmental indicators obtained from the GEMS.
They found evidence that, for most indicators, environmental quality ini-
tially deteriorated with economic growth, followed by a subsequent phase
of improvement. They also found that the turning points for the different
pollutants varied but, in most cases, were below an income per capita level
of $8000, measured in 1985 US dollars of GDP. They also found that for
seven indicators there was a statistically significant positive relationship
between environmental quality and per capita income GDP of $10 000.
Only in the case of total coliform did they find a significant adverse rela-
tionship at this level of income. They also highlighted that there were a few
points to be taken into account with regard to their findings. First, there
was no reason to believe that improvement in environmental quality as a
result of economic development was an automatic process. Second, the
reason for the downward slope and inverted U-shaped curve could be due
to countries ceasing to produce pollution-intensive goods as they imported
these goods from other countries. Third, the relationship observed in
previous research reflected the technological, political and economic con-
ditions existing at that time, from which poor countries could learn and
avoid the mistakes of earlier growth experiences.
Shafik and Bandyopadhyay (1992) explored the relationship between
economic growth and environmental quality in a background paper for
the World Development Report, which was later published by Shafik in
1994. They conducted a systematic analysis on environmental quality data
from 149 countries for the period 1960 to 1990. They used ten environ-
mental indicators: lack of clean water, lack of urban sanitation, ambient
levels of SPM, ambient sulphur oxides (SO2), change in forest area and
annual rate of deforestation between 1961 to 1986, dissolved oxygen in

rivers, faecal coliforms in rivers, municipal waste per capita and carbon
emissions per capita. The income data used were GDP per capita meas-
ured in 1985 US PPP. In order to overcome the comparability problem
of data across countries arising from differences in definitions and inac-
curate measurement sites, they applied a simple modelling technique, on
a consistent basis, to a large number of environmental quality indicators
and countries. They highlighted that there were four determinants affect-
ing environmental quality in any given country. The determinants were
(1) climate and location, which was referred to as endowment, (2) per
capita income, which reflected the production structure, urbanization and
consumption patterns of private goods, (3) exogenous factors, which were
available to all countries, but change over time, such as technology, and
(4) policies, which reflected the social decisions affecting the provision of
environmental public goods. Shafik and Bandyopadhyay in their study
focused on the relationship between per capita income and environmen-
tal quality, while taking into account the other determinants. The study
showed that some environmental indicators improved with higher income
per capita, such as water and sanitation, others worsened then improved,
such as particulates and sulphur oxides, and others deteriorated steadily,
such as dissolved oxygen in rivers, municipal solid wastes and carbon
emissions. For those environmental indicators that followed a bell-shaped
curve with rising income, the turning points varied. For instance, the
turning point for particulates was around $3280 per capita, while that of
sulphur oxides was around $3670 per capita.
Shafik (1994) concluded that individuals and countries tended to
address environmental problems at different stages of economic develop-
ment depending on the relative costs and benefits of the environmental
problem. For instance, water and sanitation were the first to be addressed,
given the relatively low costs and high private and social benefits involved.
Local air pollution came next as it imposed external but local costs and
was relatively costly to abate, with high benefits. They were also addressed
when countries reached a middle level of income where they became more
severe, as middle-income economies were often energy-intensive and
industrialized. Environmental problems that could be externalized, and
therefore, borne by poor or other countries, came last, such as solid wastes
and carbon emissions, as there were few incentives to bear the associ-
ated significant costs to abate. Moreover, the level of technology had a
positive impact on certain environmental quality indicators. If the costs
of degradation were localized, such as water sanitation and air pollution,
there would be high demand for improvements, and therefore technology
was crucial. However, where costs could be externalized and there was
uncertain knowledge of their detrimental effects, such as carbon emissions,

there was little demand for technological improvements. Furthermore, he
highlighted that environmental improvements were not automatic. They
required effective environmental policies and investments to be put in
place to reduce environmental degradation.
Panayotou (1993) examined the relationship between nominal per
capita of GDP and four environmental indicators: SO2, NOx, SPM and
deforestation using cross-sectional data. The income data used were GDP
per capita in 1985 nominal US dollars. There were 54 countries in the pol-
lution sample and 68 in the deforestation sample and the three pollutants
were measured in terms of emissions per capita on a national basis. He
found evidence of EKC for the four indicators with turning points per
capita of around $823 for deforestation, $3000 for SO2, $5500 for NOx and
$4500 for SPM, measured in 1985 US dollars of nominal GDP. Panayotou
concluded that:
at low levels of development both the quantity and intensity of environmental

degradation are limited to the impacts of subsistence economic activity on the
resource base and to limited quantities of biodegradable wastes. As economic
development accelerates with the intensification of agriculture and other
resource extraction and the take off of industrialization, the rates of resource
depletion begin to exceed the rates of resource regeneration, and waste genera-
tion increases in quantity and toxicity. At higher levels of development, struc-
tural change towards information-intensive industries and services, coupled
with increased environmental awareness, enforcement of environmental regu-
lations, better technology and higher environmental expenditures, result in
levelling off and gradual decline of environmental degradation. (Panayotou,
1993, p. 1)
In another study on deforestation, Cropper and Griffiths (1994) exam-

ined the relationship between per capita income and the rate of deforesta-
tion in three regions: Africa, Latin America and Asia. They used pooled
cross-sectional and time series data for each region for the period 1961 to
1988. Their sample included 64 developing countries. They also examined
the effects of population pressure on deforestation, which they captured
by including the rural population density and the rate of population
growth as independent variables in their equation. They reported two
findings. First, an EKC relationship for both Africa and Latin America
with turning points of $4760 and $5420, respectively. These turning points
were higher than those found by Panayotou, possibly because of the use of
panel data. Second, the rural population density for Africa shifted the rela-
tionship between income and the rate of deforestation upwards. However,
none of the variables included for Asia were statistically significant. They
concluded that reducing the rate of population growth was not necessarily
the best method for reducing the rate of deforestation in developing coun-
tries. They also said that deforestation was the problem of market failure,
which resulted from the lack of defining and enforcing property rights,
where people had no right in land ownership, and therefore no incentive
to make efficient decisions in the use of land.
Antle and Heidebrink (1995) examined the relationship between income
and two classes of environmental amenities: parklands and forests in low-
and high-income countries. The environmental indicators used were total
area of parks and protected areas (PARKS), deforestation (DEFOR),
afforestation (AFFOR) and total forest area (FOR). They gathered their
income data and area values from the World Development Report 1987,
PARKS data from the 1989/90 Environmental Data Report by the United
Nations Environment Programme (UNEP) and FOR and DEFOR data
from World Resources 1990/91. The income data used were per capita
GNP in 1985 nominal US dollars. They found a U-shaped relationship
and the income elasticity of demand for environmental quality was near
zero for countries with income per capita of around $1200–$2000, while
being positive and generally exceeding unity for countries with higher per
capita income levels.
However, in an empirical study done by Koop and Tole (1999) in
which they examined the relationship between deforestation and per
capita GDP, using less restrictive model specifications, they did not find
an EKC relationship between the two variables. They used data for 76
tropical developing countries for various years during the period 1961–92.
The deforestation indicator used was the loss in forest cover derived
from the FAO Production Yearbook. The per capita GDP data were
taken from the Penn World Table. Also they included two demographic
measures; the population density and population change, which were
derived from the FAO Production Yearbook and the Penn World Table.
The authors concluded that the extreme assumptions used in previous
studies that assumed uniformity across the countries led to the existence
of the EKC pattern between deforestation and per capita income. Upon
adopting random coefficient specifications, as they did in their study,
the inverted U-shaped relationship did not exist. They added that their
approach was more realistic given the wide variety of physical and social
characteristics across the countries.
With respect to urban environmental quality, a large number of studies
concentrated on air pollution. Selden and Song (1994) carried out an
empirical study using a cross-national panel of data on aggregate emis-
sions to examine the relationship between pollution and economic devel-
opment. They used aggregate emissions instead of urban air quality data
as they believed that the latter was valid if the objective was to understand
the factors underlying the pollution faced by urban dwellers. On the other
hand, they were trying to understand the environmental impact beyond
urban areas, and therefore the analysis using aggregate emissions was per-
ceived to provide greater insight. They concentrated on four air emissions:
SPM, SO2, NOx and CO across 30 countries and across time. Out of the
30 countries, 22 were categorized as high income, six were middle income
and two were low income. The emissions data constructed from fuel con-
sumption figures were obtained from the World Resources Institute (1991)
and were measured in terms of kilograms per capita on a national basis.
The income data used were per capita GDP in 1985 US dollars. Their
findings showed a U-shaped curve of per capita emissions for all four air
pollutants and per capita GDP. However, they found substantially higher
turning points for the similar pollutants used by Grossman and Kreuger.
The estimated turning points were $8709 for SO2, $11 217 for NOx, $10 289
for SPM and $5963 for CO. They considered this to be reasonable given
their use of aggregate emissions data as opposed to urban atmospheric
concentrations. On the other hand, their results were widely different from
those found by Panayotou, who also used aggregate emissions data. As
the turning points were high enough above the per capita income of most
countries, they concluded that global emissions of these pollutants would
continue to increase in the foreseeable future.
Cole et al. (1997) examined the relationship between per capita income
and a wide range of environmental indicators, which had indirect impact
on the environment using cross-country panel data sets. These indica-
tors were carbon dioxide, CFCs and halons, methane, nitrogen dioxide,
sulphur dioxide, suspended particulate matter, carbon monoxide, nitrates,
energy consumption, municipal waste and traffic volume. They arrived at
three main conclusions. First, the EKC existed only for local air pollut-
ants, while indicators with more global or indirect impact either increased
monotonically with income or had turning points at very high levels of per
capita income with large standard errors, unless they had been subjected
to a ‘multilateral policy initiative’, which implied that the process was not
automatic. Second, concentrations of local air pollutants in urban areas
peaked at a lower per capita income level than total emissions, which
implied that it was easier to improve the quality of urban air than to
reduce national emissions. Third, transport-generated local air pollutants
peaked at a higher per capita income level than total emissions per capita,
which implied the severity and need to tackle the problem in the transport
sector.
On the other hand, in a number of studies it has been found that CO2
follows an EKC pattern. In a study by Holtz-Eakin and Selden (1995) in
which they examined the relationship of CO2 emissions and GDP, they
found that carbon emissions increased monotonically, with a turning point

of $35 428 GDP per capita. They used a panel data of 130 countries for the
years 1951 to 1986. They also used the data to forecast global CO2 emis-
sions. They obtained the CO2 emissions data from the Oak Ridge National
Laboratory (ORNL), while the income and population data were gathered
from the Penn Mark V World Tables in Summers and Heston (1991). The
income data were per capita GDP in 1985 nominal US dollars. Using the
results of their study, they arrived at four conclusions. First, the marginal
propensity to emit (MPE) carbon dioxide diminished as GDP per capita
increased. Second, global emissions would continue to increase at 1.8 per
cent per annum. Third, the reason for the continued growth of CO2 was
the rapid population and economic growth in middle- and lower-income
countries, which had the highest MPE. Fourth, economic development
did not significantly alter the future of annual or cumulative flow of CO2
emissions.
Also Roberts and Grimes (1997) examined the relationship between
per capita GDP and national CO2 emissions for the period 1961 to 1991.
They also tracked the changes in the amount of carbon dioxide emitted (in
kilograms) per unit of their GDP, known as the National Carbon Intensity
(NCI), for low-, middle- and high-income countries, using data from
the World Bank and the Carbon Dioxide Information Analysis Center
(CDIAC). The GDP figures were adjusted for inflation using the World
Bank’s GDP deflator and the 1987 exchange rate. They found that CO2
intensity had steadily worsened for low- and middle-income countries,
but to a lower extent in the latter, however, it improved for high-income
countries. Their results were also statistically significant in the early 1970s
and then since 1982. They concluded that the EKC for carbon emissions
was not due to countries passing through stages of development but rather
due to a relatively small number of wealthy countries that became more
efficient after 1970, while the rest of the world worsened.
Schmalensee et al. (1998) conducted a study to project the emissions
of carbon dioxide from the combustion of fossil fuels to 2050. They used
national-level panel data for 47 countries for the period from 1950 to 1990,
which focused on around 80 per cent of CO2 emissions caused by human
activity. This was considered the most important source of potential future
warming. The GDP data were in purchasing power parity of 1985 US
dollars and were taken from the Penn World Table. They found evidence
of a U-shaped relationship between per capita carbon emissions and per
capita income. Their model also captured two typical patterns. First, for
developing countries, there was continuous, or even accelerating growth
of per capita CO2 emissions as per capita income increased. Second, for
highly developed countries, the relationship between per capita CO2
emissions and per capita income followed an inverted U-shaped pattern.

For instance, the turning point for the United States was $10 000 GDP per
capita. In other words, per capita CO2 emissions increased, then flattened
and then reduced at higher levels of economic development.
Similar results to those of Schmalensee et al. were found in a study by
Panayotou et al. (1999). They studied the relationship between per capita
income and CO2 emissions using panel national data with 3869 observa-
tions for 127 countries, with populations over 1 million, for the period
1960 to 1992. The sample accounted for approximately 95 per cent of the
world’s population and 90 per cent of the global CO2 emissions from fossil
fuels. The GDP data were obtained from the Penn World Table (1992) and
were in 1985 US dollars purchasing power parity. The CO2 data were taken
from Marland et al. (1999). They found evidence of an inverted U-shaped
relationship between per capita income and per capita CO2 emissions, with
a turning point of $12 000. They also found that the highest additions to
CO2 emissions were for income levels in the range of $1500 to $4000.
In a study by Ravallion et al. (1997), they found that CO2 emissions and
per capita GDP followed an inverted U-shaped pattern. They studied the
relationship between per capita GDP and carbon emissions for 42 coun-
tries using the data from the Oak Ridge National Laboratory. The data on
fuel use were obtained from the United Nations statistical division, where
the Marland and Rotty (1984) method was applied in order to convert
fuel consumption and cement manufacturing into carbon emissions. The
per capita GDP data used were in PPP-adjusted values based on 1985 US
dollars. They found that at low average income levels, carbon emissions
tended to increase, but then the relationship flattened out at middle- to
high-income levels and reversed at high levels of income. The authors
also found that at higher income inequality, there was a sizeable positive
impact on the aggregate income elasticity of carbon emissions, which
indicated that both average levels of income and inequality had effects on
carbon emissions.
Also Galeotti and Lanza (1999) attempted to shed further light on
the issue of the greenhouse gas CO2, using data covering 108 countries
around the world over the period 1971 to 1995, using new functional
forms developed by the International Energy Agency (IEA). In 1995, the
108 countries accounted for 88 per cent of the CO2 emissions generated by
fuel combustion. Their sample consisted of 2700 annual observations, of
which 700 observations were from 28 OECD countries and 2000 observa-
tions were from 80 non-OECD countries. In addition to analysing the
108 countries in the sample, they also analysed the sub-samples of OECD
and non-OECD countries in order to account for the different stages
of economic development, technological position and other structural
differences. Their findings showed a reasonable EKC relationship between

the country GDP and CO2 emissions. Under the Gamma functional form,
the turning points were $13 260 for all countries, $17 868 for non-OECD
countries and $15 582 for OECD countries, while under the Weibull func-
tional form, they were $13 648, $17 079 and $15 709, respectively. They
concluded that when using non-nested tests and applying new functional
forms, such as the Gamma and Weibull, to describe the reduced-form rela-
tionship between GDP and CO2 emissions, the emergence of a bell-shaped
EKC was possible with reasonable turning points.
In a recent study, Cavlovic et al. (2000) conducted a statistical meta-
analysis to synthesis the results of 25 existing EKC studies, with over 120
observations, and calculated new income turning points (ITPs) for 11 dif-
ferent pollutants to determine whether credible conclusions from previous
studies could be made. These pollutant categories were toxic emissions,
urban air quality, deforestation, heavy particulates, urban quality, water
quality/pollution, heavy metals, SO2, combustion by-products, hazardous
waste and CO2. They found the new ITPs ranged from $3020 for toxic
emissions to $199 345 for CO2. They concluded that (1) ITP results were
not necessarily representative across nations as their ITPs were higher
than those of previous studies for developed countries, (2) ITPs for emis-
sions and ambient concentrations of a particular pollutant were not com-
parable, (3) inclusion of trade effects variable, rather than income alone,
would tend to increase the ITPs and (4) methodological choices would
significantly affect the magnitude of ITPs.
In a study by Hettige et al. (1997) on water pollution, they examined
the relationship between economic development and industrial water pol-
lution. They used panel data collected at the factory level from national
and regional environmental protection agencies (EPAs) in 13 countries:
Brazil, China, India, Korea, Finland, Indonesia, Mexico, the Netherlands,
the Philippines, Sri Lanka, Taiwan, Thailand and the United States of
America. They basically tested the effect of income growth on three deter-
minants of pollution: the share of industry in total output, the industry
composition/share of polluting sectors and end-of-pipe pollution abate-
ment. In order to test the impact of regulation on the demand for labour,
they did a complementary study where they added a measure of regulatory
strictness to cross-country labour intensity. They found evidence of EKC
for the manufacturing share only. The sectoral composition improved to
middle-income level and then stabilized, while end-of-pipe pollution inten-
sity strongly declined with growth of income. They attributed these results
to (1) the application of stricter regulations as income increased and (2)
the complementarity of pollution and labour in production. The results
of their study also showed that income elasticities for both the pollution
and labour intensity were almost negative. Accordingly, they concluded

that the sector’s pollution/labour ratio was constant across all countries
throughout the development process.
In an earlier study by Grossman et al. (1994), they tried to study the
relationship between the composition and level of local economic activity,
in terms of current income, average lagged income and sectoral composi-
tion of income and concentrations of six criteria of air pollutants across
countries in the United States using panel data from the Aerometric
Information Retrieval System maintained by the US Environmental
Protection Agency (EPA). The air pollutants examined were total sus-
pended particulates (TSPs), SO2, NOx, CO, ozone (O3) and lead (Pb). The
income data used were per capita GDP in 1985 nominal US dollars. They
tried to separate scale, composition and technique effects in examining the
relationship between income and pollution. They concluded that composi-
tion of output had a significant effect on local air quality, however, changes
in the composition did not account for the observed relationship between
output levels and pollution in the case of the United States. With respect to
separating scale and technique effects, their results were inconclusive.
Policy and political institutions

As mentioned above, enactment of international legislation and enforce-
ment of environmental policies and regulations are vital in controlling
and minimizing the problems resulting from economic development and
its damaging effect on the environment. Despite this, very few authors
have included a policy variable in their studies of the economic-growth–
environment relationship. Congleton (1992) was among the first to explore
the effect of political institutions on the enactment of environmental
policies and regulations. Using cross-sectional data from 118 countries,
he tested the hypothesis that political institutions, rather than resource
endowments or market structure, determine enactment of environmental
policies and regulations to control pollution. His model depicted the level
of preferred environmental standard as a function of the policy-maker’s
share of national income, time horizon, market institutions in the country
and its resource base.
In his analysis, Congleton made several proxies and assumptions due
to the unavailability of data. With respect to the data on domestic envi-
ronmental policies and regulations he assumed that domestic politics
determine whether or not a country would sign a particular covenant or
protocol, so he used the data on signatories of two international agree-
ments. The first was the 1985 global convention of the United Nations in
Vienna whereby countries were committed to enact domestic legislation
to reduce emissions of the ozone-depleting substances, mainly CFCs. The
second was the 1987 Montreal Protocol that obliged developed countries
to reduce their CFC emissions to half 1986 levels by June 1999 and limits
the future consumption of CFCs by developing countries. He used the
type of regime as a proxy to the personal characteristics of decision- or
policy-makers. Gastil’s classification of countries (1987) had been used to
construct the type of regime, market structure, shares of national income
and time horizon of policy-makers. The country’s area was used as a
proxy for its resource endowments and population was proxied for labour
force. Both of these data were taken from the World Fact Book (1988).
Finally, the data on real GNP in 1987 US dollars were gathered from the
World Petroleum Institute (1990/91). Congleton arrived at three main
conclusions. First, authoritarian regimes faced a higher price for pollution
abatement than democratic governments. Second, authoritarian regimes
had uncertain career paths and shorter terms of office, and thereby would
tend to adopt less stringent environmental policies and regulations than
democratic ones. Third, they were less willing to sign international agree-
ments to protect their environments. The author concluded that political
institutions, rather than economic endowments, technology or market
institutions, would determine the level of domestic and international
environmental policies and regulations in the country.
Also Panayotou (1997) presented a study that focused on the policy
scope and provided a more analytical and structural approach to the
income–environment relationship. The roles of economic growth and pop-
ulation density were also explored. He examined the relationship between
income growth rate and policy variables against the level of sulphur dioxide
using panel data for 30 developed and developing countries for the period
1982 to 1994. The data for SO2 were gathered from the GEMS, while
those for GDP were obtained from the Penn World Table (1995) and were
measured in terms of 1985 constant US dollars as purchasing power parity
(adjusted GDP figures) was not available beyond 1992. He used proxies
for the quality of institutions to represent environmental policies. He
used five indicators to represent the quality of institutions obtained from
Knack and Keefer (1995). These indicators were enforcement of contracts,
quality of bureaucracy, the rule of law, corruption in government and the
risk of expropriation. Panayotou found an inverted U-shaped relation-
ship with a per capita turning point just under $5000. He also found that
the emission elasticity with respect to the improvements in the quality of
institutions was much higher than those of economic growth and popula-
tion density: a 50 per cent improvement in the efficacy of institutions and
policies at income levels between $10 000 and $20 000 could result in 50 per
cent reduction in the concentrations of SO2. On the other hand, at lower
income levels, the same improvement did not result in a similar reduction
of SO2. Panayotou came to three main conclusions. First, environmental

quality was influenced by the level of income rather than its growth rate.
Second, in the case of SO2, governmental policies and social institutions
played an important role in significantly reducing environmental degra-
dation at low income levels and speeding up the improvements at high
income levels, and thus mitigating the environmental externalities of eco-
nomic growth. Hence, improvements in environmental quality were not
automatic but depended on policies and institutions. Third, the structural
changes of supply and demand for goods and services throughout the
development process played an important role in determining the when
and how of environmental improvements. Therefore, markets as well as
policies determined the ‘environmental price’ of economic growth. In con-
clusion, Panayotou noted that improvements in the quality of policies and
institutions had higher payoffs for pollution abatement at higher levels of
income as it tended to improve monitoring activities. Therefore, efforts
should be focused on improving the quality of policies and institutions
rather than slowing down economic or population growth.
De Bruyn (1997) investigated the origins of change in emissions using
decomposition analysis in order to empirically explain the mechanisms
underlying the U-shaped relationship between SO2 emissions and income
in developed economies during the 1980s. In other words, he attempted
to determine the factors that shape the patterns of emissions over time
to explain why a U-shaped relationship occurred between economic
growth and pollution. The emissions data were obtained from OECD,
Environmental Data Compendium (1995), Klaassen (1995) and World
Resources (1994–95), while the income data were obtained from the
World Bank, World Table (1995) for 1993 in thousands of US dollars
based on market exchange rates. He highlighted that the main three argu-
ments explaining the EKC hypothesis according to the literature were
valid, but scarce. First, as income increased, the positive income elasticities
of demand for environmental quality and a more open political system
resulted in effective environmental policies. De Bruyn explained that the
argument was valid but was likely to occur in democracies, which are
scarce with contradictory policies. Second, the composition changes of
production and consumption were associated with rising incomes, which
assumed that economic development passed through transition stages with
respect to the structure of production. In other words, a shift took place
in countries from agriculture to industry to service-oriented economies as
income increased and that, in turn, resulted in the pattern of the inverted
U-shaped curve with the highest point occurring in the industrial stage.
Third, if changes in the production structure were not accompanied by
changes in the consumption structure, then the EKC relationship resulted
from dumping of pollutants from developed countries into the economies

of developing countries.
De Bruyn’s findings showed that structural changes failed to give
evidence of being the determinant of the reduction in SO2 emissions of
developed economies during the 1980s. However, by assuming that envi-
ronmental policy targets were representative of environmental policy
efforts, he found that richer countries had more ambitious environmental
policies. Therefore, he concluded that the scale effect of economic growth
resulted in environmental degradation unless corrected by appropri-
ate environmental policies. Hence, environmental policies, fostered by
international agreements, were the most important determinant for the
reduction of SO2 emissions at higher income levels. He also added that
income was only a minor determinant of environmental policy and that
the current state of the environment was an important determinant in
reducing SO2 emissions.
Hilton and Levinson (1998) examined the relationship between automo-
tive lead emissions and national income for 48 countries, with 1990 popu-
lations exceeding 10 million, over a period of 20 years (1972–92) using data
from Octel’s Worldwide Gasoline Survey. The survey reported the average
lead content of gasoline for over 150 countries, on a biannual basis. The
total gasoline consumption data were compiled from various OECD pub-
lications, while income and population data were obtained from the Penn
World Table documented in Summers and Heston (1991). The income
data used were real per capita GDP. They broke down total emissions into
‘pollution intensity’, which was measured as the lead per gallon of gaso-
line, and ‘pollution activity’, being the total gasoline consumption.
The authors arrived at three main conclusions. First, a U-shaped curve
relationship existed between lead emissions and income. Second, the peak
of the curve was dependent on the functional form estimated and the
time period considered. Third, lead pollution was the product of pollu-
tion intensity and polluting activity and the declining portion of the curve
depended on reducing the gasoline lead content and not gasoline use. In
other words, the existence of regulations to reduce pollution intensity, as a
policy response, was a requirement for the improvement in environmental
quality that accompanied growth in income.
Hilton and Levinson examined the relationship between income and
two separate factors of pollution: pollution activity and pollution inten-
sity. They explained that the observed inverse-U relationship might be
due to two reasons. The first reason they attributed to what was called
the ‘composition effect’. This was referred to as the pattern of economic
development, which was considered to go through transition stages, start-
ing with agriculture, which was not pollution-intensive, to manufacturing
with its high pollution, especially in the early stages, to the less pollut-
ing service industries. One of the reasons for this pattern could be that
wealthier countries dumped their polluting processes in poorer countries.
This, in turn, meant that it was not possible for all nations to experience
improvements in environmental quality. The second reason they gave was
that the EKC might be based on two separate relationships. First was the
‘scale effect’, where higher economic activity generated more pollution
and wealthy countries, with more polluting activity, generated more pollu-
tion. This was referred to as pollution activity. Second was the ‘technique
effect’, where citizens in wealthy countries demanded more environmental
quality in the form of regulations to protect the environment and to reduce
the amount of pollution per unit of activity. This was referred to as pollu-
tion intensity. Therefore, the shape of the pollution–income relationship
was the product of pollution activity and pollution intensity. So their con-
clusion was that polluting activity increased with income, while polluting
intensity decreased with income and the product of the two followed an
inverse U-shape.
Midlarsky (1998) examined the relationship between democracy and
the environment. He used six environmental indicators as the dependent
variables: CO2 emissions, deforestation, soil erosion by water, land area
protection, freshwater availability and soil erosion by chemicals. He found
that democratic countries had higher CO2 emissions, deforestation and
soil erosion. However, they tended to protect a higher percentage of their
land area. For the last two environmental indicators he found no signifi-
cant relationship and concluded that there was no uniform relationship
between democracy and the environment.
Tuan (1999) found similar results for CO2 emissions to those found by
Panayotou for SO2. He presented a paper on the relationship between
income and environmental quality, mainly CO2 emissions. He used panel
data for six countries at different stages of economic development. The
countries examined were Vietnam, Thailand, Korea, France, Japan and
the United States. In addition to the income variable, he included in his
model other social variables, such as population, economic growth and
institutional capacity in order to examine the underlying impacts on the
EKC trajectory. The environmental data for the three developed coun-
tries were obtained from the OECD Environmental Data Compendium
for the years 1993, 1995 and 1997. The energy data for the three Asian
countries were obtained from the ENERDATA database. The per capita
GDP data were in 1987 constant US dollars and were obtained from
the database of ENERDATA as well as the data on economic growth
rates and demographic density. For the political or institutional capacity
variable, the author used four indicators from the Business Environmental
Risk Intelligence (BERI), a private international investment risk service.

The indicators were bureaucratic delays, enforceability of contracts,
nationalization potential and infrastructure quality.
Tuan tested four hypotheses. The first was the existence of an inverted
U-shaped relationship between per capita income and CO2 emissions.
The second was the impact of the rate of economic growth on the EKC
trajectory. The third was the impact of population density on EKC. He
hypothesized that the higher the population density, the lower the turning
point of the EKC, and hence, a better environmental quality. The fourth
hypothesis was the role of institutions and policies in determining the
EKC path. Again, he hypothesized that effective policies and institutions
would lower the EKC turning point, and therefore would improve the
environment along with economic growth. Tuan found an EKC relation-
ship between per capita income and CO2 emissions with a turning point
of approximately $18 000. With respect to the impact of economic growth
and population rate on CO2 emissions, the author found that their impact
varied according to the level of income and the country’s stage of develop-
ment. The author’s main conclusion was that the quality of policies and
institutions played an important role in smoothing out the EKC, however,
the payoffs were not the same for developing and developed countries. He
noted that a good institutional capacity and sound environmental policies
could, to a large extent, contribute to a better environment.
Similar results were also found in a study by Bhattarai and Hammig
(2001) for deforestation. They studied the relationship between income
and deforestation in 21 countries in Latin America for the period 1972
to 1995. They used data from various sources, such as the FAO, WRI
and the UNEP. They also used the Penn World Table from Summers
and Heston (1991) for the national income data, exchange rates and
trade, the Freedom House Tables and Knack and Keefer (1995) for the
index measures for sociopolitical institutions and the World Development
Report (1998) for other variables. The per capita GDP data were adjusted
1998 US dollars PPP. Their findings confirmed the existence of an EKC
in Latin America with turning point around $6800. They concluded that
(1) macroeconomic factors, such as indebtedness, inflation and exchange
rate policies would shift the intercept of the EKC and (2) the present level
of deforestation in Latin America would be reduced by strengthening the
sociopolitical institutions in the region.
A number of studies examined the relationship between economic
growth and the environment across different political systems. Torras and
Boyce (1998) investigated the causal linkages between changes in income
and pollution levels. They hypothesized that more equitable distribution
of power was an essential link for the improvement of some air and water
pollutants as a result of higher per capita income. The air and water pol-
lutants examined were sulphur dioxide, smoke, heavy particles, dissolved
oxygen, faecal coliform, access to safe water and access to sanitation. For
the first five pollutants, they used data from the GEMS for the period
1977 to 1991. The air pollution data contained observations from 19 to
42 countries and that for water pollution contained observations from 58
countries. The percentages of population with safe water and sanitation
were obtained from the United Nations Development Programme (1994)
and were national-level variables with no time series dimension. The per
capita income used were measured in real PPP adjusted for 1985 US
dollars. The income data for per capita GDP were taken from Summers
and Heston (1991), while the national income levels data were taken from
the UNDP (1994).
The authors predicted that the higher the power inequality, the higher
the levels of pollution, so they included other variables as proxies for
power inequality, such as Gini ratio of income inequality, literacy rates
and political rights and civil liberties. Four of the seven environmental
indicators showed EKC patterns when excluding the power inequality
variable. These indicators were sulphur dioxide, smoke, access to safe
water and access to sanitation with turning points of $3890, $4350, $11 255
and $10 957, respectively. Heavy particles monotonically decreased with
rising income while dissolved oxygen and faecal coliform monotonically
increased. When including the power equality variable, only three pol-
lutants showed EKC patterns, which were sulphur dioxide, dissolved
oxygen and access to safe water with turning points of $3360, $19 865 and
$6900, respectively. Access to sanitation increased monotonically, while
the relationships for smoke, heavy particles and faecal coliform were sta-
tistically insignificant. They arrived at two main conclusions. First, equi-
table distribution of power, in the form of equitable income distribution,
wider literacy and greater political liberties and civil rights, would lead to
better environmental quality, especially in low-income nations. Second,
improvements in environmental quality would not automatically accom-
pany continued growth in per capita income.
Also, Deacon (1999) argued that the form of government in a country
played a crucial role in the way economic growth would affect the envi-
ronment and that omitting it as a determinant of environmental quality
could lead to biased results and interpretation. He presented a study to
examine the effect of the government regime on the provision of envi-
ronmental public goods and environmental policies using cross-sectional
analysis of 118 countries. Deacon classified governments into groups with
similar political attributes using compiled data sets from two sources. The
first was the Cross-national Time-series Data Archive, which was first
compiled by Arthur Banks and listed attributes of the government systems

in almost all countries. The second was the Polity III database maintained
by Keith Jaggers and T.R. Gurr (1995) that had information on the politi-
cal systems of all countries with populations over 500 000. The data sets in
these two sources went back to the nineteenth century and up to the late
1990s.
He examined two environmental public goods; the rural population
with access to sanitation facilities and safe drinking water, using data from
the World Health Organization reported in the World Resources Institute
database. Deacon also examined two other public goods; roads and public
education. The data for roads were taken from the International Road
Federation and those for public education were taken from Banks. The
environmental policy measure he used was the lead content of gasoline
as reported by Octel Corporation. He showed that there was an inverted
U-shaped relationship between pollution and income, however, the shape
of the curve was determined by the form of government. In other words,
income elasticities of provision of public goods were affected by the type
of governance. He concluded that income levels and government regimes
were highly correlated. He showed that incomes were higher in countries
with democratic governments than autocratic ones and environmental
quality tended to be lower in the latter. His explanations were that (1)
property rights were less defined and the rule of law was absent in less
democratic countries, which in turn suppressed investment in environ-
mental quality and other public goods, and (2) the governments were
generally corrupted, and hence drained any economic surplus that could
be produced by the private sector.
Barrett and Graddy (2000) made an attempt to determine the rela-
tionship between a selected number of air and water pollutants and per
capita income. The air pollutants were sulphur dioxide, smoke and heavy
particles, while the water pollutants were dissolved oxygen, biological
oxygen demand, chemical oxygen demand, nitrates, faecal coliforms, total
coliforms, lead, cadmium, arsenic, mercury and nickel. They relied on the
data used by Grossman and Krueger, which were gathered by the Global
Environmental Monitoring System and they added freedom variables that
were constructed from the civil and political indices developed by Freedom
House. The civil freedoms index reflected constraints imposed on the level
of freedom of the press and individuals, while the political freedoms index
reflected the level of freedom of choice of government, existence of opposi-
tion, and so on. The income data were taken from Summers and Heston
(1991) and were adjusted for differences in PPP.
Barrett and Graddy’s hypothesis was that improvement in air and water
pollutants was due to improvement in the political system in a country,
as they believed that politics played an intermediate role between income

and pollution. Their findings showed that as income per capita increased,
certain measures of pollution deteriorated then later improved. They
also found that for some measures of pollution, environmental quality
improved according to the level of civil and political freedoms allowed
in the countries, while for other measures environmental quality did not
seem to depend on freedoms. For air pollutants, they found three main
observations. First, income and freedom were positively correlated. In
other words, the high-freedom countries were rich, while the low- or
medium-freedom countries were poorer. Second, with the exception of
concentrations of sulphur dioxide, pollution levels and freedom were
negatively correlated. Third, as civil and political freedoms were highly
correlated, the correlations between freedoms and income and freedoms
and pollution were the same for both civil and political freedoms. On the
other hand, the relationship between freedom and water pollution levels
was not always negative. Some of the water pollutants increased with
higher freedom, such as biological and chemical oxygen demand, nitrates
and total coliform. Faecal coliform, on the other hand, reduced with
increased freedom. With respect to heavy metals, freedoms were jointly
significant in the case of arsenic and cadmium, where the former reduced
with higher freedom, and therefore was consistent with their hypothesis
of an induced policy response, while the latter increased. The combined
effect of freedom and income was very significant in the case of lead, with
a negative relationship, which also might be the result of an induced policy
response. However, the relationships for nickel and mercury were statisti-
cally insignificant. They concluded that effective environmental protec-
tion required both economic and political reforms. Nevertheless, they
could not explain why freedoms affected some measures of environmental
quality and not others.
The role of international trade

A number of empirical studies attempted to examine the role of interna-
tional trade and direct investment in the income-growth–environment
relationship. They hypothesized that the shift of pollution dumping from
richer to poorer countries either through international trade or direct
investment was the reason behind the reduction of pollution in the richer
countries. Among the authors who examined this hypothesis were Hettige
et al. (1992). They examined the relationship between the level of economic
development and toxic intensity of industrial production in 80 countries
for the period 1960 to 1988. They used industrial data gathered by the
United Nations to calculate the shares of total manufactured output for 37
sectors as defined in the International Standard Industrial Classification
(ISIC). The income data were GDP per capita in 1985 nominal US dollars.
They found an inverted U-shaped relationship for pollution intensity per
unit of GDP, however, they did not find the same pattern for toxic inten-
sity per unit of industrial output. On the contrary, they found that manu-
facturing output increased steadily with income. They concluded that this
was the result of development in these countries, and, therefore, a shift
in the production structure from industry towards services, which were
lower in pollution. They also analysed the relationship of toxic intensity
and trade policy in less developed countries (LDCs) and they found that
the closer the country to international trade, the more rapid the growth of
toxic intensity of manufacturing. They concluded that, for the majority of
countries, there would be an increasing long-term trend of industrial emis-
sions, with respect to both GDP and manufacturing output. This would
also be higher in the poorer countries mainly due to the tighter environ-
mental regulations in the industrialized nations since 1970.
Grossman and Krueger (1991) also included a trade variable in their
study as they studied the effect of reduction in trade barriers on the envi-
ronment in Mexico and assessed the relative magnitudes of change in the
level of pollution resulting from further trade liberalization. They found
significant evidence that trade helped to reduce the level of pollution for
only one of the three pollution variables, which was urban concentrations
of SO2. Shafik and Bandyopadhyay (1992) tested the hypothesis that trade
induced the use of cleaner technology, and thereby would reduce pollution.
However, they found weak evidence between the two variables.
Rock (1996) studied the relationship between trade policy and the
environment in rich and poor countries using cross-country data over the
period 1973 to 1985. He used the pollution data on toxic chemical intensity
of GDP, which were previously used by Hettige et al. (1992) in their study.
However, he included two other variables: the manufacturing share of
GDP and the energy intensity of GDP. The first variable was included to
separate the effect of the broad composition of output on pollution inten-
sity from that of trade policy, while the latter was included to separate the
effect of energy price policy from trade policy. The trade orientation data
were taken from the World Bank (1987) where countries were grouped
into four categories depending on their trade policies. The four groups
were strongly outward-oriented, moderately outward-oriented, moder-
ately inward-oriented and strongly inward-oriented. He found evidence of
higher pollution intensity in countries with more open trade policies and
that the most inward-oriented developing countries had the lowest toxic
chemical pollution intensity of GDP.
The relationship between per capita income and consumption of
primary commercial energy was examined by Suri and Chapman (1998)
using pooled cross-country and time series data in order to quantify the
impact of the actual movement of goods between countries. They used
observations from 33 countries over the period 1971 to 1990. The energy
data, expressed in oil equivalents, were obtained from the IEA, while the
income and trade data were obtained from the Penn World Table and the
income data were expressed in 1985 US dollars PPP. The authors pre-
sented two types of models. The first model implicitly captured the impacts
of both structural change and international trade, while the second model
explicitly analysed the effect of international trade on commercial energy
use. The variable representing structural change was the share of total
manufacturing in GDP (that is, manufacturing for domestic market and
production for exports). International trade variables were represented
by import- and export-manufacturing ratios. In both models, the EKC
pattern was depicted, with a turning point of $55 000 in the first model and
$224 000 in the second one. Accordingly, they concluded that, all things
being equal, international trade would tend to increase the pollutant
emissions related to commercial energy consumption. The authors also
found that energy use increased in both industrialized and industrializing
countries, however, it was substantially higher in the latter. Moreover, as
opposed to industrializing countries, whose imports were largely interme-
diate and capital goods, which were essential for building an industrial
base, industrialized nations benefited from importing manufactured goods
from industrializing countries. Therefore, their relative increase in energy
requirements to GDP was substantially lower. The authors concluded that
the imports of manufactured goods, in addition to the structural change
from manufacturing to non-energy-intensive service sector, were the result
of this decline.
Some authors concentrated on trade patterns and consumption, rather
than production, in examining the income–environment relationship.
Rothman (1998) studied the relationship between per capita GDP and
consumption activities using data on eight categories of consumer goods
from the United Nations International Comparison Programme (1994).
Among the commodities used were food, beverages and tobacco, clothing
and footwear, gross rent, fuel and power, medical care and services, trans-
port and communications, recreation, entertainment and education. The
unit for each commodity was the quantity that could be bought for $1 at
average international prices. The data used for the per capita GDP were in
1985 US dollars PPP. The only commodity they found showing evidence
of EKC was food, beverages and tobacco with a turning point of $12 889.
Other commodities, in terms of shares, also changed with income, such as
clothing and footwear, gross rent, fuel and power and medical care and
services, with turning points of $35 263, $23 278 and $47 171, respectively.
However, this was due to changes in relative share rather than a decline in
consumption.
The extent of the so-called pollution haven hypothesis (PHH) in
explaining the EKC relationship was assessed by Cole (2004). The PHH
was defined as the migration or displacement of pollution-intensive
industries from developed to developing countries due to more stringent
environmental regulations in the former countries. The author estimated
the EKC model for ten air and water pollutants for a sample of OECD
countries over the period 1980 to 1997. The air pollutants were CO2, NOx,
SO2, CO, SPM and volatile organic compounds (VOC), while the water
pollutants were the level of nitrates, the level of phosphorous, BOD and
the levels of dissolved oxygen. To capture the PHH effects, the author
included the share of pollution-intensive exports and imports to and from
non-OECD countries in total exports and imports. He also included the
share of manufacturing in GNP to capture the effect of structural change
within the economy on pollution. To reflect trade openness/intensity, the
author used the ratio of the sum of imports and exports to GNP. Cole
found that, with the exception of VOC and CO, all the other eight pollut-
ants exhibited an EKC relationship with per capita income, with turning
points ranging between $1973 and $35 140. He also found a positive rela-
tionship between the manufacturing share of GNP and environmental
quality for eight pollutants, of which seven were statistically significant.
This indicated that structural change within the economy was partially
responsible for pollution reduction at higher income levels. With respect
to PHH, Cole found mixed results. There was a negative relationship
between the share of pollution-intensive imports to total imports and envi-
ronmental quality for seven pollutants (five air pollutants and two water
pollutants), of which five were statistically significant. On the other hand,
there was a positive relationship between the share of pollution-intensive
exports to total exports and environmental quality for five pollutants (one
air pollutant and four water pollutants), of which four were statistically
significant. Nevertheless, upon omitting the pollution haven variables, the
EKC turning points were lower for eight out of the ten pollutants, suggest-
ing that these effects contributed to the reduction of emissions at higher
levels of income. With respect to trade openness, the author found a nega-
tive relationship between trade openness and environmental quality for
eight out of the ten pollutants, of which six were statistically significant.
The author concluded that the reduction in emissions at higher levels
of income was due to higher demand for environmental regulation and
investment in abatement technologies, trade openness, structural change
within the economy and increased imports of pollution-intensive indus-
tries. Nevertheless, the author could not ascertain that the developing
countries would be able to follow an income–pollution path similar to that

of the developed countries as the demand of the developed countries for
pollution-intensive products was being met by the developing countries.
Therefore, the developing countries did not have anyone to pass on the
production of such products to.
The role of population density

In the economic-growth–environment literature, there are a number of
empirical studies that included population density variables as it is per-
ceived that population density affects the relationship between income
and the environment. As mentioned earlier, Cropper and Griffiths (1994)
examined the effects of population pressure on deforestation by including
the rural population density and the rate of population growth as inde-
pendent variables in their equation. They found an EKC relationship for
both Africa and Latin America with turning points of $4760 and $5420,
respectively. They found that the higher the rural population density, the
higher the deforestation rate. An increase in rural population density by
100 persons per 1000 hectares implied an increase of the deforestation rate
by 0.33 per cent in Africa. However, none of the variables included for Asia
were statistically significant. The authors also found that the rural popu-
lation density for Africa shifted the relationship between income and the
rate of deforestation upwards, which implied the existence of a large trade-
off between per capita income and rural population density. For instance,
a country with a population density of 0.1 persons per hectare, being the
average population density for the sample of African countries, had a peak
deforestation rate of 1.26 per cent per year at a per capita income level of
$4760. For a country with a population density of 0.7 persons per hectare,
it would need a per capita income level of $11 650 per year to achieve the
same deforestation rate. Nevertheless, the authors concluded that defor-
estation was the problem of market failure, which resulted from the lack
of defining and enforcing property rights. Therefore, reducing the rate of
population growth was not necessarily the best method for reducing the
rate of deforestation in developing countries.
Also as mentioned earlier, Panayotou (1997) explored the role of popu-
lation density using two approaches: the reduced-form analysis and the
income decomposition analysis with its three constituents, scale, composi-
tion and abatement. Under the reduced-form approach, Panayotou found
an inverted U-shaped relationship with a per capita turning point of just
under $5000. However, he found that the higher the population density, the
higher the level of ambient SO2 at every income level. With respect to the
decomposition approach, the author found that, controlling for all other
factors, low levels of population density of under 50 persons per square
kilometre were associated with high levels of ambient SO2 emissions, being
70 kg/km3. As population density increased, SO2 levels dropped, reaching
their lowest level of 45 kg/km3 at a population density level of 170 persons
per square kilometre and then started to increase again. The author attrib-
uted the higher SO2 levels at lower population density to the lack of pres-
sure in countries with spare population to control emissions. However, as
population density increased, more pressure was exerted to control emis-
sions as more people were exposed to pollution. This continued up to the
level where the household use of coal and non-commercial fuels exceeded
the pressure for pollution abatement.
Similarly, as mentioned before, Tuan (1999) tested the impact of popu-
lation density on the EKC trajectory. He hypothesized that the higher the
population density, the lower the turning point of the EKC, and, thus, a
better environmental quality. He found an EKC relationship between per
capita income and CO2 emissions with a turning point of approximately
$18 000. However, with respect to the impact of economic growth and
population rate on CO2 emissions, Tuan found that their impact varied
according to the level of income and the country’s stage of development.
At income levels below $2000 per capita, a highly dense population had a
negative impact on the environment. On the other hand, at income levels
of $2000 per capita or more, a highly dense population had a positive
effect on the environment. For instance, at an income level of $1000 per
capita, an increase in the population density of 67 per cent resulted in a
12 per cent increase in CO2 emissions. However, at an income level of
$10 000 per capita, a similar increase in the population density resulted in
a 22 per cent reduction in CO2 emissions. The author concluded that at
higher income levels, the higher the population density, the more pressure
was being exerted to control emissions as more people were being exposed
to the pollution.
Vincent (1997) presented an analysis for a single country, Malaysia, in
an attempt to examine the pollution–income relationship. Unlike previous
studies that conducted an analysis on cross-sectional or panel data for a
sample of developed and developing countries, Vincent used a panel data
set for the 13 Malaysian states and he analysed the relationship between
income and a number of air and water pollutants over time (1970s to
1990s). He also measured the impact of population density on the various
pollutants. The pollutants were TSPs, BOD, COD, ammoniacal nitrogen,
pH and suspended solids. The data used were ambient levels measured by
monitoring stations.
He chose Malaysia for several reasons. First, the country had a rich data
set with nearly two decades of readings on ambient air and water pollu-
tion. Second, the country’s economy had been one of the fastest growing
in the world since the 1970s, and therefore its income data contained
ample variation over time, even within the states as growth had not been
equally rapid across all states. Third, income data overlapped between
the states. Fourth, the borders of most states were determined by ridges
that separated the major water basins and, to a certain extent, air basins.
Fifth, the states shared the same pollution policies, which varied across
pollutants.
According to Vincent’s study, there was no evidence of EKCs for any of
the six income–pollution relationships in Malaysia. This result was incon-
sistent with previous research conducted using cross-country relationships.
With respect to population density, Vincent found that the net impact of
population density on TSP concentrations was positive and the interaction
term between population density and time was statistically significant but
negative. He attributed the positive impact to the various household activi-
ties, which were important sources of TSP concentrations, such as heating,
cooking, rubbish disposal and transportation. The negative interaction
term indicated a reduction in the per capita TSP concentrations due to the
pressure exerted through the enforcement of effective anti-pollution regula-
tions. For the water pollutants, and while holding income constant, Vincent
found that higher population densities were associated with worse water
quality for BOD and ammoniacal nitrogen, and better water quality for sus-
pended solids. He attributed this to the higher sewage discharge as a result
of higher population, in the case of BOD and ammoniacal nitrogen, and
the movement of people from rural to urban areas in the case of suspended
solids. The interaction term between population density and time was also
significant. It was negative in the case of BOD and ammoniacal nitrogen as
a result of the introduction of regulations to reduce BOD discharge and the
percentage of population without access to sanitary facilities. The interac-
tion term was positive for suspended solids indicating agriculture and set-
tlement movements to steeper and erosion-prone areas. In conclusion, the
author said that his findings did not mean that EKC did not exist anywhere,
but it meant that policy-makers should not assume that economic growth
was an automatic solution for the air and water pollution problems.
Kaufmann et al. (1998) attempted to explore the effects of income and
spatial intensity of economic activity on the atmospheric concentration
of SO2. They used a panel of international data for 23 countries with
observations between 1974 and 1989. Out of the 23 countries, 13 were
developed nations, seven were developing nations and three were centrally
planned economies. The data on SO2 were obtained from the United
Nations Statistical Yearbook (1993) and GDP per capita and population
data were obtained from Summers and Heston (1993). The income data
were expressed in 1985 nominal US dollars. They used two variables to
proxy the spatial intensity of economic activity, one for cities and the other
one for the entire nation. The variable for spatial intensity of economic
activities in cities was the product of a city’s population density and the
per capita GDP of the nation in which the city was located, while that for
the nation was the GDP divided by the nation’s area. The result of their
study indicated a U-shaped relationship between both per capita GDP
and atmospheric concentration of SO2. The concentration of SO2 tended
to reduce as per capita income rose from $3000 to $12 500, however,
it increased beyond $12 500. On the other hand, there was an inverted
U-shaped relationship between spatial intensity of economic activity and
atmospheric concentration of SO2 with turning points of $6.7 million per
square mile for the nation variable and $153 million per square mile for the
city variable. They suggested that the impetus for policies and technologies
to reduce SO2 atmospheric concentration was spatial intensity of economic
activity rather than income. This implied that atmospheric concentration
of SO2 would decline faster than indicated in previous studies, however, it
would depend on the rate of income growth relative to population growth.
They suggested that the inverted U-shape found in previous studies might
be due to the omission of the variables representing changes in the mix
and spatial intensity of economic activity, and therefore the pattern was a
proxy for changes in the mix of economic activity associated with changes
in per capita GDP.
Household preferences and income elasticity of demand

Income elasticity of demand for environmental quality changes as income
rises. One of the explanations given for the downward sloping section of
the EKC is that as per capita income increases, the demand or preference
for environmental amenity also rises. A number of studies attempted
to study the role of preferences in the economic-growth–environment
relationship.
In a study, McConnell (1997) tried to understand the role of prefer-
ences, and in particular income elasticity of demand for environmental
quality in the EKC relationship, by decomposing the reduced-form effect
of income changes on pollution. He used panel data of approximately 50
countries from the 1970s to 1990s. The types of environmental indicators
were for ambient urban air and water pollutants, such as sulphur dioxide,
suspended particulates, nitrogen oxides, carbon monoxide, dark matter,
dissolved oxygen, faecal coliform and emissions, in the case of toxic
substances and carbon dioxide. He concluded that:
preferences that were consistent with a high income elasticity of demand were
neither necessary nor sufficient for pollution to grow with income. Simple
models showed that preferences that were consistent with a high income elas-
ticity of demand for environmental quality were attenuated by high and rising
abatement costs or a high impact of pollution on production. Hence, it would
be feasible for pollution to decline with a zero income elasticity of demand for
environmental quality, or to increase with a high income elasticity of demand
for environmental quality.
In another study, Kahn (1998) examined the effect of per capita income
on hydrocarbon emissions of vehicles. He used the 1993 Random Roadside
Test created by the California Department of Consumer Affairs Bureau of
Automotive Repairs, whereby vehicles at random are selected to have
an emissions test. He used the 1990 Census of Population and Housing,
which included the zip code median household income as a proxy for
household income. The hydrocarbon emissions were measured in parts per
million (ppm) and household income was measured in thousands of US
dollars. The author tried to study the total annual contribution of a house-
hold to local pollution in relation to its income. He presented evidence of
an inverted U-shaped relationship between income and emissions, with a
turning point between $25 000 and $35 000. He also found that average
emissions for vehicles owned by households with income levels below
$35 000 were twice as high as those for households with income levels
over $45 000, showing a higher demand for better vehicle quality at higher
levels of income.
Pfaff et al. (2001) examined the linkage between income and house-
hold choices and its impact on the environment. The model they used
emphasized two main features. First, environmental degradation was a
by-product of the household activities. Second, households could recog-
nize their activities, and therefore would substitute for more expensive,
less environmentally degrading commodities, with higher income. They
highlighted that this would occur with the assumption that environmental
quality was a normal good and given the elimination of natural constraints
of desirability and feasibility of substitution. They found a non-monotonic
relationship between income and environmental quality, generating an
EKC at the household level as substitution constraints were eliminated.
In another paper, Pfaff et al. (2002) provided an explanation for the
EKC using their household-choice framework. They showed that at low
income levels, the marginal rate of substitution between household con-
sumption and the environment made abatement undesirable for the house-
hold. As income increased, both consumption and abatement expenditure
rose. They concluded that both household consumption preferences and
abatement technologies were necessary and sufficient conditions to gener-
ate the EKC path. They also applied their model to a set of externalities
and multiple agents who voted for environmental spending and taxation.
They found evidence for the EKC. As the chosen tax rate increased with
income, abatement occurred after a certain range, and therefore environ-
mental quality deteriorated at first and then improved.
Also Chaudhuri and Pfaff (2002), using the above model, studied the
linkages between household fuel-choice preference and income on the envi-
ronment, specifically indoor air pollution, in Pakistan. The individual and
house-level data were obtained from the Pakistan Integrated Household
Survey (1991), which covered the energy consumption of 4800 households.
To derive their sample, they used a multi-state stratified sampling proce-
dure from the Federal Bureau of Statistics based on the 1981 census. The
authors found significant evidence of an inverted U-shaped relationship
between monthly household income and monthly household emissions, in
terms of the quantity of the monthly fuel consumption due to the transi-
tion in the fuel-choice behaviour of the household. As per capita income
increased, the household moved from using traditional fuel and invested
in cleaner modern fuel. However, for all modern fuel consumption, their
quantities increased with higher incomes, forming a concave curve.
In a recent study by Khanna and Plassmann (2004), they examined the
impact of demand for environmental quality on the relationship between
income and pollution. They argued that the ability of the consumers to
separate themselves from the source of pollution played a decisive role
in the consumer’s decision to limit their exposure to pollution as income
increased. Also they argued that the turning point of the EKC would be
lower for pollutants where spatial separation was possible. Unlike Shafik
and Bandhopadhyay (1992) and Suri and Chapman (1998) who used
multi-country panel data that combined the effects of other factors, such
as structural changes, technology and consumer preferences for environ-
mental quality, Khanna and Plassmann used cross-sectional census-tract-
level data for the United States to isolate the effects of changes in income
from those in other factors. The census-tract-level data were the smallest
geographical unit for which detailed socioeconomic data were available.
The authors tested their hypothesis on five air pollutants: SO2, particulate
matter (PM), CO, ground level O3 and NOx. They used ambient con-
centrations data from the EPA AIRS database. They found an EKC
relationship for SO2 and PM, with turning points of $8653 and $11 412,
respectively. For the other three pollutants, the income–pollution relation-
ship continued to increase.
The authors concluded that the EKC hypothesis was an equilibrium
relationship between income and pollution based on the interaction
between consumer preferences and technology. For pollutants where
spatial separation was relatively straightforward, the opportunity cost of
pollution abatement was relatively low, and therefore the turning point of
the EKC or the equilibrium income elasticity of pollution, as referred to

by the authors, changed its sign from positive to negative at lower income
levels. However, for pollutants, where spatial separation was more dif-
ficult and costly, the equilibrium income elasticity of pollution remained
positive.
The role of technology

Technology shifts is one of the explanations conjectured for the economic-
growth–environment relationship. Hence, some of the empirical results
were attributed to the use of cleaner technology. Komen et al. (1997)
examined the increase in income and its role in promoting new technolo-
gies to improve environmental quality. They gathered data from 19 OECD
countries for the period 1980 to 1994. The countries were Canada, United
States, Australia, Austria, Belgium, Denmark, Finland, France, Germany,
Greece, Ireland, Italy, Norway, Spain, Sweden, United Kingdom, the
Netherlands, Portugal and Japan. They carried out an empirical analysis
focusing on the relationship between real budget per capita allocated for
public research and development (PRD) and real gross per capita GDP.
The PRD, for each country, was measured as nominal expenditures
converted to 1991 US dollars at prevailing purchasing power parities.
Similarly, GDP for each country was measured in thousands of US dollars
at prevailing purchasing power parities. The results of their study showed
a positive income elasticity of public research and development funding for
environmental protection. The income elasticity was approximately equal
to unity, which implied that emissions of some pollutants might decline
after an income-level threshold was reached. The authors concluded that
their results should be interpreted with caution due to (1) the small size of
funding allocated to public research and development relative to the size
of overall spending on environmental protection, (2) the limited ability of
the country to substitute between public and private research and develop-
ment spending and alternative policy instruments and (3) the possibility
that funding for public research and development might act as a form of
industrial subsidy in some countries. The study made two main contribu-
tions. First, unlike previous research that focused on the overall linkage
between per capita income and environmental degradation, this study
focused on research and technology development, which were the underly-
ing individual components. Second, it helped to explain why the EKC had
a negative slope for some pollutants in relatively industrialized countries.
Carson et al. (1997) did not study the role of technology per se in the
income–environment relationship, however, they attributed their findings
to the use of the latest-vintage technology. Carson et al. examined the
existence of an EKC relationship between income and seven types of air
emissions across the 50 US states. In order to overcome the problem of

comparability and quality of available environmental data, they moved
away from developing countries and used data from the 50 US states.
They also used state-level emissions for the seven major air pollutants
(greenhouse gases, air toxics, carbon monoxide, nitrogen oxides, sulphur
dioxide, volatile organic carbon and particulate matter). They estimated
the relationship using a panel data set consisting of the different states
over time. In addition, access to technology and air pollution control
regulations were likely to be the same across the United States, or at
least the differences would be less than that across different countries.
The income data used were expressed in thousands of 1982 US dollars.
They arrived at two main findings. First, all seven pollutants decreased
with increase in income even when industrial composition and popula-
tion density/urbanization were controlled for. Second, high-income states
had low per capita emissions while emissions in low-income states were
highly variable. They attributed this finding to two main factors. First,
richer countries produced goods using less polluting technology per unit
of output, either in the form of designing technology for this purpose or
using the latest technology, which was usually efficient in terms of energy
consumption. Second, consumers demanded more environmental quality
as they became wealthier, such as shifting their consumption to less pollut-
ing products, moving to cleaner areas or demanding that their government
strictly regulate the output of pollution.
De Bruyn et al. (1998) examined the relationship between economic
growth rate and environmental quality for the Netherlands, West
Germany, the United Kingdom and the United States for various time
intervals between 1960 and 1993. They used emissions data for CO2,
NOx and SO2. The CO2 emissions data were taken from the Oak Ridge
National Laboratory (1994), while the SO2 and NOx data were taken
from the US Environmental Protection Agency (National Air Quality and
Emission Trends Report, 1984) and OECD (Environmental Data Report,
1993). They found a positive and significant effect of economic growth on
the growth of emissions in all of the cases, with the exception of SO2 in
the Netherlands. The analysis suggested that in only half the cases was the
rise in income the reason for the reduction of emissions. They also found
a positive correlation between time patterns of emissions and economic
growth. The authors concluded that there was no evidence in their country
analysis to confirm that economic growth would improve environmental
quality. However, the relationship between income and emissions that
could be reduced by using cleaner technology would follow an inverted
U-shaped pattern when using panel data. The decline in emissions over
time would be due to technological and structural changes.
Decomposition and structural models of the income–environment

relationship
A number of studies attempted to understand the underlying determinants
of the relationship between economic growth and the environment. In
doing so they used more structural and analytical models in their studies
instead of the reduced-form models, in order to identify the structural
forces that would influence the relationship. Panayotou (1997), in examin-
ing the relationship between income and the level of SO2 using panel data
for 30 countries, identified three forces/effects influencing environmental
quality: (1) the scale effect, (2) the composition effect and (3) the abate-
ment effect. He found that expansion of the scale of an economy increased
the level of SO2 concentrations monotonically, but at a diminishing rate.
It was particularly strong at income levels up to $3 million per square
kilometre. The composition effect, represented by the industry share of
GDP, also monotonically increased the level of SO2 concentrations as its
share of GDP increased from 20 per cent to 43 per cent at income levels up
to $8000, then declined to 37 per cent when income levels reached $17 000
and then started to rise again. The abatement effect, which is the ‘pure’
income effect after being stripped out of the scale and composition effect,
had a negative relationship with SO2 concentrations up to income levels of
about $13 000.
Similarly, in a study, Islam et al. (1999), identified the same three forces
in examining the relationship between GDP per unit of area and ambient
level of SPM using data from the GEMS. The data contained 901 obser-
vations from 23 countries, including 56 cities, for the period 1977 to 1988.
They also used data from the World Bank Tables. The per capita GDP
data were in 1985 US dollars purchasing power parity. The authors found
that the level or scale effect increased monotonically, showing a positive
relationship between ambient level of SPM and the level of GDP per unit
of area. The composition effect showed an inverted U-shape with a hump-
shaped relationship between the level of SPM and the share of industry in
GDP. However, the peak was reached at a very high industry share level
at which the SPM level increased almost steadily. The abatement effect
generally declined.
Moomaw and Unruh (1997) tried to provide insight into the processes
that generated the changes in environmental quality and national economic
growth. They examined the structural transition changes in per capita CO2
emissions and per capita GDP over 16 countries for the period 1950 to
1992 and compared the structural transition models to EKC models. They
used the data from the Oak Ridge National Laboratory (1995) for the CO2
emissions and the Penn World Table for the income data in real per capita
GDP, measured in 1985 US dollars. The 16 countries were characterized
by having sustainable income growth with a stable or decreasing level of

CO2 emissions per capita over time. From their findings they made three
conclusions. First, the transition initiated in the countries did not correlate
to changes in income levels but to historic exogenous events, such as oil
price shocks in the 1970s, and policies. Second, the positive CO2 emissions
elasticities (also called marginal propensity to emit) in these countries hap-
pened as a sudden, discontinued transition, instead of a gradual change,
and therefore, the decrease in CO2 emissions did not correlate to specific
income levels, but to specific points in time. Third, the relationship had
an N-shape, which was the result of data aggregation instead of income
dependence. The turning points were $12 813 and $18 333, which implied
a very narrow range for CO2 to decline as emissions increased again once
the second turning point was passed.
They concluded that neither the U- nor the N-shaped relationship
between CO2 emissions and income provided a reliable indication that
at low income levels, environmental quality would deteriorate and then
would improve above a certain income threshold or transition value,
and therefore the model was inappropriate for forecasting future emis-
sions behaviour. They also added that CO2 emissions would continue to
rise as economic growth was pursued in countries. Moomaw and Unruh
provided two main explanations for their conclusions. First, individuals
would not sacrifice consumption for investment in environmental quality
and that was why pollutant emissions increased at low income levels. This
was because the environment was assumed to be a luxury good, and there-
fore individuals would only be willing to trade consumption for improve-
ments in environmental quality at high levels of income. Second, CO2
emissions created global and not local disutility to the public, who would
only demand controls on the level of pollutants that would create disutility
at the local level.
Also Moomaw and Unruh (1998) evaluated whether income was the
determining variable for the reduction of CO2 emissions by applying tech-
niques of non-linear dynamical analysis. This technique was used in order
to account for temporal patterns and discontinuous changes that might
have taken place. The study was conducted on 16 OECD countries that
showed an EKC pattern in previous studies using data from Penn World
Table (1994). The countries were Austria, Belgium, Canada, Denmark,
Finland, France, West Germany, Iceland, Italy, Japan, Luxembourg, the
Netherlands, Sweden, Switzerland, United Kingdom and United States.
The per capita income used was measured in 1985 US dollars PPP. The
authors arrived at two conclusions. First, income was not the determin-
ing variable for reducing CO2. They added that emissions were expected
to follow a regular, incremental path until they were subjected to a shock
that would lead to a new trajectory. In the case of CO2, the oil price shock
was the impetus that changed its trajectory. Second, a national capacity
existed for a rapid and persistent change under the proper stimuli. In the
case of CO2, the change in its trajectory happened within a single year and
continued at a particular emission level despite the continued economic
growth and the decline in oil prices.
Stern and Common (2001) criticized the EKC models by saying that they
lacked adequate specification due to the omissions of a number of variables
that were correlated with GDP. Their criticism was based on the results of
statistical tests from their study on the relationship between sulphur emis-
sions and per capita GDP. They applied fixed and random effects models
to a subset of panel data from 73 countries over the period 1960 to 1990.
They used data from the ASL and Associates (1997) database, consisting
of data on sulphur emissions for most of the countries of the world over
the period 1850 to 1990. An EKC relationship was found in both models,
however, the turning points were much lower when using data for only the
OECD countries (fixed effects: $9239, random effects: $9181), than when
estimating EKC for the whole sample (fixed effects: $101 166, random
effects: $54 199). Similarly the estimated turning points were extremely
high for non-OECD countries (fixed effects: $908 178, random effects:
$343 689), implying a monotonic EKC relationship between per capita
income and sulphur emissions in the case of non-OECD countries and the
whole sample. They said that their findings suggested that the inclusion of
trade variables would be important as OECD countries could outsource
the production of pollution-intensive products to the rest of the world.
However, inclusion of global macroeconomic trends and shocks, such as
the oil crisis of the 1970s, seemed unimportant.
In a study by Halkos (2003) using the same sample data from ASL and
Associates (1997) as Stern and Common (2001), he empirically tested
the relationship between environmental damage from sulphur emissions
and per capita GDP. He applied two econometric methods: random coef-
ficients and the Arellano–Bond Generalized Method of Moments (A–B
GMM). The per capita GDP data were derived from the Penn World
Table in Summers and Heston (1991), together with the population data.
The sample of countries chosen by Halkos in his study represented 81
per cent of the world’s population. He found an EKC pattern relation-
ship in the case of an A–B GMM model with turning points well within
the sample for all cases (global, OECD and non-OECD), ranging from
$2805 to $6230. However, unlike the results found by Stern and Common
(2001), the turning point for only the OECD countries was higher than
those for the global and non-OECD countries. On the other hand, no
support for the EKC hypothesis was found using the random coefficients
model. The author concluded that the econometric techniques adopted in

the empirical studies were crucial to the results and the level of the turning
points of the EKC.
Ecological thresholds and sustainability

There have been a number of criticisms of the Environmental Kuznets
Curve as it suggests that countries can overcome their environmental
problems for certain pollutants simply by economic growth without
paying attention to the environment and ecological thresholds. The criti-
cisms were started by Arrow et al. (1995), who attributed the existence of
the relationship between income and a selected set of pollutants to the fact
that people in poor countries could not afford to emphasize environmental
amenities over material well-being. As their standard of living improved
and reached a sufficiently high level, their demand for environmental
amenities increased. This led the governments to create environmental
legislation and institutions to protect the environment. They also said
that while the inverted U-shaped curve indicated that while improvements
of some environmental indicators could be associated with economic
growth, it did not mean that economic growth was sufficient to induce
environmental improvement. Moreover, the effects of economic growth
on the earth’s resource base should not be ignored as it could not support
indefinite economic growth. In addition, irreversible degradation in this
base could put economic development at risk.
Furthermore, Arrow et al. highlighted that in interpreting the inverted
U-shaped curves, a number of factors should be taken into account. First,
research showed that the inverted U-shaped relationship between income
and environmental quality was only valid for pollutants involving local
and short-term costs (for example, sulphur, particulates and faecal colif-
orms), not for those involving long-term and more dispersed costs (such
as CO2). The latter type of pollutants showed in most research that they
had an increasing function with income. Second, the relationship was valid
for emissions of pollutants and not stocks of resources of the earth’s base,
such as soil, forests and other ecosystems. Third, the inverted-U curve
relationship did not explain the effects of reductions in emissions on the
wider system, in terms of increasing other emissions in the same country
or transferring pollutants to other countries. Fourth, in most cases the
reduction in emissions as income increased was due to local institutional
reforms, such as environmental legislation and governmental policies,
which often ignored the international and intergenerational consequences
on other countries or future generations.
A number of environmentalists and ecological economists commented
on the article by Arrow et al., including Ayres (1995). He rejected the
proposition that economic growth improved the environment due to

reasons other than the ones mentioned by Arrow et al. He referred in
his argument to two well-known relationships: (1) the close correlation
between economic growth and the increase in energy consumption and
(2) most of the environmental problems were directly traceable to the
unsustainable use of fossil fuels and/or other materials, such as toxic heavy
metals and chlorinated chemicals, being a potential waste according to the
basic law of conservation.
Stern et al. (1996) critically examined the concept of the EKC and they
concluded that there were three major generic problems to the estimation
and testing of EKCs. First was the assumption of unidirectional relation-
ship from economic growth to environmental quality and no feedback
from the environment to production. Second was the assumption that
international trade associated with development did not affect environ-
mental quality, while standard trade theory implied that under free trade,
developing countries would specialize in producing labour-intensive goods
and use their natural resources, while developed countries would special-
ize towards human capital and manufacture of capital-intensive goods.
This specialization and the existence of environmental regulation would
reduce environmental degradation in developed countries while increasing
it in developing ones. In other words, developed countries would migrate
polluting activities to poorer countries by increasing substitution towards
importing manufactured goods. They also added that historical experi-
ences of some economies could not be extrapolated to the whole global
economy because once poor countries adopted similar levels of environ-
mental policies, they would be faced with the problem of abating these
activities as there would be no unregulated countries to migrate pollution
to. Therefore, the whole world could not achieve a similar transformation.
Third, the data on environmental problems were patchy in coverage and
poor in quality, and therefore were insufficient to provide conclusions or
to project future trends. They suggested that in order to project future
trends using econometric analysis, they had to take the form of structural
models rather than the generally reduced-form equations used to examine
the EKC hypothesis. Such models would inform choices for policy-makers
and would require policy adjustments in order to sustain environmental
development of the global economy.
Farber (1995) also criticized the notion that economic growth induced
improvements in environmental quality as naive as it was based on obser-
vations of only a limited number of environmental variables, which were
local pollutants, and ignored global pollutants and their impact on the
environment. Moreover, most findings resulted from empirical research
and observations using cross-sectional country data. This was very limited
in terms of the long-term, time-series relations across countries. In other

words, any country could not be expected to follow the cross-country
data points as its income increased. Furthermore, all countries with low
income levels could not be expected to grow as (1) their environmental
capital resource could be so depleted that economic growth would not
be feasible and (2) the empirical relation did not measure the extent to
which high-income countries substitute environmental degradation in
poorer countries in the process of economic growth. Farber concluded
that environmental and economic policies could not be set based on this
relationship.
The answer to the U-shaped relationship between income and envi-
ronmental quality was attributed by Page (1995) to the idea of harmony.
This concept occurred in a number of studies, such as population studies,
where there was an empirical claim of demographic transition at a low
rate of economic development, then the rate of population growth would
slowly increase up to a certain level beyond which further increase in
economic growth and development would lead to reduction in the popula-
tion growth rate to environmentally sustainable levels. Another example
was the Laffer curve, which suggested that with very low tax rates there
would be little government revenue, with increasing tax rates the govern-
ment revenue would increase, but with further increasing tax rates the
government revenue would decrease. A third example was the scarcity
and growth argument by Barnett and Morse (1961), which stated that as
depletion in natural resources continued, people would anticipate future
scarcities, therefore prices would rise, and thereby incentives would be
created for developing new technologies and substitutes, and hence the
resource base would be renewed. In other words, natural competition
among the individual players in the market would take care of the larger
environment. A fourth example is Coase’s policy (1990), which stated that
if bargaining costs were kept low enough, the results would be close to
efficient and good for the society as a whole.
The idea of harmony in economics goes back to Adam Smith’s invis-
ible hand (1776 [1994]). Nowadays, this idea is also strongly present in the
form of structural adjustment and trade liberalization and other reforms,
which aim at lower taxes, less regulation, freer markets and free trade. In
the last decade, this resulted in shifting economic policies in International
Financial Institutions (IFIs), such as the World Bank, International
Monetary Fund, and so on, towards developing countries. It also resulted
in large policy shifts in the world as a whole.
Concern regarding the structural adjustment programmes and eco-
nomic reforms that took place in poor countries to accelerate economic
growth was expressed by Munasinghe (1999). He argued that economic
reforms could be at the expense of violating ecological thresholds, espe-

cially during the crossing-over stage of the EKC. The author said that eco-
nomic reform was good for both the economy and the environment only if
the reform introduced complementary environmental measures as part of
the reform package in order to address the imperfections associated with
it to mitigate environmental harm. Addressing such imperfections, which
include policy distortions, market failures and institution constraints,
would allow the reform programme to go forward without adverse
impacts on the environment. Finally, the author recommended (1) caution
in introducing policy reforms with great consideration to the timing and
sequencing of the policies introduced and (2) addressing specific distor-
tions in the economy rather than a wide economic reform programme with
fewer environmental gains.
Panayotou (2000) argued that the EKC relationship between economic
growth and the environment was not optimal or inevitable. He gave three
reasons. First, the environment did not improve because of higher income
levels, but as a result of environmental legislation and institutions due
to the higher demand for better environmental quality. It was the policy
intervention that determined the turning of the shape and height of the
EKC. Second, the damage that could accumulate in a country until the
environmental improvements were realized could exceed by far the present
value of economic growth and better environmental quality. Again envi-
ronmental policy was necessary to mitigate the environmental damage at
early stages of development. Third, the height of the EKC, which reflected
the environmental damage or what the author called ‘environmental price’
as a result of economic growth, was determined by the effectiveness of
policies and efficiency of markets. The higher the externalities, undefined
property rights and harmful subsidies, the higher the environmental price
as a result of economic growth. In other words, market and policy fail-
ures resulted in unnecessary environmental degradation and economic
inefficiency. Panayotou argued that better management of the resource
base, especially in the presence of ecological thresholds that could be irre-
versible, would result in an optimal economic growth and environmental
improvement.
The process of dematerialization, which was defined as the ‘unlinking
of income and the use of nature’, was examined by Canas et al. (2003).
Dematerialization was related to the concept of strong sustainability:
the maintenance of natural capital. It could mean an absolute reduction
in the use of material (strong dematerialization) or just a reduction in
the material intensity of income (weak dematerialization). They tested the
EKC hypothesis using the direct material input (DMI) per capita as the
dependent variable and per capita GDP as the independent variable for
16 industrialized countries, using panel data over the period 1960 to 1998.
Their results showed robust support for weak dematerialization in indus-
trialized countries, whereby the aggregate material intensity, measured
in mass units, reduced as per capita income levels increased. The authors
suggested that this relationship could be due to (1) the higher material use
at lower levels of income to respond to infrastructural needs and (2) the
structural change within the economy towards an increase in the service
sector as income increased.
Dinda (2004) reviewed a number of studies on the EKC hypothesis and
ended up by saying that the subject was open-ended and the existence of
the EKC was inconclusive. He also highlighted several areas that required
further research. First, the choice of economic models was important
to properly reflect the physical and ecological aspects of the economy.
Second, unfolding the underlying factors behind the EKC was a priority
for any research. Third, moving towards structural rather than reduced-
form models and decomposition analysis might be needed. Fourth, using
of time series rather than panel data was essential to identify the develop-
ment of pollution associated with economic growth in individual coun-
tries. Finally, determining the policy measures and regulations necessary
to allow sustainable development.
Conclusion
This chapter reviewed the extensive research conducted in examining
the relationship between economic growth and the environment. Earlier
studies on the income–environment relationship used the reduced-form
single-equation approach to show the relationship between economic
growth and environmental degradation, without understanding the
dynamics of this relationship. In order to test whether income was the
determinant of the behaviour of the environmental indicator(s), recent
studies moved towards decomposition analysis and more structural and
analytical models to identify the real factors or influences behind the
income–environment relationship. Most of the studies used cross-sectional
country or panel data instead of time series data across countries due to
the difficulty and limited availability of the latter. Furthermore, although
some research explored the role of policies and institutions, a limited
number of empirical studies examined the influence of international envi-
ronmental legislation on the relationship between economic growth and
the environment.
References
Andreoni, J. and A. Levinson (1998), ‘The Simple Analytics of the Environmental Kuznets
Curve’, National Bureau of Economic Research, Inc, NBER Working Paper No. 6739.
Antle, J.M. and G. Heidebrink (1995), ‘Environment and Development: Theory and
International Evidence’, Economic Development and Cultural Change, 43(3), 603–25.
Arrow, K., B. Bolin, R. Costanza, P. Dasgupta, C. Folke, C.S. Holling, B.-O. Jansson, S.
Levin, K.-G. Maler, C. Perrings and D. Pimentel (1995), ‘Economic Growth, Carrying
Capacity, and the Environment’, Ecological Economics, 15(2), 89–95.
Ayres, R.U. (1995), ‘Economic Growth: Politically Necessary but not Environmentally
Friendly’, Ecological Economics, 15(2), 97–9.
Barnett, H. and C. Morse (1961), Scarcity and Growth: The Economics of Natural Resource
Availability, Baltimore, MD: Johns Hopkins Press.
Barrett, S. and K. Graddy (2000), ‘Freedom, Growth, and the Environment’, Environment
and Development Economics, 5(1), 433–56.
Beckerman, W. (1992), ‘Economic Growth and the Environment: Whose Growth? Whose
Environment?’, World Development, 20(4), 481–96.
Bhattarai, M. and M. Hammig (2001), ‘Institutions and the Environmental Kuznets Curve
for Deforestation: A Cross-country Analysis for Latin America, Africa and Asia’, World
Development, 29(6), 995–1010.
Brundtland, G. (1987), Our Common Future: The World Commission on Environment and
Development, Oxford, UK: Oxford University Press.
Canas, A., P. Ferrao and P. Conceicao (2003), ‘A New Environmental Kuznets Curve?
Relationship between Direct Material Input and Income per Capita: Evidence from
Industrialized Countries’, Ecological Economics, 46(2), 217–29.
Carson, R.T., Y. Jeon and D.R. McCubbin (1997), ‘The Relationship between Air
Pollution Emissions and Income: US Data’, Environment and Development Economics,
2(4), 433–50.
Cavlovic, T.A., K.H. Baker, R.P. Berrens and K. Gawande (2000), ‘A Meta-analysis of
Environmental Kuznets Curve Studies’, Agricultural and Resource Economics Review,
29(1), 32–42.
Chaudhuri, S. and A. Pfaff (2002), ‘Economic Growth and the Environment: What Can
We Learn from Household Data’, School of International and Public Affairs, Columbia
University, Paper No. 0102-51.
Coase, R.H. (1990), The Firm, the Market, and the Law, Chicago: University of Chicago
Press.
Cole, M.A. (2004), ‘Trade, the Pollution Haven Hypothesis and the Environmental Kuznets
Curve: Examining the Linkages’, Ecological Economics, 48(1), 71–81.
Cole, M.A., A.J. Rayner and J.M. Bates (1997), ‘The Environmental Kuznets Curve: An
Empirical Analysis’, Environment and Development Economics, 2(4), 401–16.
Common, M. (1995), Sustainability and Policy: Limits to Economics, Cambridge, UK:
Congleton, R.D. (1992), ‘Political Institutions and Pollution Control’, Review of Economics
and Statistics, 74(3), 412–21.
Cropper, M. and C. Griffiths (1994), ‘The Interaction of Population Growth and
Environmental Quality’, Population Economics, 84(2), 250–54.
De Bruyn, S.M. (1997), ‘Explaining the Environmental Kuznets Curve: Structural Change
and International Agreements in Reducing Sulphur Emissions’, Environment and
Development Economics, 27(2), 485–503.
De Bruyn, S.M., J.C.J.M. Van Den Bergh and J.B. Opschoor (1998), ‘Economic Growth
and Emissions: Reconsidering the Empirical Basis of Environmental Kuznets Curves’,
Deacon, R. (1999), ‘The Political Economy of Environmental Development
Relationships’, Santa Barbara, Preliminary Framework Working Paper, University
of California.
Dinda, S. (2004), ‘Environmental Kuznets Curve Hypothesis: A Survey’, Ecological
Economics, 49(4), 431–55.
Farber, S. (1995), ‘Economic Resilience and Economic Policy’, Ecological Economics, 15(2),
105–7.
Galeotti, M. and A. Lanza (1999), ‘Desperately Seeking (Environmental) Kuznets’,

Discussion Paper, International Energy Agency.
Gastil, R.D. (1987), Freedom in the World: Political Rights and Civil Liberties 1986–1987,
New York: Greenwood Press.
Goodstein, E. (1999), Economics and the Environment, Englewood Cliffs: Prentice Hall.
Grossman, G.M. and A.B. Krueger (1991), ‘Environmental Impacts of a North American
Free Trade Agreement’, Discussion Papers in Economics, Woodrow Wilson School of
Public and International Affairs, Princeton, NJ.
Grossman, G.M. and A.B. Krueger (1995), ‘Economic Growth and the Environment’,
Quarterly Journal of Economics, 110(2), 353–77.
Grossman, G.M., J.A. Laity and A.B. Krueger (1994), ‘Determinants of Air Pollution in
US Countries’, Discussion Papers in Economics, Woodrow Wilson School of Public and
International Affairs, Princeton, NJ.
Halkos, G.E. (2003), ‘Environmental Kuznets Curve for Sulfur: Evidence Using GMM
Estimation and Random Coefficient Panel Data Models’, Environment and Development
Economics, 8(4), 581–601.
Hettige, H., R.E.B. Lucas and D. Wheeler (1992), ‘The Toxic Intensity of Industrial
Production: Global Patterns, Trends, and Trade Policy’, American Economic Review,
82(2), 478–81.
Hettige, H., M. Mani and D. Wheeler (1997), ‘Industrial Pollution in Economic Development:
Kuznets Revisited’, Policy Research Working Paper No. WPSIF76, World Bank.
Hilton, H.F.G. and A. Levinson (1998), ‘Factoring the Environmental Kuznets Curve:
Evidence from Automotive Lead Emissions’, Journal of Environmental Economics and
Management, 35(2), 126–41.
Holtz-Eakin, D. and T.M. Selden (1995), ‘Stoking the Fires? CO2 Emissions and Economic
Growth’, Journal of Public Economics, 57(1), 85–101.
IBRD (1992), World Development Report, ‘Development and the Environment’, New York:
Oxford University Press.
Islam, N., J. Vincent and T. Panayotou (1999), ‘Unveiling the Income–Environment
Relationship: And Exploration into the Determinants of Environmental Quality’, Working
Paper, Department of Economics and Harvard Institute for International Development.
Jaggers, K. and T.R. Gurr (1995), ‘Tracking Democracy’s Third Wave with the Polity III
Data’, Journal of Peace Research, 32, 469–82.
Kahn, M.E. (1998), ‘A Household Level Environmental Kuznets Curve’, Economics Letters,
59(2), 269–73.
Kaufman, R., B. Davidsdotter and D. Garnham (1998), ‘The Determinants of Atmospheric
SO2 Concentration: Reconsidering the Environmental Kuznets Curve’, Ecological
Economics, 25(2), 209–20.
Khanna, N. and F. Plassmann (2004), ‘The Demand for Environmental Quality and the
Environmental Kuznets Curve Hypothesis’, Ecological Economics, 51(3/4), 225–36.
Klaassen, G. (1995), ‘Trading Sulfur Emission Reduction Commitments in Europe: A
Theoretical and Empirical Analysis’, International Institute for Applied System Analysis
(IIASA), Laxenburg, Austria.
Knack, S. and P. Keefer (1995), ‘Institutions and Economic Performance: Cross Country
Tests Using Alternative Institutional Measures’, Economic and Politics, 7(3), 207–27.
Komen, M.H.C., S. Gerking and H. Folmer (1997), ‘Income and Environmental R&D:
Empirical Evidence from OECD Countries’, Environment and Development Economics,
2(4), 505–15.
Koop, G. and L. Tole (1999), ‘Is there an Environmental Kuznets Curve for Deforestation?’,
Journal of Development Economics, 58(1), 231–44.
Kuznets, S. (1955), ‘Economic Growth and Income Inequality’, American Economic Review,
45(1), 1–28.
Marland, G. and R.M. Rotty (1984), ‘Carbon Dioxide Emissions from Fossil Fuels: A
Procedure for Estimation and Results for 1950–1982’, Tellus, 36(B), 232–61.
Marland, G. R.J. Andres, T.A. Boden, C. Johnson and A. Brenkert (1999), Global Regional
and National CO2 Estimates from Fossil Fuel Burning, Cement Production, and Gas-flaring:
1751–1996, Oak Ridge, TN: Carbon Dioxide Information Analysis Center.
McConnell, K.E. (1997), ‘Income and the Demand for Environmental Quality’, Environment
and Development Economics, 2(1), 383–99.
Meadows, D.H., D.L. Meadows, J. Randers and W.W. Behrends (1972), The Limits to
Growth, London: Earth Island Limited.
Midlarsky, M.I. (1998), ‘Democracy and the Environment: An Empirical Assessment’,
Journal of Peace Research, Special Issue on Environmental Conflict, 35(3), 341–61.
Moomaw, W.R. and G.C. Unruh (1997), ‘Are Environmental Kuznets Curves Misleading
Us? The Case of CO2 Emissions’, Environment and Development Economics, 2(4),
451–63.
Moomaw, W.R. and G.C. Unruh (1998), ‘An Alternative Analysis of Apparent EKC-type
Transactions’, Ecological Economics, 25(2), 221–9.
Munasinghe, M. (1999), ‘Is Environmental Degradation an Inevitable Consequence of
Economic Growth: Tunneling through the Environmental Kuznets Curve’, Ecological
Economics, 29(1), 89–109.
Page, T. (1995), ‘Harmony and Pathology’, Ecological Economics, 15(2), 141–4.
Panayotou, T. (1993), ‘Empirical Tests and Policy Analysis of Environmental Degradation
at Different Stages of Economic Development’, World Employment Programme Research
Working Paper No. WEP2-22/WP 238, Geneva: International Labour Force.
Panayotou, T. (1997), ‘Demystifying the Environmental Kuznets Curve: Turning a Black
Box into a Policy Tool’, Environment and Development Economics, 2(4), 465–84.
Panayotou, T. (2000), ‘Economic Growth and the Environment’, CID Working Paper No.
56, Center for International Development, Cambridge, MA: Harvard University.
Panayotou, T., J. Sachs and Peterson (1999), ‘Developing Countries and the Control of
Climate Change: A Theoretical Perspective and Policy Implications’, CAER II Discussion
Paper No. 45, August.
Perrings, C.A. and A. Ansuategi (2000), ‘Sustainability, Growth and Development’, Journal
of Economic Studies, 27(1/2), 19–55.
Pfaff, A.S.P., S. Chaudhuri and H.L.M. Nye (2001), ‘Why Might One Expect Environmental
Kuznets Curves? Examining the Desirability and Feasibility of Substitution’, Columbia
University, 9 July, Paper No. 0102-45.
Pfaff, A.S.P., S. Chaudhuri and H.L.M. Nye (2002), ‘Endowments, Preferences, Abatement
and Voting: Microfoundations of Environmental Kuznets Curves’, Columbia University,
Paper No. 0102-46.
Ravallion, M., M. Heil and J. Jalan (1997), ‘A Less Poor World, but a Hotter One?
Carbon Emissions, Economic Growth and Income Inequality’, Working Paper No. 13,
Washington, DC: World Bank.
Roberts, J.T. and P.E. Grimes (1997), ‘Carbon Intensity and Economic Development
1962–91: A Brief Exploration of the Environmental Kuznets Curve’, World Development,
25(2), 191–8.
Roca, J. (2003), ‘Do Individual Preferences Explain the Environmental Kuznets Curve?’,
Rock, M. (1996), ‘Pollution Intensity of GDP and Trade Policy: Can the World Bank Be
Wrong?’, World Development, 24(3), 471–9.
Rothman, D.S. (1998), ‘Environmental Kuznets Curves – Real Progress or Passing the Buck?
A Case for Consumption-based Approaches’, Ecological Economics, 25(2), 177–94.
Schmalensee, R., T.M. Stoker and R.A. Judson (1998), ‘World Carbon Dioxide Emissions:
1950–2050’, The Review of Economics and Statistics, 80(1), 15–27.
Selden, T.M. and D. Song (1994), ‘Environmental Quality and Development: Is There a
Kuznets Curve for Air Pollution Emissions?’, Journal of Environmental Economics and
Management, 27(2), 147–62.
Selden, T.M. and D. Song (1995), ‘Neoclassical Growth, the J Curve for Abatement, and the
Inverted U Curve for Pollution’, Journal of Environmental Economics and Management,
27(2), 162–8.
Shafik, N. (1994), ‘Economic Development and Environmental Quality: An Econometric

Analysis’, Oxford Economic Papers, 46(5), 757–73.
Shafik, N. and S. Bandyopadhyay (1992), ‘Economic Growth and Environmental Quality:
Time-series and Cross-country Evidence’, World Bank Working Papers No. WPS 904,
Washington, DC: World Bank.
Smith, A. (1776 [1994]), The Wealth of Nations, New York: Random House.
Stavins, R.N. (2000), ‘Economic Analysis of Global Climate Change Policy: A Primer’, John
F. Kennedy School of Government, Harvard University and Resources for the Future –
Working Paper No. 00-003.
Stern, D.I. and M.S. Common (2001), ‘Is There an Environmental Kuznets Curve for
Sulphur?’, Journal of Environmental Economics and Management, 41(2), 162–78.
Stern, D.I., M.S. Common and E.B. Barbier (1996), ‘Economic Growth and Environmental
Degradation: A Critique of the Environmental Kuznets Curve’, World Bank Development,
24(7), 1151–60.
Summers, R. and A. Heston (1991), ‘The Penn World Table (Mark 5): An Expanded Set
of International Comparisons, 1950–1988’, The Quarterly Journal of Economics, 106(2),
327–68.
Suri, V. and D. Chapman (1998), ‘Economic Growth, Trade and Energy: Implications for
the Environmental Kuznets Curve’, Ecological Economics, 25(2), 195–208.
Torras, M. and J. Boyce (1998), ‘Income, Inequality, and Pollution; A Reassessment of the
Environmental Kuznets Curve’, Ecological Economics, 25(2), 147–60.
Tuan, N.A. (1999), ‘Evidences of Environmental Kuznets Curve from CO2 Emissions in
Six-country Analysis’, Working Paper, Institute d’Economie et de Polique de L’Energie
(IEPE) BP 47, 38040 Grenoble Cedex 09.
Vincent, J.R. (1997), ‘Testing for Environmental Kuznets Curves within a Developing
Country’, Environment and Development Economics, 2(4), 417–31.
World Bank (1987), World Bank Development Report, New York: Oxford University Press.
13. Biodiesel as the potential alternative
vehicle fuel: European policy and global
environmental concern
Mahesh Poudyal and Jon C. Lovett
Introduction
The widely accepted principles of sustainable development are that the
present generation should be able to meet its own needs without compro-
mising the needs of future generations. Essentially this implies that we can
continue to have economic growth, but the means by which we achieve
this should not do so much damage that our children look back in anguish
and question our actions. In the last two centuries economic growth has
been powered by burning fossil fuels with the consequent release of carbon
dioxide. In recent years this release has been increasing rapidly: between
1961 and 2002 humanity’s carbon footprint grew more than 700 per cent
(Kitzes et al., 2007). In the last few decades there has been widespread
concern that observed increases in atmospheric carbon dioxide and global
temperatures are causally linked. Confidence in the general scientific
consensus has reached the point where policy-makers are willing to take
firm action, for example the 10th Session of Working Group I of the
Intergovernmental Panel on Climate Change (IPCC) in Paris (February
2007) concluded that:
Most of the observed increase in globally averaged temperatures since the

mid-20th century is very likely due to the observed increase in anthropogenic
greenhouse gas concentrations. . . . Discernible human influences now extend
to other aspects of climate, including ocean warming, continental-average tem-
peratures, temperature extremes and wind patterns. (Intergovernmental Panel
on Climate Change 2007, emphasis original)
Policy instruments for tackling greenhouse gas (GHG) induced warming

have been in place for some time. The United Nations Framework
Convention on Climate Change (UNFCCC) entered into force on 21
March 1994 and led to the 1997 Kyoto Protocol. Ratification of the
Kyoto Protocol by the European Union (EU) in 2002 meant that all the
Member States came under obligation to cut their GHG emissions. The
Protocol entered into effect on 16 February 2005 following ratification by
Russia in November 2004. Although it is still not ratified by the United
408
Biodiesel as the potential alternative vehicle fuel 409
States, a major carbon emitter, and big emitters like China and India are
not required to reduce their emissions at present, the policy and practices
within the EU – the largest economic bloc in the world – provide a much-
needed credibility to the Kyoto Protocol and impetus to development of
technological innovations that will allow present generations to have their
energy supply without leaving a drastically altered planet for the future.
Among various measures to cut GHG emissions within the EU, the use
of biofuels derived from agricultural or forestry products is considered
a viable alternative to fossil fuels, especially for the transport sector. As
such, the EU Directive 2003/30/EC set targets for biofuels to be used in
EU transport at 2 per cent by the end of 2005 and 5.75 per cent by the
end of 2010. This directive with other complementary directives, resolu-
tions and legislation on renewable fuels could play a major role in the
EU’s attempts to reduce GHG emissions from the transport sector. This
drive has been further strengthened by the recent announcement about the
minimum efficiency requirements set for vehicle manufacturers that limit
the emissions from the vehicles to be manufactured in the future.1
Substitution of biofuels for fossil fuels also helps to fulfil other policy
objectives. First, the world’s main oil reserves are located in geopoliti-
cally sensitive areas. Climate change notwithstanding, this is perhaps the
main incentive for switching from oil to a more costly fuel supply. The
economic rise of countries such as China and India mean that there
are major players in the global quest for natural resources that are not
necessarily inside the traditional European sphere of political influence.
Environmental considerations are thus in line with the strategic needs of
national security. Second, agricultural subsidies have historically been a
major European expenditure and source of considerable debate within the
Union. Support for growth of biofuel feedstocks could provide a bridge
to resolve differences between the Member States. Third, rapid expansion
of the European Union has led to economic disparities between the new
and old members. The political need for harmonization requires invest-
ment from the rich and markets for the poor. For example, Bulgaria and
Romania, who joined the Union in January 2007, have abundant agricul-
tural land and could potentially become major biofuel suppliers. Fourth,
biofuels open a huge new market for developing countries that have spare
land and cheap labour. Biofuel production could help these countries
meet the Millennium Development Goal of poverty alleviation through
economic growth based on primary production.
These potential multiple benefits of biofuel use make them attractive to
policy-makers, but there have been criticisms too. For example, although
some calculations demonstrate a positive energy balance for biofuel use
(Hill et al., 2006), other work suggests that the GHG footprint of biofuel
production may not be as attractive as hoped because biofuel production

requires considerable energy and fertilizer inputs and so does not meet
the primary policy goal of reduced GHG emissions (Dias de Oliveira et
al. 2005; Crutzen et al., 2007). Indeed, it may be that avoiding deforesta-
tion for agriculture and restoring forests is a better option (Righelato and
Spracklen, 2007). Second, a switch to intensive biofuel production on the
scale needed to supply energy markets will result in a transformation of
species-rich habitats such as tropical rainforests in biodiversity hotspots
(Koh, 2007) and so perhaps have a greater negative effect on biodiver-
sity than that of global warming. Third, replacing food production with
biofuel crops could cause market distortions and increase food prices
(Doornbosch and Steenblik, 2007). However, the situation here is compli-
cated by global changes in eating habitats, for example, increased intensive
livestock-rearing and meat consumption in China have been driving the
rise in soya bean production (a biofuel feedstock) and export in countries
such as Brazil (Naylor et al., 2005).
The outline of this chapter is as follows. The next section looks at the
EU renewable fuel policy broadly, and at its biofuels policy specifically.
The biofuel targets set by the EU and the policies in place to meet such
targets are analysed. Furthermore, we discuss the importance of biodiesel
in helping to meet the EU’s biofuels targets, and the issues related to pro-
duction of biodiesel within the EU and in other parts of the world in the
context of the EU’s biofuels policy. We then assess the costs and benefits
of biodiesel production and use, mainly in the transport market to which
it is geared. In addition, we critically review the results from a number of
studies that have looked into the emissions from vehicles using biodiesel.
Then we ask whether biofuels in general, and biodiesel in particular, have
the potential to provide the double benefit of being a secure and cheap
source of energy and at the same time being environmentally friendly – as
is often argued by its promoters. We assess the potential of biodiesel to
achieve this based on our review of the costs and benefits of biodiesel, as
well as on our analysis of the promotion of biofuels in the EU market and
its policies on biofuels. We conclude the chapter by summarizing the main
issues surrounding biodiesel at present, and looking into the future of this
particular type of renewable fuel, in the context of EU policy, climate
change issues and global energy demand.
EU renewable fuels policy and biodiesel

The European Union has been promoting production and use of renew-
able energy within its Member States, especially during the last decade. In
the past few years, a number of Council resolutions and directives have
been passed in its drive to promote renewable energy production and use
in the Member States (for example, The Council of the European Union,
1998; The European Parliament and the Council of the European Union,
2001, 2003). The Council Resolution of 1998 on renewable sources of
energy states:
there is need to promote a sustained and substantially increased use of renew-

able sources of energy throughout the Community in the light of the valuable
contribution renewables can make to environmental protection and the imple-
mentation of the commitments under the Kyoto Protocol, to security of supply
and the preservation of finite energy resources, and to economic and social
development generally, including in relation to employment and the strength-
ening of the economic structure of the outermost, isolated and island regions.
(The Council of the European Union, 1998, p. 1)
The 1998 Council Resolution was influenced by the Commission’s

‘White’ and ‘Green’ Papers on renewable sources of energy, and endorsed
an indicative target of 12 per cent renewable energy use for the Community
as a whole by 2010 as a useful guidance for all Member States pursuing
policies towards increasing renewable energy production and use (The
Council of the European Union, 1998, p. 1). This resolution also encour-
aged the Member States to ‘choose the most appropriate means of pro-
moting use of renewables’, suggesting a number of instruments, such as
subsidies, preferential tariffs and purchase obligations to name a few.
The 1998 resolution was followed by a number of other directives
promoting use of renewable energy sources, including one on biofuels or
other renewable fuels for transport – the Council Directive 2003/30/EC
(The European Parliament and the Council of the European Union, 2003).
Among various measures to cut GHG emissions within the EU, the use of
biofuels has been considered a potential alternative to reducing emissions,
especially from transport, as biofuels are seen as the most viable alterna-
tive to replace or complement fossil fuel. As such, the European Directive
2003/30/EC not only aims to promote the use of biofuels in the transport
sector, but also sets out the target for biofuels to be used in transport in the
EU Member States. According to the first article of the directive, it
aims at promoting the use of biofuels or other renewable fuels to replace diesel
or petrol for transport purposes in each Member State, with a view to con-
tributing to objectives such as meeting climate change commitments, environ-
mentally friendly security of supply and promoting renewable energy sources.
(Article 1, Directive 2003/30/EC)
Thus, the directive is especially concerned with the promotion of bio-

fuels as the viable alternative to the diesel or petrol used in transport in
the EU, and considers this as one of the means to meet its climate change
commitments as set out in the Kyoto Protocol. The directive lists ten
products it classifies as biofuels, including biodiesel and bioethanol – the
biofuels that are already available and used in significant proportions
around the world. In Article 3, this directive also sets minimum targets for
all the Member States, and encourages them to set their own ‘indicative
targets’.2
Furthermore, the directive provides flexibility as to how the biofuels for
transport are made available in the market, allowing for the supply of pure
biofuels (biodiesel and bioethanol) or blended with fossil fuels. Where the
biofuels are supplied as blends, the directive requires specific labelling of
such fuels only when the quantity of biofuels in such blends exceed 5 per
cent (Directive 2003/30/EC, OJ L 123, p. 45). Finally, the directive sets out
the guidelines for reporting the annual progress in meeting the set targets
by the Member States, before 1 July each year. The directive stated that
the Commission would draw up an evaluation report on the progress
made in the first phase by 31 December 2006, adding that:
If this [evaluative] report concludes that the indicative targets are not likely to
be achieved for reasons that are unjustified and/or do not relate to new scien-
tific evidence, these proposals shall address national targets, including possible
mandatory targets, in the appropriate form. (Directive 2003/30/EC, OJ L 123,
p. 46)
So, although the targets set by the directive were non-binding and more
as guidelines, if the Member States were not doing enough to promote
the use of biofuels in transport during the first phase as set out by the
directive, without justifiable reasons, the directive clearly stated that the
Commission could make the targets mandatory in the second phase.
Indeed, despite a relatively low minimum target set by the directive for the
first phase, it is now clear that the 2 per cent target for the minimum pro-
portion of biofuels and other renewable fuels that should have been placed
on the markets of the EC Member States by the end of 2005 has not been
met by the Member States except for Germany and Sweden (European
Commission, 2006, 2007b). According to the 2006 progress report, the
share of biofuels in the transport market for the 21 Member States where
biofuels are in use was only 1 per cent by the end of 2005. Although this
1 per cent market share is ‘a good rate of progress – a doubling in two
years’ according to the 2006 progress report, it is only half the target set
by Directive 2003/30/EC, and less than the 1.4 per cent share that would
have been achieved if all the Member States had met their annual indica-
tive targets (European Commission, 2007b, p. 6). Furthermore, both the
progress among the Member States and the rate of adoption of biofuels
seemed very uneven – with only Germany (3.8 per cent) and Sweden (2.2
per cent) exceeding the 2 per cent target. In terms of adoption of biofuels,
biodiesel was significantly ahead of ethanol – with biodiesel achieving a 1.6
per cent share of the diesel market, whereas ethanol achieved only a 0.4 per
cent share of the petrol market (European Commission, 2007b, p. 6).
The main policy instrument to facilitate meeting this target was the
EU Directive 2003/96/EC (The Council of the European Union, 2003)
on energy products taxation, which allowed Member States to reduce, or
apply total or partial exemption in the level of taxation to fuels from renew-
able sources. Moreover, schemes like Energy Crop Payments – a premium
payment of €45/ha for growing energy crops under the 2003 Common
Agricultural Policy (CAP) reform – and allowing energy crops to be grown
on set-aside land, were thought to encourage farmers in producing energy
crops to meet the demand for biofuel feedstocks. In addition, increasing
prices of fossil fuel and insecurity of supply, particularly in the past two
to three years, either due to natural disasters like Hurricane Katrina or
due to conflicts in the Middle East and elsewhere (such as the supply dis-
putes between Russia and Ukraine or Russia and Georgia), should have
provided further incentives, in theory, to invest in, produce and promote
biofuels in the Member States. Even with these biofuel-favourable poli-
cies, and geopolitical conditions, the share of biofuels on the transport fuel
market has been negligible in most of the Member States.
The next milestone in the EU Directive 2003/30/EC on promotion of
the use of biofuels or other renewable fuels for transport is the target of
5.75 per cent – the proportion of biofuels and other renewable fuels that
should be placed on the transport fuel market by the end of the year 2010.
As in the first phase, most of the Member States have set their own annual
indicative targets for the second phase in the promotion of biofuels in the
transport fuel market. The indicative targets set by the 19 Member States,
if achieved, will increase the share of biofuels in the transport market to
5.45 per cent by 2010, slightly less than the 5.75 per cent target. However,
the 2006 progress report concludes, judging by the progress made by a
majority of the Member States during the first phase of biofuels promo-
tion in their transport fuel markets, that ‘the biofuels directive’s target [of
5.75 per cent] for 2010 is not likely to be achieved’ (European Commission,
2007b, p. 6).
Although a number of the EU Member States have put in place ‘biofuel
obligations’,3 the 2006 biofuels progress report calls for legislative meas-
ures to support the promotion of biofuels in transport in the EU, arguing
that such a legislative measure will ‘send a signal of the [European] Union’s
determination to reduce its dependence on oil use in transport’ (European
Commission, 2007b, p. 14). The report further argues that the legislative
framework in favour of biofuels will
give support to national, regional and local authorities working towards the
objective of reducing dependence on oil use in transport; give confidence to
companies, investors and scientists who are working on more efficient ways
to do this; and give pause to those who believe that European consumers will
always remain hostage to oil prices, whatever the price. (European Commission,
2007b, p. 7)
Positing that ‘a signal in the form of legally binding targets is stronger

than a purely voluntary commitment’, the biofuels progress report calls
for the EU to set minimum targets for the future share of biofuels in
transport, and suggests a minimum target of 10 per cent biofuels in the
transport market by 2020 (European Commission, 2007b, p. 8). Moreover,
the report presses for ‘efficiency in biofuel policy’ so as to build investor
confidence, reduce administrative burden and encourage production of
biofuels such that it helps meet the directive’s objectives (ibid.).
As mentioned above, biodiesel is the major biofuel adopted within
the EU Member States with a significantly higher share of total trans-
port market fuels than bioethanol. Of the two Member States exceeding
the 2 per cent target by the end of 2005, Germany’s biofuel market is
mainly biodiesel, whereas that of Sweden is mainly bioethanol (European
Commission, 2007b). However, in aggregate the total proportion of bio-
fuels in the EU transport fuel market is very skewed towards biodiesel.
The main reason for higher use of biodiesel is the availability of the
biodiesel feedstocks, such as oilseed rape, within the EU. Furthermore,
biodiesel requires very little adjustment in the engine as well as the supply
infrastructure, which is a great incentive for the investors. In fact Rudolph
Diesel’s first diesel engine was designed to run on vegetable oil (Demirbas,
2003). On the other hand, bioethanol is produced mainly from corn and
sugarcane at this early stage of technological progress, and has to be
imported from countries like Brazil. Although the next generation (the
so-called ‘second generation’) of biofuel production technologies are cur-
rently being developed and tested, such as the production of ethanol from
straw and wood chips, it might be a number of years before their produc-
tion is commercially viable and they are supplied to the market (Herrera,
2006; Schubert, 2006). Thus, in the meantime, EU Member States have to
either import bioethanol, as Sweden is doing from Brazil, or rely largely on
biodiesel produced within the EU to meet their biofuels targets.
Production of biodiesel within the EU Member States should be able to
supply enough fuel to meet the 2 per cent target set for 2005, however, it
will be difficult to meet the 5.75 per cent target set for 2010 from domestic
production alone. Apart from increasing the production of the biofuels
within the EU, meeting this target will be further complicated by the fact
that all of the EU member countries will not be able to grow all of the
feedstocks required to produce the biofuels to meet this target unless food-
crops are replaced by biofuel crops.4 For example, the United Kingdom
could produce enough feedstocks to meet the 2 per cent target by cultivat-
ing oilseed crops in less that 1 million hectares of arable farmland, which
could be available by using part of the set-asides, part of grasslands under
five years old (which are considered arable land) and without replacing
a significant proportion of the food-crops. However, to produce enough
feedstocks to meet the target of 5.75 per cent, the United Kingdom will
require around 2 million hectares of arable farmland, which in a country
with total arable land of just 5.8 million hectares5 (of which about 1.2
million hectares is grassland under five years old), is not feasible without
significantly altering the production of food-crops and the agricultural
landscape. The area of the set-asides is only about 0.5 million hectares, so
using part of this land will not contribute a significant proportion of the
total land required to produce the feedstocks. Thus, the only alternative
would be to import the biofuels and/or biofuel feedstocks from countries
outside the EU – mainly from countries like Indonesia and Malaysia (for
biodiesel) and Brazil (for bioethanol).
The EU Strategy for Biofuels (European Commission, 2006), and the
Biomass Action Plan (European Commission, 2005) call for ‘supporting
developing countries’ to develop internal and export markets for biofuels,
making sure minimum sustainability standards are met in their cultiva-
tion. The potential environmental and socioeconomic impacts from the
cultivation of biofuel feedstocks and the production (and consumption)
of biofuels are similar in the developing countries to in the EU. However,
increasing demand for biofuels from the EU and other developed coun-
tries is expected to not only provide opportunities for developing countries
to benefit from their production and export, but also to generate nega-
tive environmental and socioeconomic consequences, especially where
there are few laws in place to ensure sustainable production of biofuels.
Although the biofuels progress report refutes the claim that Europe’s
biodiesel consumption has caused tropical deforestation in palm-oil-
producing countries such as Malaysia and Indonesia, and suggests that the
global palm oil demand is mainly driven by the food market (European
Commission, 2007b), the increasing use and demand for such oils, not only
for foods but for biofuels in the West as well as in fast-growing economies
like China and India, is likely to be responsible for tropical deforestation
both in the Brazilian Amazon and in Indonesia (Casson, 2003; Monbiot,
2004, 2005; Mortished, 2006b). Hence, even if the EU restricts its import
of biofuels and biofuel feedstocks to those produced meeting its sustain-
ability criteria, other large-scale importers such as China may not apply
similar restrictions. Unless the countries producing biofuels have in place
strict environmental regulation with regard to biofuels production, there

will be an increasing possibility of negative environmental consequences
in those countries as the demand for biofuels increases globally in the
future.
The biofuels progress report, however, sees more benefits than costs
for both the EU and its trading partners from growth in the biofuels
industry. In addition to the much discussed benefits from the reduction
of greenhouse gas emissions, it not only sees short- and long-term secu-
rity of energy supply in the EU through the increasing use of biofuels,
it also posits that the demand for biofuel imports from the EU can help
improve trade relations and provide opportunities for developing coun-
tries to produce and export biofuels at competitive prices (European
Commission, 2007b, p. 10). In the likely absence of second generation
biofuels in the highly competitive fuel market for another decade or so,
Europe has to rely heavily on countries such as Malaysia, Indonesia (for
biodiesel) and Brazil (for bioethanol) to meet its growing biofuel needs,
at least in the foreseeable future. Taking a similar line to the EU Strategy
for Biofuels (European Commission, 2006) and the Biomass Action Plan
(European Commission, 2005), the biofuels progress report calls for non-
discriminatory access to the biofuels market for both domestic production
and imports, as long as they meet sustainability criteria in the production of
biofuels (European Commission, 2007b). Furthermore, the report argues
that the target of 10 per cent biofuels by 2010 could be met with limited
use of the second generation biofuels through (1) further development of
rapeseed cultivation in the EU and its neighbours to the east; (2) proper
incentives to biofuels producers, both in the EU and other countries, for
environmentally friendly production of biofuels and (3) ‘implementation
of the balanced approach to international trade in biofuels, so that both
exporting countries and domestic producers can invest with confidence in
the opportunities created by the growing European market’ (European
Commission 2007b, p. 13). However, the report argues that the likely
failure to meet the directive’s biofuels target for 2010 (as a whole in the
EU) cannot be described as ‘justified’, and hence calls for the revision of
the directive to make it more effective. It calls for the biofuels directive to
be revised to (1) reiterate the EU’s determination in reducing its depend-
ence in oil for transport; (2) set minimum targets for the share of biofuels
and (3) discourage unsustainable production of biofuels in favour of those
produced in an environmentally friendly fashion (European Commission,
2007b). Nonetheless, it recognizes that revision of the biofuel directive will
not work by itself, and that the changes will require ‘sustained effort on
the part of industry, agriculture and Member States as well as the EU’ to
make them work (European Commission, 2007b, p. 13).
Finally, the biofuels progress report is also in line with a broader policy
document from the EU – the Renewable Energy Road Map (European
Commission, 2007a). Like the biofuels progress report, the Renewable
Energy Road Map also calls for revision to existing voluntary targets for
use of biofuels in the EU, proposing instead a ‘mandatory (legally binding)
target of 20% for renewable energy’s share of energy consumption in the
EU by 2020’, along with the proposal for necessary legislative frameworks
to make sure such targets are met (p. 3). Like most other documents from
the EU relating to use of renewable energy, the Renewable Energy Road
Map is inherently optimistic: ‘The challenge [that is, meeting renewable
energy targets] is huge, but the proposed target can be achieved with
determined and concerted efforts at all levels of government assuming
the energy industry plays its full part in the undertaking’ (European
Commission, 2007a, p. 3).
This document also sees biofuels as the ‘only available large-scale sub-
stitute’ for transport fuels, recognizing barriers in trade as one of the key
factors that could hinder the EU Member States in meeting their biofuels
target, especially if they are dependent on biofuels imports from outside
the EU (European Commission, 2007a, p. 7). Furthermore, the document
is keen to stress the need to take an incentive-based approach, such as cer-
tification schemes with preferential trade agreements for biofuels, when it
comes to importing biofuels from developing countries, in order to ‘avoid
rain forest destruction’ and other forms of losses in biodiversity.
Biodiesel: benefits and costs

Some of the main arguments in favour of the use of biofuels in general,
and biodiesel in particular, is their potential to reduce greenhouse gas
emissions, their potential to supply a secure energy source, and their per-
ceived carbon neutrality and provision of a positive net energy balance.
Although the potential of biodiesel to reduce GHG emissions is well
established and much less contested, the latter two issues are subject to a
lot of debate. In order to assess these claims, it is essential to look into the
actual benefits and costs of biodiesel, from the early stages of production
to end use. In this section, we try to summarize the arguments surrounding
the benefits and costs of biodiesel based on a number of earlier studies. We
assess both the economic and environmental costs and benefits, and try
to outline whether biodiesel is or could be a beneficial alternative fuel at
present and in the future.
A number of life-cycle analyses have looked into the direct environ-
mental impacts of biodiesel, arising mainly from vehicle emissions. A
general conclusion from these studies is that biodiesel fuel provides a
‘net positive energy balance’6 and reduced CO2 emissions (Poitrat, 1999;
Mortimer et al., 2003; Puppan, 2002; Booth et al., 2005; Bozbas, 2005).
Individual studies differ in many respects however, making it difficult to
generalize all aspects of environmental impacts of biodiesel. Using life-
cycle assessment of RME (rapeseed methyl ester) and conventional diesel,
Franke and Reinhardt (1998) present a comparative overview of the
environmental impacts of both these diesel fuels. Of the six environmental
impacts considered, three are shown to favour RME and the remaining
three to favour fossil fuel diesel. Although RME could have a potentially
lower greenhouse effect (CO2 equivalents), low eco and human toxicity
(NOx) and lower resource demand (finite energy), it is shown to have a
significant ozone depletion potential due to nitrous oxide (N2O) emissions.
Furthermore, potential eutrophication and acidification from the use of
fertilizers, pesticides and herbicides in rapeseed production adds to the
potential negative environmental impacts of biodiesel.
Franke and Reinhardt (1998) is one of the few studies that can be termed
‘complete’ life-cycle assessment in that they cover almost all aspects of the
life cycle of the products in question for both RME and conventional
diesel. As such, the results from the study also seem very balanced with
regard to which fuel is more ecologically sound. Nevertheless, the authors
conclude – ‘under certain assumptions, RME has or can have an “overall”
ecological advantage against diesel oil’ (Franke and Reinhardt, 1998,
p. 1032).
The carbon balance and ecological footprint of ethanol as a fuel were
analysed by Dias de Oliveira et al. (2005) for the examples of Brazil and
the United States. Controversially they conclude that using ethanol as a
substitute for petroleum is not environmentally sustainable when emis-
sions from agricultural inputs (in the case of sugarcane) and conversion (in
the case of maize) are taken into account. In contrast, a study on ethanol
from maize and biodiesel from soybeans showed that ethanol yields 25 per
cent more energy than that invested in its production and biodiesel yields
93 per cent more (Hill et al., 2006). The situation is further complicated by
the use of nitrogen fertilizers in biofuel production because N2O release
from fertilizer application significantly contributes towards GHG emis-
sions (Crutzen et al., 2007), though if nitrogen demand is reduced by using
second generation biofuel feedstocks such as grasses and woody coppice,
this problem is alleviated.
One of the major environmental impacts from the burning of fossil fuel
diesel comes from the emission of sulphur. Biodiesel, on the contrary, pro-
duces virtually zero sulphur emissions, making it much cleaner compared
with fossil fuel diesel (Puppan, 2002; Bozbas, 2005). With biodiesel, there
is also a significant reduction in other emissions, such as carbon monoxide,
hydrocarbons and soot (Puppan, 2002; Mortimer et al., 2003; Nwafor,
2004; Schmidt, 2004). Although NOx emissions from vehicles using bio-
diesel are found to be generally higher than from those using conventional
diesel, studies have shown that this can be reduced by a slight modification
of the diesel engine (Schmidt, 2004; Booth et al., 2005). A major drawback
in using biodiesel compared with conventional diesel seems to be the higher
emission of N2O, both from production processes (that is, from feed-
stock production) and vehicle emissions, which has stratospheric ozone-
depleting potential (Franke and Reinhardt, 1998; Poitrat, 1999). Although
Poitrat (1999) contests this drawback, stating that the overall greenhouse
effect of biodiesel is four or five times less than that of conventional diesel,
Franke and Reinhardt (1998, p. 1037) are more cautious and state that this
drawback actually favours conventional diesel over RME given the ‘high
to very high ecological importance’ of this ozone depletion potential.
Many different studies have come up with different figures as to the
level of emissions and emission reductions for various gases and particu-
late matters when using biodiesel. While Ryan et al. (2006) state that for
biodiesel the CO2 equivalent emissions savings could range from 36 per
cent to 83 per cent compared with conventional diesel, Mortimer et al.
(2003) conclude that savings in total CO2 emissions of 72 per cent to 86 per
cent are possible by using biodiesel derived by conventional and modified
production respectively compared with ultra low sulphur diesel. Similarly,
net energy balance (measured as NER – Net Energy Ratio) for biodiesel
use has been reported to be from between 1.59 and 2.08 (Turley et al.,
2002) through to between 1.9 and 2.7 (Poitrat, 1999).
Feedstocks, comprising mainly oilseed rape and other competitor oils
like soybean, sunflower and palm oil, are the main input in the production
of biodiesel fuel. The production of these feedstocks could have two major
environmental impacts, namely an impact on land use and an impact on
biodiversity. The general consensus in terms of the potential impacts of
biodiesel feedstock production on the environment is that there would be
little or no negative impact if the feedstock were grown on existing agri-
cultural land (Anderson et al., 2004). However, Puppan (2002) warns of
dangers to the environment from large monoculture oilseed farms, as well
as land and water pollution from excess use of fertilizers and pesticides
in commercial oilseed plantations. In terms of the impact of biodiesel
feedstock production on biodiversity, Anderson et al. (2004) argue that
production of oilseed rape could have a positive impact on biodiversity,
especially for farmland bird species that feed on seeds and invertebrate
(that is insect) species within the crop. However, they warn against using
large areas of set-aside land for oilseed production, as that could have
negative impacts on the birds inhabiting such land (Anderson et al., 2004;
Anderson and Ferguson, 2006).
Franke and Reinhardt (1998) posit that the production of biodiesel

feedstocks could be quite an intensive land use, often with greater trans-
portation efforts compared with fossil fuels. Furthermore, commercial
production of these feedstocks could pollute water from leaching of ferti-
lizers, pesticides and herbicides – adversely impacting local and regional
biodiversity. Turley et al. (2002, 2003) state that biodiesel feedstocks
grown in set-asides or on other uncultivated lands could increase traffic
impacts in rural areas, whereas those grown on already cultivated lands by
replacing current crops will have little or no such impact. In addition to
increased traffic impacts, the use or possible use of the set-asides to grow
biodiesel feedstocks could have negative environmental impacts according
to a number of studies (for example, Turley et al., 2002, 2003; Anderson et
al., 2004; Anderson and Ferguson, 2006).
In terms of the impacts of biodiesel on human health, emissions from
biodiesel are reported to pose lower direct or indirect risks compared with
fossil fuel diesel (Poitrat, 1999). However, reports are conflicting on the
impacts of biodiesel feedstock production on human health. Major health
concerns from the production of biodiesel feedstocks, such as oilseed rape,
are the allergies related to these crops – complaints and concerns about
which seem to come mainly from the British public rather than from those
in other oilseed rape growing countries like Canada, France and Germany
(Hemmer, 1998; Turley et al., 2002). As most of these complaints (from
the British public) seem to be from anecdotal rather than scientifically
proven ill-health impacts, Hemmer (1998, pp. 1327–8) questions if there
is ‘prejudice because the expansion of this crop is subsidised by the EU’
or if ‘people simply dislike [oilseed rape’s] intense smell and flashy yellow
flowers’, concluding that there is ‘little evidence to incriminate a versatile
crop of economic importance as a cause of ill health’.
Most studies comparing emissions from biofuel-run vehicles with those
of fossil-fuel-run vehicles conclude that biofuel could be significantly less
harmful, implying significant environmental benefits. However, the carbon
neutrality and provision of net energy balance by biodiesel is a contested
issue. Although combustion of biodiesel itself is carbon neutral in that it
releases no more carbon than that sequestered during growth, inputs in
production, transportation and marketing of biodiesel usually increase net
carbon emissions due to the use of fossil fuels in these operations. Thus,
strictly speaking, biodiesel cannot be 100 per cent carbon neutral unless
biofuels themselves are used for the energy component in all parts of the
manufacturing process. It is, however, possible to reduce the net carbon
emissions from biodiesel further by reducing the use of inputs, such as
nitrogen fertilizers, in the production of the feedstocks, the production
of which emits the highest amount of CO2 among the agricultural inputs
(Powlson et al., 2005). Furthermore, reduction in the amount of transpor-

tation necessary to move biodiesel feedstocks to oil refineries, by building
refineries as close to the feedstock production sites as possible, could help
reduce CO2 emissions further (Turley et al., 2002, 2003).
An equally contested issue is that of the energy balance of biodiesel.
We have reported studies earlier in this section that have shown the net
energy balance of biodiesel varying from around 1.6 to 2.7 (Poitrat, 1999;
Turley et al., 2002). However, in a controversial paper, Pimentel and
Patzek (2005) report that the net energy balance of biodiesel is on the
negative side, meaning that production of biodiesel requires more energy
inputs than the energy output from the biodiesel produced. Presenting
their analysis on the energy requirement for the production of biodiesel
from soybean and sunflower in the United States, they show that biodiesel
production using soybean required 27 per cent more fossil energy, and that
using sunflower required 118 per cent more fossil energy than the energy
output from the biodiesel produced. Pimentel and Patzek (2005) argued
that a number of earlier studies showing net positive energy balance from
biofuels were incomplete as they had omitted some of the energy inputs
in the production system. However, Wesseler (2007), commenting on
Pimentel and Patzek (2005), argues that the latter’s study was flawed as it
ignored opportunity costs, especially those associated with the alternative
use of the land used to produce feedstocks or alternative use of the feed-
stocks themselves. Revising Pimentel and Patzek’s analysis by including
the opportunity costs associated, Wesseler (2007) shows a positive energy
balance for the biodiesel produced from both soybean and sunflower.
Moving away from the environmental costs and benefits of biodiesel
and focusing solely on the economics of biodiesel production, we find
that in the majority of cases, the cost of a litre of biodiesel is significantly
higher than the cost of a litre of fossil fuel diesel (Barnwal and Sharma,
2005; Demirbas and Balat, 2006). Demirbas and Balat (2006) report that,
at the time of writing, pure biodiesel is 120–175 per cent more expensive
than fossil fuel diesel, though of course fossil fuel prices can change rapidly
in response to demand and geopolitical events. Most of the biodiesel cur-
rently produced comes from soybean, sunflower, palm or rapeseed oil. All
of these oils are widely used in the food production market, resulting in a
higher price. Thus, the use of these oils in biodiesel production increases
the costs of production, thereby increasing the market price of biodiesel.
Even in countries like India, the cost of biodiesel is significantly higher
than the cost of diesel, mainly because of the higher cost of vegetable
oils used in biodiesel production (Barnwal and Sharma, 2005). Hence,
at present, biodiesel cannot compete with diesel in the market in terms
of price without government subsidies. Studies suggest that the cost of
biodiesel could be reduced by using low cost oils, such as used frying oils
from restaurants and non-edible oils that are usually cheaper than edible
oils (Azam et al., 2005; Barnwal and Sharma, 2005; Demirbas and Balat,
2006).
Although the higher environmental benefits from the use of biodiesel
could justify a higher cost (price), for the majority of the customers it
is the price they pay that matters. Thus, unless the price of biodiesel is
competitive with that of fossil fuel diesel, it is unlikely that the average
consumer will look into using biodiesel instead of diesel. Environmentally
aware ‘green consumers’ may be willing to pay a higher price for biodiesel,
but their use will be negligible in comparison with the use of diesel by the
majority of other consumers. Provision of government subsidies could
lower the price of biodiesel and increase its adoption in the short run,
but subsidies are not a long-term solution. Thus, the only way to increase
adoption of biodiesel by an average consumer in the long run would be to
make it competitive with fossil fuel diesel in terms of price.
What potential for biodiesel to provide double benefits?

During the past few years biofuels have grown out of being a niche market
fuel into being a mainstream market fuel – more so in countries such as
Brazil, which developed a biofuels programme in the 1970s because of
increased oil prices combined with a low sugar price (Oliveira, 2002).
However, the adoption of biofuels has generally not been very encourag-
ing, especially in developed countries. Of the EU states that were required
to supply their transport fuel market with 2 per cent of biofuels by the
end of 2005, only two countries (Germany and Sweden) met the target.
Biodiesel has been the more popular biofuel in the EU, taking over 80 per
cent of the biofuels market. However, even with a number of incentives,
such as fuel tax subsidies, agricultural subsidies and so on, the adoption
of biodiesel has been poor in EU countries to say the least. In a context
where biodiesel has been touted as a reliable source to supply growing fuel
demands, at the same time as being environmentally friendly, it is essential
to ask if this optimism about biodiesel providing double benefits – that is,
being a cheaper and more secure source of fuel supply, at the same time as
providing environmental benefits – is justifiable. In this section we explore
this issue from various angles – the demand and supply of biodiesel, the
policies driving its production and use or potential use, and whether this
optimism about environmental benefits is actually well founded.
It is fair to say that the demand for biodiesel is growing steadily over
the years. This has come about due to a number of reasons: (1) due to
a great insecurity surrounding the production and supply of fossil fuels
from all major oil-producing regions; (2) due to growing environmental
awareness of governments as fuel consumers in terms of GHG emissions

and global warming and (3) due to policies driving the use of renewable
fuels in general and biodiesel in particular in the EU and in other regions.
Security of supply has been a major cause of concern for fossil fuel diesel
for decades now. Most of the largest oil-producing regions can, in one way
or another, be classified as conflict zones. The Middle East is an obvious
example, but a recent dispute between Russia and Ukraine and Russia
and Georgia unexpectedly threatened supply of oil to Western Europe. In
this context, alternative fuel sources, especially those that can ensure the
security of supply in the long run, have become very important. A number
of renewable energy sources, such as biomass, wind and solar power and
biofuels are seen as potential alternatives and secure sources of energy.
For the transport sector biodiesel has the potential to become a secure
source of renewable energy in the EU as it can be ‘homegrown’. One of
the most common feedstocks for biodiesel production, rapeseed, can be
easily grown within the EU Member States, which increases the security of
supply. However, the EU has a limited area of arable land that can be used
to produce rapeseed or other feedstocks for biodiesel, as the production
of biodiesel feedstocks will be competing with food production, which for
obvious reasons will always have a higher priority. Thus, although most
of the biodiesel used in the EU at present is produced within the EU, it
is clear that the higher proportions of biodiesel required to meet future
targets cannot be supplied internally. At some point, the EU has to rely on
imported biodiesel just as it relies on imported fossil fuel diesel now. Thus,
guaranteeing the security of supply of biodiesel could become as much of
an issue as the security of supply for fossil fuel diesel at present.
The tropical countries, such as Brazil, Malaysia, Indonesia and a
number of African countries could be the potential source of feedstocks
for the future EU market. A major concern, however, seems to be the
potential deforestation and land use changes due to increased demand
for arable lands to grow biofuel feedstocks like palm oil, oilseed rape,
soybean and sugarcane in these tropical countries. The public perception
about biodiesel being the ‘green’ alternative fuel is influenced more by
reports and opinion pieces in the newspapers than by articles in scientific
journals. And, judging by the number of news reports and opinion pieces
on how biodiesel could bring an environmental disaster rather than being
a greener alternative to fossil fuels (for example, Monbiot, 2004, 2005;
Mortished, 2006a, 2006b), it is likely that it will take a lot more effort to
make people switch from fossil fuel diesel to biodiesel, even those with a
concern for the environment. The European Commission (2005) seems to
share this concern as it fears that public support for biofuels will dwindle
if worries about possible deforestation and destruction of natural habitats
are not addressed properly. Moreover, if biofuels lose their reputation as

the ‘cleaner, greener fuel’ compared with fossil fuels, meeting and sustain-
ing any target level will be much more difficult.
There are also moral and ethical dimensions to the issue when it comes
to sourcing biodiesel feedstocks from countries in the tropics that have
low food security. The livelihood impacts of switching from cereal pro-
duction for domestic use to biodiesel feedstocks production for export,
especially in the countries that are not self-sufficient in food grains, could
be devastating. Economists call for the most efficient use of the resource
or mode of production, so according to them, it makes perfect sense to
produce biodiesel feedstocks as cash crops in these developing countries
and import cheaper grain from those who produce in surplus. However,
this is disputed by those concerned with sustainability, and also on moral
and ethical grounds. Local livelihood effects of changes in crop yields and
prices can be devasting, the examples of farmers committing suicide in
the Indian state of Kerala due to cash crop failures have become all too
common in recent years to justify the switch from cereal to cash crops in
such countries.7
The review of cost and benefits of biofuel production above shows that
just relying on financial cost–benefit analysis is not sufficient to assess the
true costs and benefits from the biodiesel. Although the cost of biodiesel
is significantly higher than that of fossil fuel diesel at present, it is offset
by significantly greater environmental benefits from biodiesel because of
lower GHG emissions and carbon neutrality. However, from both an eco-
nomic and policy perspective, the costs of switching to a more expensive
fuel source are incurred in the present, but the environmental benefits
of the reduced GHG from biodiesel may not accrue until 40 or 50 years
from now. This makes analyses of costs and benefits difficult and complex
because of different time horizons involved. It is also not surprising that in
the early days of adoption, biodiesel, like every new technology, is likely
to incur high short-run costs and low adoption rates, usually in favour of
long-term benefits. Furthermore, despite growing environmental aware-
ness among consumers, lack of competitive prices of biodiesel compared
with fossil fuel diesel is slowing the adoption rate. It is essential to have a
higher adoption rate of biodiesel if it is to make a significant impact on
GHG emissions reduction.
In light of the growing production and use of biodiesel in the EU and
biofuels in general in the EU, as well as elsewhere in the world, it is not an
over-exaggeration to say that biofuels have a real prospect for providing
double benefits, especially when issues such as sustainable production are
resolved. There are both positives and negatives from the use of biofuels
on a small scale and at a large scale. On a small scale, production costs of
biofuels are usually higher, thereby making them more expensive com-
pared with fossil fuels, which enjoy an economy of scale. However, the
short- and long-term benefits of biofuel use even at a small scale could be
significant with regard to savings in carbon emissions and those of other
greenhouse gases. Moreover, competition for land use to grow feedstocks
for biofuels on a small scale is minimal, which means it is less likely that
the problems related to land use and land degradation will be significant at
this level of biofuels production and use.
In large-scale production and use of biofuels the savings on carbon
emissions will certainly be significantly higher, especially when production
does not involve intensive use of fossil fuels. Moreover, greater production
and use of the biofuels is likely to reduce their production costs, and hence
their market price, encouraging even more use of such fuels as substitutes
for fossil fuel. However, there is a growing concern that this large-scale
production of biofuels would encourage extensive land conversion to
produce feedstocks, such as soybean, corn, rapeseed and sugarcane, espe-
cially in the tropics. If this is the case then benefits of reduced greenhouse
emission from such fuels are easily cancelled out by tropical deforestation
and biodiversity loss resulting from land conversion to provide feedstocks
for biofuel production. In this case, biofuels, instead of being an environ-
mentally friendly alternative to the fossil fuels, could become an environ-
mentally harmful source of energy. This means they might not be able to
provide double benefits as they are currently promoted as being able to do,
at present and in the future.
Conclusion
Growing demands for fuel, uncertainty and conflicts surrounding fossil
fuels, and increasing concern for the environment from the use of fossil
fuels have inevitably led to the search for alternatives to fossil fuels, espe-
cially renewable sources of fuels. Biofuels have been considered one of the
best alternatives to fossil fuels for the transport sector. The use of biofuels
in transport has been growing worldwide with bioethanol and biodiesel
leading the way. Although the price of biofuels is generally higher than
fossil fuels at this stage of biofuel production and use, incentives provided
by governments, such as subsidies, have helped promote the growing
use of biofuels. Favourable policies are in place, especially in the EU, to
promote use of biofuels as transport fuels. Despite the Council directive
on the use of biofuels in transport, which sets the target for the Member
States to ensure a minimum supply of biofuels in their transport fuel
market up to 2010, and favourable policies to go with the directive, the
target set by the directive has not been met and according to the Council’s
own progress report, the target set for 2010 is not likely to be met either.
This shows that legally non-binding targets, such as the biofuels targets,
will be difficult to meet despite favourable policies and economic incen-
tives – the reason why the Council’s biofuels progress report calls for a
legislative framework to make such targets legally binding (European
Commission, 2007b). Furthermore, the EU is proposing legislation
to force vehicle manufacturers to make technological improvement in
vehicles to reduce emissions.8
There is clear evidence of environmental benefits from using biofuels
generally, and biodiesel particularly, when reductions in greenhouse gas
emissions by vehicles running on biodiesel are considered. However, the
benefits are not so clear-cut when the environmental impacts of biofuels
production are accounted for, especially production of biofuel feedstocks
such as palm, soybean, rapeseed, sugarcane and corn – mainly in the
tropics. In its policy documents, such as the Biomass Action Plan, and
the Strategy for Biofuels, the EU sets out policies for promoting biofuels
production in developing countries, setting guidelines for ‘sustainable
production’ and supply of such fuels, and appropriate incentives, such
as guaranteed investments, for such ‘good practices’. However, it is not
guaranteed that producer countries will follow the EU guidelines, when
they are faced with growing demands for fuels from the rapidly indus-
trializing countries such as China and India. Indeed, lack of stringent
environmental policies and/or lack of enforcement of environmental
policies in place has been blamed for the growing loss of tropical rain-
forest in Malaysia and Indonesia to make way for palm oil plantations.
Recent years have seen a massive growth in such plantations mainly
to meet the demands for oil, so far mainly for the food industry, from
China. However, countries like China will have to rely more and more on
alternative fuel sources in the future to meet their insatiable demand for
energy, which would mean more pressure on the production of biofuels
among other sources of energy. This will no doubt increase pressure on
land use, especially in the tropics where most of the biofuels feedstocks
come from.
Certification schemes, which will guarantee that the biofuels produced
in any country, especially in the developing countries, are produced in an
environmentally sustainable way, are seen as a possible solution to concern
about potential tropical deforestation and habitat destruction during the
production of such fuels (European Commission, 2005). Restrictions on
the imports of biofuels and biofuel feedstocks to the EU based on such
certification schemes could encourage producers to produce biofuels in an
environmentally sustainable way; however, implementing such a scheme
is not without problems. International trade agreements and WTO rules
require that such certification schemes be non-discriminatory between
domestically produced biofuels and imports (European Commission,

2005), however, it is obvious that the developing countries, where most
of the biofuels are or will be produced, do not always have the same
environmental sustainability standards in production as the EU. Making
sure that biofuel producers in developing countries comply with the same
environmental sustainability standards as EU producers will increase
the costs of production of biofuels in those developing countries. These
increased costs could lead to a bifurcation in the biofuels market – one
market for the export of standard-compliant biofuels to the EU, and
another for standard-non-compliant biofuels to the countries, mostly fast-
growing and needy developing countries, with less stringent environmental
regulations.
Countries like China, with an ever-growing fuel demand, are already
investing heavily in palm oil industries in Indonesia, without due consid-
eration to the potential tropical deforestation caused by the expansion
of such industries (Perlez, 2005). And the companies supplying palm oil,
either for food products or as processed biodiesel to China, which has less
stringent environmental regulations, are unlikely to practice environmen-
tally sustainable production schemes at a greater cost when they are not
required to do so. In fact this could lead to a shortage in biofuels supply to
regions like the EU, with more stringent environmental regulations, from
countries like Indonesia in the future, when they could have easy access to
large markets like that of China. Thus, the challenge for the EU in meeting
biofuels targets such as 5.75 per cent of all transport fuels by 2010 is
twofold. First, meeting as much demand for biofuels as possible from the
domestic supply without negatively affecting the agriculture and food pro-
duction sector, also making sure biofuels are produced in an environmen-
tally sustainable manner. Second, putting effective certification schemes
in place that guarantee the environmentally sustainable production of the
biofuels and making sure they are followed when importing biofuels from
outside the EU, mainly from the developing countries.
Many of the arguments regarding biofuels providing double benefits
are not well founded if one considers the total environmental impacts of
biofuels, from production to their combustion. It is too early to conclude
that the biofuels could be the panacea for the issues surrounding green-
house gas emissions and global warming, or even pollution from vehicles.
However, sustainable production of biofuels could help supply a cleaner
fuel to the transport market to a certain extent and help in greenhouse
gas emissions reduction in the process, thereby providing double benefits.
The prospect is there for biofuels to make a difference, but it is up to all
concerned – the state, the market and the consumers – on how to make it
work.
Notes
1. http://news.bbc.co.uk/1/hi/world/europe/6334327.stm, accessed 31 August 2009.
2. 1. (a) Member States should ensure that a minimum proportion of biofuels and
other renewable fuels is placed on their markets, and, to that effect, shall set
national indicative targets.
(b) (i) A reference value for these targets shall be 2%, calculated on the basis of
energy content, of all petrol and diesel for transport purposes placed on
their markets by 31 December 2005.
(ii) A reference value for these targets shall be 5.75%, calculated on the basis
of energy content, of all petrol and diesel for transport purposes placed
on their markets by 31 December 2010.
(Article 3, Directive 2003/30/EC, OJ L 123, pp. 44–5)
3. A measure that requires the oil suppliers to put a certain percentage of biofuels in the fuel
they supply to the market.
4. It is estimated that 17 million hectares of arable land will be needed to produce biofuels
to meet this target completely from the domestic production. Given the total arable land
in the EU of 97 million hectares, meeting the biofuels target set for 2010 entirely from the
EU domestic production is thought to be ‘technologically feasible in principle’. However,
due to WTO regulations and other trade agreements, the EU cannot block the import of
(potentially cheaper) biofuels and biofuel feedstocks (mainly from developing countries).
‘Therefore, the scenario of 100% domestic production is a theoretical one and would not
be possible in practice’ (European Commission, 2005).
5 http://statistics.defra.gov.uk/esg/quick/agri.asp, accessed 31 August 2009.
6. Net energy balance is the difference between energy in the biofuels and the energy
equivalence of fossil fuel inputs used in producing, harvesting and processing the fuel.
Net energy balance is usually calculated as a ratio – called Net Energy Ratio (NER) – of
biofuel energy to the fossil fuel energy used in production of the biofuel. NER greater
than 1 shows a ‘positive’ net energy balance, which indicates that the biofuel generates
more energy than that used to produce it.
7. http://news.bbc.co.uk/1/hi/world/south_asia/4988018.stm, accessed 31 August 2009.
8. http://news.bbc.co.uk/1/hi/world/europe/6334327.stm, accessed 31 August 2009.
References
Anderson, G.Q.A. and M.J. Ferguson (2006), ‘Energy from biomass in the UK: sources,
processes and biodiversity implications’, Ibis, 148(1), 180–83.
Anderson, G.Q.A., L.R. Haskins and S.H. Nelson (2004), ‘The effects of bioenergy crops
on farmland birds in the UK: a review of current knowledge and future predictions’,
in K. Parris and T. Poincet (eds), Biomass and Agriculture: Sustainability, Markets and
Policies, Paris: OECD.
Azam, M.M., A. Waris and N.M. Nahar (2005), ‘Prospects and potential of fatty acid
methyl esters of some non-traditional seed oils for use as biodiesel in India’, Biomass and
Bioenergy, 29(4), 293–302.
Barnwal, B.K. and M.P. Sharma (2005), ‘Prospects of biodiesel production from vegetable
oils in India’, Renewable and Sustainable Energy Reviews, 9(4), 363–78.
Booth, E., J. Booth, P. Cook, B. Ferguson and K. Walker (2005), Economic Evaluation of
Biodiesel Production from Oilseed Rape Grown in North and East Scotland, report prepared
by SAC Consultancy Division.
Bozbas, K. (2005), ‘Biodiesel as an alternative motor fuel: production and policies in the
European Union’, Renewable and Sustainable Energy Reviews, 12(2), 542–62.
Casson, A. (2003), Oil Palm, Soybeans & Critical Habitat Loss: A Review Prepared for the
WWF Forest Conservation Initiative, Zurich, Switzerland: WWF Forest Conservation
Initiative.
Crutzen, P.J., A.R. Mosier, K.A. Smith and W. Winiwarter (2007), ‘N2O release from
agro-biofuel production negates global warming reduction by replacing fossil fuels’,

Atmospheric Chemistry and Physics, 7(4), 11191–205.
Demirbas, A. (2003), ‘Biodiesel fuels from vegetable oils via catalytic and non-catalytic
supercritical alcohol transesterifications and other methods: a survey’, Energy Conversion
and Management, 44(13), 2093–109.
Demirbas, M.F. and M. Balat (2006), ‘Recent advances on the production and utilization
trends of bio-fuels: a global perspective’, Energy Conversion and Management, 47(15/6),
2371–81.
Dias de Oliveira, M.E., B.E. Vaughan and E.J. Rykiel (2005), ‘Ethanol as fuel: energy,
carbon dioxide balances, and ecological footprint’, BioScience, 55(7), 593–602.
Doornbosch, R. and R. Steenblik (2007), Biofuels: Is the Cure Worse than the Disease? Paris:
Organisation for Economic Co-operation and Development.
European Commission (2005), Communication from the Commission – Biomass Action Plan
{SEC(2005) 1573}, Brussels: European Commission.
European Commission (2006), Communication from the Commission – An EU Strategy for
Biofuels {SEC(2006) 142}, Brussels: European Commission.
European Commission (2007a), Communication from the Commission to the Council and
the European Parliament – Renewable Energy Road Map, Renewable energies in the 21st
century: building a more sustainable future {COM(2006) 845 final} {Sec(2006) 1719}
{Sec(2006) 1720} {Sec(2007) 12}, Brussels: Commission of the European Communities.
European Commission (2007b), Communication from the Commission to the Council and the
European Parliament – Biofuels Progress Report: Report on the progress made in the use of
biofuels and other renewable fuels in the Member States of the European Union {COM(2006)
845 final} {SEC(2006) 1721} {SEC(2007) 12}, Brussels: Commission of the European
Communities.
Franke, B. and G. Reinhardt (1998), ‘Environmental impacts of biodiesel use’, paper read
at BioEnergy ’98: Expanding BioEnergy Partnerships, 4–8 October 1998, at Madison,
Wisconsin.
Hemmer, W. (1998), ‘The health effects of oilseed rape: myth or reality?’, British Medical
Journal, 316(7141), 1327–8.
Herrera, S. (2006), ‘Bonkers about biofuels’, Nature Biotechnology, 24(7), 755–60.
Hill, J., E. Nelson, D. Tilman, S. Polasky and D. Tiffany (2006), ‘Environmental, economic,
and energetic costs and benefits of biodiesel and ethanol biofuels’, Proceedings of the
National Academy of Sciences, 103(30), 11206–10.
Intergovernmental Panel on Climate Change (2007), ‘Climate Change 2007: The Physical
Science Basis – Summary for Policymakers’, Contribution of Working Group I to the
Fourth Assessment Report of the Intergovernmental Panel on Climate Change (Draft),
Geneva, Switzerland: IPCC Secretariat. Available online at http://www.ipcc.ch/, accessed
31 August 2009.
Kitzes, J., M. Wackernagel, J. Loh, A. Peller, S. Goldfinger, D. Cheng and K. Tea (2007),
‘Shrink and share: humanity’s present and future ecological footprint’, Philosophical
Transactions of the Royal Society B, 363(1491), 467–75.
Koh, L.P. (2007), ‘Potential habitat and biodiversity losses from intensified biodiesel feed-
stock production’, Conservation Biology, 21(5), 1373–5.
Monbiot, G. (2004), ‘The adoption of biofuels would be a humanitarian and environmental
disaster for the planet’, The Guardian, 22 November 2004.
Monbiot, G. (2005), ‘The most destructive crop on earth is no solution to the energy crisis’,
The Guardian, 6 December 2005.
Mortimer, N.D., P. Cormack, M.A. Elsayed and R.E. Horne (2003), Evaluation of the
Comparative Energy, Global Warming and Socio-economic Costs and Benefits of Biodiesel,
report for DEFRA/, Sheffield Hallam University.
Mortished, C. (2006a), ‘Food prices would soar in biofuels switch, says Unilever’, The Times,
7 August 2006.
Mortished, C. (2006b), ‘Clean and green, but is biofuel a winner?’, The Times, 7 August
2006.
Naylor, R., H. Steinfeld, W. Falcon, J. Galloway, V. Smil, E. Bradford, J. Alder and

H. Mooney (2005), ‘Losing the links between livestock and land’, Science, 310(5754),
1621–2.
Nwafor, O.M.I. (2004), ‘Emission characteristics of diesel engine operating on rapeseed
methyl ester’, Renewable Energy, 29(1), 119–29.
Oliveira, J.A.P. (2002), ‘The policymaking process for creating competitive assets for the use
of biomass energy: the Brazilian alcohol programme’, Renewable and Sustainable Energy
Reviews, 6(1/2), 129–40.
Perlez, J. (2005), ‘The end of Borneo’s tropical forests?’, International Herald Tribune, 28
April 2006.
Pimentel, D. and T.W. Patzek (2005), ‘Ethanol production using corn, switchgrass, and
wood; biodiesel production using soybean and sunflower’, Natural Resources Research,
14(1), 65–76.
Poitrat, E. (1999), ‘The potential of liquid biofuels in France’, Renewable Energy, 16(1–4),
1084–9.
Powlson, D.S., A.B. Riche and I. Shield (2005), ‘Biofuels and other approaches for decreas-
ing fossil fuel emissions from agriculture’, Annals of Applied Biology, 146(2), 193–201.
Puppan, D. (2002), ‘Environmental evaluation of biofuels’, Periodica Polytechnica Ser. Soc.
Man. Sci., 10(1), 95–116.
Righelato, R. and D.V. Spracklen (2007) ‘Carbon mitigation by biofuels or by saving and
restoring forests?’, Science, 317(5840), 902.
Ryan, L., F. Convery and S. Ferreira (2006), ‘Stimulating the use of biofuels in the European
Union: implications for climate change policy’, Energy Policy, 34(17), 3184–4.
Schmidt, L. (2004), ‘Biodiesel vehicle fuel: GHG reductions, air emissions, supply and eco-
nomic overview’, Discussion Paper No. C3-015.
Schubert, C. (2006), ‘Can biofuels finally take center stage?’, Nature Biotechnology, 24(7),
777–84.
The Council of the European Union (1998), ‘Council Resolution of 8 June 1998 on renewable
sources of energy’, Official Journal of the European Communities, C198, 1–3.
The Council of the European Union (2003), ‘Council Directive 2003/96/EC of 27 October
2003 restructuring the Community framework for the taxation of energy products and
electricity’ (text with EEA relevance), OJ L 283, 51–70.
The European Parliament and the Council of the European Union (2001), ‘Directive
2001/77/EC of the European Parliament and of the Council of 27 September 2001 on the
promotion of electricity produced from renewable energy sources in the internal electricity
market’, OJ L 283, 33–40.
The European Parliament and the Council of the European Union (2003), ‘Directive 2003/30/
EC of the European Parliament and of the Council of 8 May 2003 on the promotion of the
use of biofuels or other renewable fuels for transport’, OJ L 123, 42–6.
Turley, D., G. Ceddia, M. Bullard and D. Martin (2003), Liquid Biofuels – Industry Support,
Cost of Carbon Savings and Agricultural Implications, report prepared by CSL, ADAS,
Ecofys for DEFRA, Organic Farming and Industrial Crops Division.
Turley, D.B., N.D. Boatman, G. Ceddia, D. Barker and G. Watola (2002), Liquid Biofuels
– Prospects and Potential Impacts on UK Agriculture, the Farmed Environment, Landscape
and Rural Economy, report prepared by Central Science Laboratory for DEFRA,
Organics, Forestry and Industrial Crops Division.
Wesseler, J. (2007), ‘Opportunities (costs) matter: a comment on Pimentel and Patzek
“Ethanol production using corn, switchgrass, and wood; biodiesel production using
soybean and sunflower”’, Energy Policy, 35(2), 1414–6.
Index
Aarhus Convention 214 Agenda 21 (Rio Declaration on

abatement costs 392 Environment and Development)
abatement effect 388, 396 218, 229
Abbot, J. 37, 64 Aggarwal, R.M. 140, 144, 148–9
abdomen 222 agricultural cover, direct use value of
Abel, N. 299 345–8
ability to pay 151 agricultural land coverage, percentage
Aborigines’ policy stance on burning of 75
176–7, 178, 180, 181, 182 agricultural products, prices of 252–3
discourses adopted in support of see also food prices
182–3, 185–8, 192 agricultural subsidies 7, 97, 105, 106,
absolute forest cover 13 409, 413, 422, 425
acanthurids 248 agri-environmental programmes 148
accountability 178 Agrostis capillaris 94
acid rain 199, 215, 362 Agrostis castellana 85
Adamowicz, W.L. 340 Agrostis-Festuca 87, 94
Adams, W.M. 31, 33 Ahlén, I. 88
Addy, N.D. 93 air pollution see under names of
Adhikari, B. 139, 143, 309 individual air pollutants, e.g.
afforestation 371, 410 carbon dioxide (CO2) emissions;
Africa sulphur dioxide (SO2) emissions
biodiversity conservation priorities see also greenhouse gas emissions
in 14, 20 air toxics 395
biodiversity in managed landscapes Albania 221
in 84–5, 86, 87, 94 Albertine Rift, Central Africa 42
biofuel feedstocks sourced from 423 Alchian, A.A. 120, 135
economic growth and deforestation Alés, Fernandez R. 84
in 370, 388 Algeria 221
entitlement to products of the Alliance for Zero Extinction 16
commons in 161 Alonso, J.A. 83
fuel wood shortage in 2, 4, 171–2 Alpert, P. 31
Integrated Conservation and alpine forests 93–4
Development Projects (ICDPs) Altieri, M. 103
in 30–70 altruism 99
savanna valuation studies in 338–40 Amazonian rain forests 14, 337
see also under names of individual ammoniacal nitrogen 389, 390
African countries, e.g. Kenya; amphibians 17, 20
Tanzania Anand, P. 282
African whitespotted rabbitfish anarchy, cooperation under 200,
(Siganus sutor) 247 201–2, 219
agency and motivation 176 Anderson, A.B. 337
in Cape York seminar transcript Anderson, D. 31, 299, 336
187, 188, 189, 190, 191 Anderson, G.Q.A. 419, 420
431
Andersson, J. 255, 269, 270 fire management in Cape York in

Andes 14 4–5, 176–94
Andreoni, J. 366 functional diversity in 79
Andresen, S. 206 structural transition model (STM)
Ansuategi, A. 366 used in 91
Antarctic Minerals Treaty 215 Australian Labor Party 215
Antle, J.M. 366, 371 Austria 394, 397
aquatic systems 21 Ausubel, J.H. 213–14
Araucaria cunninghamii 93 automotive emissions 379, 383, 392,
Araújo, M.B. 22 409, 417–18, 419, 420, 426, 427
Arcese, P. 31 average cost 129
Aredo, D. 120 Axelrod, R. 125
Arellano–Bond Generalized Method Ayres, R.U. 399–400
of Moments (A–B GMM) 398 Azam, M.M. 422
argumentation theory 175
ark shells (Barbatia fusca) 248, 252, Bach, S.B. 275
267, 268, 270 backward induction 312, 327
Arkansas National Wildlife Refuge Baden, J.A. 143
102 Bailey, C. 269
Armitage, D.R. 287, 291, 298, 303 Baillie, J.E.M. 8, 17, 20, 21, 22
Arrow, K.J. 98, 135, 360, 367, 399 Bakker, K. 173
arsenic 368, 384 Baland, J. 121, 128, 150–51, 154, 155,
arthropods 17 157, 161, 316
Arusha Resolution on Integrated Balat, M. 421, 422
Coastal Zone Management in bald eagle 102
Eastern Africa (including Island Baldock, D. 82
States) 271 Balmford, A. 8, 12, 13, 14, 22
Ashley, C. 43 balsam fir (Abies balsamea) 93
Asia 241, 370 Bamenda 58
see also central Asia; South Asia; Bamenda Highlands, Cameroon
South East Asia 56–64, 68, 69
ASL and Associates database 298 Bamenda Highlands Forest Project
aspen 92, 93 69, 70
asset specificity 136, 138, 141, 159 banana cultivation 52–3, 55
Assetto, V.J. 202 banded wattle-eye (Platysteira
assurance game 317–18, 322, 323 laticincta) 56
see also stag hunt game Bandyopadhyay, S. 360, 363, 368–9,
asymmetric equilibrium 316 385, 393
asymmetric information 136, 138, 141, Banks, Arthur 383
143 Bannerman’s turaco (Tauraco
Ataroff, V. 336 bannermani) 56, 63
Athens 227 Bantu tribes 291, 292
Attwell, C.A.M. 30 Barbier, E.B. 76, 97, 99, 102, 105, 107,
Australia 334, 338, 341
biodiversity conservation priorities Barcelona Convention 217, 219,
in 14 220–26, 229–30, 231, 232, 233
economic growth and pollution in Dumping Protocol 221–3, 229
394 Hazardous Wastes Protocol 223, 226
ecosystem productivity in managed LBS (Land-Based Sources) Protocol
landscapes in 93 222, 224, 229, 233
Index 433
Offshore Protocol 223, 225 biodiversity hotspots 9, 10, 16, 18, 410
Prevention and Emergency Protocol biofuels
222, 223–4 benefits and costs of 7, 409–10, 416,
SPA (Specially Protected Areas) and 417–22, 424, 426
Biodiversity Protocol 222, 225 certification schemes for 426–7
Bardhan, P. 119, 150, 152, 153, 155, EU renewable fuels policy and 7,
157, 158, 161, 311, 313, 314 409, 410–17, 425–7
bargaining costs 135, 138, 146, 148, potential to provide double benefits
149, 401 422–5, 427
Barnes, D.K.A. 245, 269 price of 421–2, 424–5
Barnett, H. 401 products classified as 412
Barnwal, B.K. 421, 422 biogeographic spatial units 13–14
Barreto, P. 331 biological diversity, definition of 76
Barrett, C.B. 31 biological oxygen demand (BOD) 368,
Barrett, S. 214, 383–4 384, 387, 389, 390
barter 259, 279 Biomass Action Plan 415, 416, 426
Barzel, Y. 121 bird species
Bateman, I. 101 biofuel feedstock production and
Baumann, O. 300 419
Bazaruto Archipelago marine conservation of 9, 10, 12, 14, 17, 18,
protected area 271 20, 22–3, 35, 56, 62, 63, 83, 87,
beavers 78 88, 92, 102, 104
Beckerman, W. 359, 360, 361 bird watching 98
beech (Fagus silvatica) 89–90 BirdLife International 23, 59, 60, 69
Begon, M. 296 Birkeland, C. 241
Behnke, R.H. 296, 297, 299 Birner, R. 136, 137, 140, 141, 142, 143,
Beira 271 144, 145, 146, 147
Belgium 394, 397 Bishop, R.C. 102, 107, 126, 131, 132
Bell, F.W. 102 black list 222, 223
Bembridge, J. 331 Black Sea 221
Benham, A. 140 blackcap (Sylvia atricapilla) 78
Benson, J.F. 100, 105 blacktip mojarra fish (Gerres oyena)
bequest value 99 247
Berkes, F. 126, 132, 143 Blench, R. 287
Bernauer, T. 206 Blomley, T. 31
Betula 88 Blomquist, W. 150
Bevir, M. 172 Blue Plan 217, 230, 231
Bhatia, A. 158 Blue Plan Regional Activity Centre
Bhatia, Z. 48, 52 (BP/RAC) 228, 230
Bhattarai, M. 365, 381 Bokdam, J. 84
Biermann, F. 203 Booth, E. 418, 419
Binmore, K. 310 Borchert, M.I. 87
biodiversity conservation in managed Borrini-Feyerabend, G. 30
landscapes see managed Bosnia and Herzegovina 221
landscapes, biodiversity Bosphorus 221
conservation in Botswana 97
biodiversity conservation priorities, Boulding, K.R. 176
global see global biodiversity bounded rationality 135
conservation priorities Bourn, D. 287
biodiversity conservation treaties 199 Bowes, M. 100
Bowker, J. 102 Buszko, J. 14

Bowman, D.M.J.S. 176–7 Buttel, F.H. 173
Boxer, B. 231
Boxer, C.R. 243 C-PLAN 41
boxfish (Ostraciidae) 259–60 Cabo Delgado province 242, 243, 244,
Boyce, J. 365, 381–2 250, 251, 263, 264, 270, 271, 272
Boyle, K.J. 102 Cactoblastis cactorum 79
Bozbas, K. 418 cadmium 368, 384
BP/RAC (Blue Plan Regional Activity CAFNEC (Cairns and Far North
Centre) 228, 230 Environment Centre) 180, 181,
Brachystegia trees 285 191
Bradshaw, R. 89, 93 Cairngorm mountains, Scotland 93
branch lines 311 Caldecott, J. 31, 40
Brandon, K. 31, 32 California 14
Brazil California Department of Consumer
biofuels sourced from 410, 414, 415, Affairs Bureau of Automotive
416, 418, 422, 423 Repairs 392
economic growth and pollution in Callaway, R.M. 87
375 Callmander, M.W. 20
economic valuation of forests in 337 CAMCOF (CAmeroon Mountains
Brazilian Atlantic forest 14 COnservation Foundation) 62
Briggs, J.C. 21 camels 85
Briske, D.D. 296, 297, 298 Cameroon 35
broadleaved species 88, 89 Integrated Conservation and
Brokaw, N. 92 Development Project in
Bromley, D.W. 124, 126, 127, 128, 131, Bamenda Highlands of 56–64,
132, 139, 158, 160 68, 69
Brooke, C. 293 Cameroon Ministry of Forestry and
Brooks, T.M. 11, 13, 14, 15, 20, 21, 23 Wildlife (MINFOF) 58, 60–61, 63
Brookshire, D. 101 Cameroon Ministry of the
Brown, G.M. 75, 99, 102 Environment and Forestry 69
Brown, J.R. 296, 297 Campbell, B.M. 301, 334, 335, 339–40,
Brown, M. 31 350
Brown Weiss, E. 212 Campbell, R.W. 87
Brummit, N. 12, 14, 17, 21, 22 CAMPFIRE (Communal Areas
Brundtland, G. 360 Management Programme for
Brundtland Report 282 Indigenous Resources) 87
Bruner, A. 30 Canada 14, 394, 397
Bryant, D. 9, 12, 13, 16 Canas, A. 402–3
Buchan, D. 30 Canney, S. 17, 21, 22
Buckley, P. 48, 52 Cape York (Australia), fire
Bulgaria 409 management in 4–5, 176–94
bureaucracy, quality of 377, 381 Cape York Peninsula Development
Burgess, J. 170 Association (CYPDA) 190
Burgess, N.D. 17, 35, 48, 49, 53, 56 Cape York Peninsula Sustainable Fire
burning 60, 64, 92, 95, 269, 275 Management Programme 177–8
management in Cape York, capitalism 202, 219–20, 282
Australia 4–5, 176–94 Caprinus 88
Business Environmental Risk capulana (cotton wrap) fishing 255, 267
Intelligence (BERI) 380–81 carbon balance 417, 418, 420–21, 424
Index 435
carbon dating 188 chemical oxygen demand (COD) 368,

carbon dioxide (CO2) emissions 384, 389
biofuels and 417–18, 419, 420–21 Cheung, S.N. 121, 135, 148
economic growth and 363, 372–5, chicken game 201, 317, 318–19
380–81, 387, 389, 391, 395, children, fishing by 249, 268, 269
396–8, 399, 408 Chile 14
see also greenhouse gas emissions China 14, 375, 409, 410, 415, 426, 427
Carbon Dioxide Information Analysis Chong, D. 170
Center (CDIAC) 373 Cincotta, R.P. 13, 22
carbon emissions 369, 370 Ciriacy-Wantrup, S.V. 107, 126, 128,
see also carbon dioxide (CO2) 131, 132
emissions; carbon monoxide Clapp, J. 202–3
(CO) emissions; greenhouse gas Clarke, G.P. 48
emissions Clawson, M. 100
carbon monoxide (CO) emissions 363, Crayfish (Panulirus spp.) 249
372, 376, 387, 391, 393, 395, 418 Cleaner Production Regional Activity
carbon payments 43 Centre (CP/RAC) 228
carbon sequestration 42, 100, 336, 340 Clements, F.E. 296
Cardillo, M. 16 climate change 13, 21, 176, 202,
CARE development agency 48, 49, 50, 359–60, 408
55, 67 see also global warming
care-intensity 137–8, 142, 143 climax community 296
Caribbean 21 Clout, M.N. 92
Carney, D. 32 clover 86–7
Carrascal, L.M. 83, 92 Club of Rome 175
carrying capacity 296, 360, 363, 366–7 Coase, R. 96, 119, 120, 122, 135, 213,
Carson, R.T. 101, 364, 394–5 401
Carter, N.T. 215, 216, 282 Coase theorem 120–21, 213
Carvalho, M. 264 coconut plantations 246, 252, 255, 256,
Casey, D. 92 262, 264–5
Casson, A. 415 Coe, M.J. 352
Catchment Forest Project 54 Cohen, S. 285
catfish (Plotosidae) 260 Cole, E. 88
cattle 86–7, 88, 89, 93, 178, 184, 246, Cole, M.A. 365, 372, 387–8
291, 293, 338, 339–40 Coleman, J.S. 162
Caucasus 14 collective action approach to
Cavlovic, T.A. 375 institutional analysis 119–20
Ceballos, G. 17 co-management 323
Census of Population and Housing 392 transaction costs of 142–6, 149–50
central Asia 14 command and control regulations
central Europe 93–4 107–8
centres of plant diversity 9, 10, 18 commercial vehicles 335
certification schemes 426–7 Common, M. 95, 97, 107, 341, 362,
CFC emissions 214, 372, 376–7 398
chamba (plot of land) 246, 268 Common Agricultural Policy (CAP)
Chambers, R. 31, 242 97, 106, 413
chance decision nodes 312 common goods 127
Chapman, D. 365, 385–6, 393 common pool resources (CPRs)
Chaudhuri, S. 393 economic valuation of different
Chavas, J.-P. 103 forms of land use 331–56
game theory as a tool for mapping consumption, relationship between

strategic interactions in 120, income and 386–7
121, 125, 160, 316–28 contingent valuation method (CVM)
see also game theory 100–102
role of social institutions in contracts 4, 119, 138, 320, 377, 381
management of 4, 119–62 Convention on Biodiversity 20, 23, 76,
see also institutions and 107
the management of conversion 190
environmental resources Conway, A. 86
common property regimes cooperative game theory 310
efficiency of 124, 125, 139 Cordery, I. 336
group heterogeneity and core conservation areas 32, 37, 54,
participation in 150–58, 161 56–7, 60
open access versus 124–7, 131 ‘core values’ of conservation 33
theoretical aspects of 130–34, 160 corruption in government 377, 383
transactions costs of 139, 142, 145 Cossins, N. 338
empirical studies of 146–50, cost–benefit analysis 337, 338, 339, 424
160–61 Costanza, R. 94
communication spaces 170 Cotterill, F.P.D. 30
community-based monitoring 63, 67, Coulson, A.C. 293
150 Council of the European Union
community-based natural resource Council Directive 2003/30/EC 409,
management 411–12, 416, 425
distinction between ICDP and 31–2 Council Directive 2003/96/EC 413
see also common property regimes Council Resolution 1998 on
community forestry 146–8, 149, 155, renewable sources of energy 411
156, 157–8 counterfactual analysis 209–11, 234–5
complete information condition 97–8 Cousins, S.H. 78
complete set of markets condition 97 Cowling, R.M. 23, 90
compliance in international climate Cox, S.J.B. 132
regimes 213–14, 230 CP/RAC (Cleaner Production
composition effect 388, 396 Regional Activity Centre) 228
concession purchase models 68 Craig, S. 101
Congleton, R.D. 361, 362, 376–7 cranes 83
Congo 14 crisis ecoregions 9, 10, 18
conifers 88 Critical Ecosystem Partnership Fund
Connolly, J. 86 16
consensus, creation of 170–71 Croatia 221, 228
conservation goals Crocker, T.D. 148
definition of 35 crop diversity, measuring value of 103
development activities consistent crop production function 340, 344
with 39 rainfall as input into 351–5
conservation targets Cropper, M. 364, 365, 370–71, 388
definition of 35, 37 Cross-national Time-series Data
direct payments to communities in Archive 382–3
return for 68 Crowley, G.M. 179
formulation of 38–41, 50–53, 60, Crozier, R.H. 77
62–3 Crutzen, P.J. 410, 418
consultants 38, 148 Cumbria, UK 89
consumer surplus 98, 100 Cumming, D.H.M. 86, 87
Index 437
Cushitic tribes 291, 292 demand curves 100

Cyprus 221 demand for labour 375–6
dematerialization 402–3
Dactylis glomerata 85 Demirbas, A. 414
Dahalani, Y. 255 Demirbas, M.F. 421, 422
Dahlman, C.J. 135–6 democracy 378, 380, 383
Daily, G.C. 75, 76, 78, 102 Demsetz, H. 121, 122, 123, 124, 131,
Dalton, R. 16 135
Dambach, C.A. 92 Denmark 93, 394, 397
Danielsen, F. 43, 67 desertification 287, 298
Danish International Development Deudney, D. 215
Agency (DANIDA) 47, 48, 51, development assistance funds 32, 44,
52, 69 67
Dansk Ornitologisk Forening (DOF) Di Falco, S. 103
48, 49, 69 Diamond, P.A. 101
Dar es Salaam 48, 272 Dias de Oliveira, M.E. 410, 418
Darcy, M. 168 Díaz, M. 82
Darwall, W. 21 Dinda, S. 403
Dasgupta, P.S. 128, 158, 161 Dinerstein, E. 9, 12, 13, 16, 21, 22
Dauvergne, P. 202 direct land purchase 68
Davies, J. 146–8 direct material input (DMI) 402–3
Dawes, R.M. 319, 320 direct payments to communities 68–9
Dayaratne, P. 242 direct use value 99–100, 335
Dayton-Johnson, J. 150, 152, 155–6, measurement of 101–2, 336–40,
158 342–3, 345–50, 355
De Bruyn, S.M. 366, 378–9, 395 discourse, definition of 168
De Miguel, J.M. 84 discourse analysis 4, 168–94
Deacon, R. 382–3 arguments for a discursive approach
Debussche, M. 78 to policy analysis 168–72
deciduous tree species 88 and fire management in Cape York,
decision costs 135, 141–4 passim, 149 Australia 4–5, 176–94
decision support software 41 applying Dryzek’s analytic
decomposition analysis 363, 378, framework 185–92
388–9, 396–9 applying Hajer’s analytic
deer 92 framework 181–5
see also giant Irish deer; red deer conclusions drawn from 192–4
(Cervus elaphus); roe deer theoretical perspectives on 172–6
(Capreolus capreolus) Dryzek 174–6
deforestation 49, 64, 97, 296, 333, 341, Hajer 172–4
410 discourse coalitions 173, 181–5 passim,
biofuel crop production and 7, 415, 186, 189, 190, 192
423–4, 425, 426, 427 discursive affinities 173, 184, 190, 191,
economic growth and 368–9, 192, 193
370–71, 380, 381, 388 discursive closure 173
open access regimes and 129 discursive contamination 173
soil erosion caused by 301, 322, 344 dissolved oxygen 368, 369, 382, 387,
DeGraaf, R.M. 92 391
dehesas 82–3, 85 distribution school view of property
deliberation 170 rights 123
DeLuca, K. 169 Dixon, J.A. 332
DNR (Queensland Department of decomposition analysis and 363,

Natural Resources, Mines and 378, 388–9, 396–9
Energy) 178, 179, 180 empirical models on 363–6
Dobson, A.P. 17 empirical studies on 366–7
Dobzhansky, T. 77 income elasticity of demand for
Doggart, N. 48 environmental quality 366,
domestic politics 214–15 371, 378, 391–4
dominant strategy 125, 315, 316 role of ecological thresholds 363,
Doornbosch, R. 410 399–403
Douglas fir (Pseudotsuga menziesii) 104 role of household preferences
Dove, M.R. 132 391–4
Dowling, E.T. 356 role of international trade 365,
Downing, J.A. 80 367–8, 375, 384–8, 398, 400
Drake, L. 148 role of policies and institutions
Drennan, L.G. 135, 136–7 202–3, 363, 365–6, 369,
Dresher, M. 329 376–84, 399, 402
dried fish 250, 251, 252, 259, 264, 270, role of population density 364,
272 370, 371, 377, 381, 388–91
drought 80, 83, 90, 294 role of technology shifts 366,
decision problem dependent on 369–70, 385, 394–5
pattern of 311–14 economic incentives 97, 104–6
Dryzek, J.S. 168, 171, 173, 174–6, 185, see also agricultural subsidies;
186, 187, 190, 192, 193 taxes
Dubois, O. 32 economic rents 122, 135, 136, 159
dugongs 260 economic school view of property
Dulvy, N.K. 21 rights 122–3
dumping of pollution 221–3, 229, 379, ecoregions 13–14
380, 384, 387–8, 400 ecosystem disturbance and biodiversity
dynamite fishing 261, 272 87–91
moderate ecosystem disturbance and
ecological paradigms 296–8 biodiversity 91–5
ecological thresholds 91, 94, 363, ecosystem services 22, 100
399–403 valuation of 76, 102–4, 331–2,
Economic Commission for Europe 335–6, 340, 343–4, 351–5
(ECE) 227 ecotourism 39, 43
economic efficiency Edelman, M. 168
conditions for 95, 96, 97–8 education, access to 276, 383
distinction between equity and 95 education and awareness programmes
institutions and 119, 121, 122, 124, 57, 58, 61, 149, 277
125, 129–30, 139, 143–5, 152, Edwards, K.J. 83
153, 154 effort-intensity 137
international agreements 207, 213 Eggertsson, T. 122, 135, 158
economic growth Egypt 221
discourse of 169–70, 175 Eken, G. 23
environmental effects of 6, 359–403, El Niño Southern Oscillation 284
408 Elliott, L. 175
in Asia 241 Ellis, G.M. 76, 102
basic studies of environmental Ellis, J.E. 298
Kuznets curve 367–76 Elton, C. 79
debate concerning 359–63 emission reductions 210
Index 439
endangered species 102, 136, 137, 179, epistemic communities 5, 203, 205,
185 220, 231–2, 233, 235
endemic bird areas 9, 10, 12, 14, 18 equal-area grids 13
endowment 369, 377 equilibrium theory, ecological
ENERDATA database 380 paradigms based on 5, 296–8,
energy consumption 372, 385–6, 400, 302–3
410 policy discourse associated with 6,
see also net energy balance 299–300, 301, 303, 304
Energy Crop Payments 413 equity
enforcement costs 136, 138, 142, 145, development projects and 65
146, 149, 150 distinction between economic
England 88 efficiency and 95
environment and society institutions and 119, 137, 145–6,
ecological and social paradigms and 149, 152, 153, 157–8, 161
5–6, 295–304 non-equilibrium theory and 301
reciprocal relationship between see also income and wealth
283–95, 303–4 inequality
environmental effectiveness 206, 208, Er, K.B.H. 13
233, 234, 235 Erdelen, M. 92
Environmental Impact Assessment Erwin, T.L. 21
(EIA) 221, 223, 225 Espelta, J.S. 87
environmental Kuznets curve 6, 360, ethanol 412, 413, 414, 415, 416, 418,
362 425
basic studies of 367–76 Ethiopia 291
decomposition analysis and 378, EU Directive 2003/30/EC 409, 411–12,
388–9, 396–9 416, 425
ecological thresholds and 399–403 EU Directive 2003/96/EC 413
household preferences and 391–4 EU Strategy for Biofuels 415, 416, 426
international trade and 367–8, 375, European Commission 411, 412,
384–8, 398, 400 413–14, 415, 416, 417, 423–4, 426,
policies and political institutions and 427, 428
369, 376–84, 399, 402 European Convention on Long-range
population density and 370, 371, Transboundary Air Pollution
377, 381, 388–91 (LRTAP) 211
technology shifts and 369–70, 385, European Habitats Directive 83
394–5 European Parliament 411
environmental price 402 European Union
Environmental Protection Act Common Agricultural Policy (CAP)
(Queensland, 1994) 180 97, 106, 413
environmental protection agencies funding for ICDPs 48, 69
(EPAs) 375 participation in agri-environmental
see also Queensland Environmental programmes in 148
Protection Agency (EPA); participation in Mediterranean
US Environmental Protection Action Plan 217, 221, 231
Agency policy on biofuels 7, 409, 410–17,
Environmentally Sensitive Area (ESA) 425–7
scheme 105–6 potential for double benefits from
EPA (Queensland Environmental biofuel use 422–5, 427
Protection Agency) 178, 179–80 evening grosbeak (Hesperiphona
EPA AIRS database 393 vespertina) 104
evergreen oak forests 82 fishing boat owners 250, 255–8, 263,

evolutionary processes 21–2, 94 265–6, 269, 278
exclusion of non-members 127, 131, fishing regulations 225, 244, 274, 277
132, 133 fishing traditions 260–63
exclusivity in consumption 126–7 fixed effects model 398
see also non-exclusivity in Fjeldså, J. 22
consumption flagship species 77
existence value 99 Flannery, T. 176
exit options 153 Flintan, F. 31
expropriation, risk of 377 Flood, M. 329
externalities 96, 120, 121, 125, 128, Florida mangroves 94
129, 133, 134, 137, 153, 335–6 floristic diversity 82, 92
Flournoy, A.C. 171
Fa, J.E. 22 focal biological elements, definition of
faecal coliform 368, 369, 382, 384, 391, 35, 37
399 Folke, C. 132
Falconer, K. 139, 140 fon (traditional chief) 60, 61
Faliñski, J.B. 88 Fonseca, G.A.B. da 16, 17
FAO (Food and Agriculture Food and Agriculture Organization
Organization) 75 (FAO) 218, 227, 287, 288, 371,
Faraco, A.M. 84 381
Farber, S. 400–401 food prices 410
Farquhar, G.D. 285 Forboseh, P. 63
faunal products, valuation of 337, 338 Ford, H.A. 88
Fay, P.A. 89 FORECE model 93
Feeny, D. 131, 132, 134 forest boundaries, demarcation of 57,
fence traps 248, 251, 267 59
fences 91, 139, 155, 333 forest conservation
Fenoaltea, S. 137 Bamenda Highlands, Cameroon
Ferguson, M.J. 419, 420 56–64, 68, 69
Ferraro, P.J. 33, 68 development activities consistent
Ferrier, S. 20 with 39
fertilizers 418, 419, 420 Uluguru Mountains, Tanzania
Festuca rubra 94 45–56, 67, 68–9
financial costs of conservation 105, 106 value of 42, 45, 55, 61, 64–5
Finland 375, 394, 397 Forest Management Institutions
Fire Wardens 179 (FMIs) 59, 61, 62, 63
fires 83, 84, 87, 91, 92, 94–5, 285, 292 Forest Officers 54, 60–61
management in Cape York, Forest Reserves 45, 46, 47, 48, 52, 53,
Australia 4–5, 176–94 54, 56, 60, 62, 68, 301
see also burning forest tent caterpillars 93
fish prices 250, 251–3, 257, 259, 263–4, Forestry Commission 101
270 forests, economic valuation of 102,
Fisher, A.C. 76, 102 331, 334–8, 340, 341, 344,
Fisher, K.C. 98 348–9
Fisher, R.J. 31 FORGRA simulation model 88–9
fisheries 125, 128, 129, 149–50 fossil fuels 373–4, 400, 408, 409, 411,
see also seagrass fishery on Quirimba 412, 418–19, 420, 421–2, 423–5
Island, socioeconomic aspects Foucault, M. 172, 173
of Fox, L.R. 89
Index 441
France Garrod, G.D. 101

economic growth and pollution in Garton, E.O. 104
380, 394, 397 gasoline lead content 379, 383
role in Mediterranean Action Plan Gastil, R.D. 377
(MAP) 219, 221, 228, 230, 234 Gayanilo, F.C. 259
Franke, B. 418, 419, 420 Gaze, P.D. 92
Franks, P. 31, 32, 35 GEF (Global Environment Facility)
free-riding 125, 130, 132, 154, 317, 322, 16–17, 48, 49, 69, 277
362 Gell, F.R. 242
freedom, effect on income–pollution gene pool 77
relationship 384 general equilibrium theory 95
Freedom House Tables 381 genetic diversity 76–7
FRELIMO government 262 Gengenbach, H. 261, 262
French, D.D. 93 Genista scorpius 84
frequency of transactions 136, 137, 159 GeoNetWeaver 41
freshwater biodiversity 21 Georgia 423
frontier forests 9, 10, 13, 18 Germany
Frontier-Moçambique Quirimba biofuel usage in 412–13, 414, 422
Archipelago Marine Research economic growth and pollution in
Programme 242, 243 394, 395, 397
fruit production function 348–9 position in international
fuel wood negotiations 215
collection by locals 52, 53, 54, 259, Gessner, J. 243, 244, 274
273–4, 339 Ghimire, K.B. 30
harvesting and production of 61 giant Irish deer 89
shortage in Africa 2, 4, 171–2 Gibbs, J.N. 124, 132
Fuentes, E. 335 Giblin, J. 302
Fuller, R.J. 88 Gill, A.M. 177, 178
functional diversity 77–9 gill nets 247, 248, 255
functional redundancy 78 Ginsberg, J. 10
fur trade 123 Gleichman, M. 84
Global 200 ecoregions 9, 10, 18
Galeotti, M. 374–5 global biodiversity conservation
Galicia, Spain 89 priorities 3, 8–23
game models 310 challenges facing global
game theory prioritization 17–22
CPR problems analyzed using 120, from global to local priorities 22–3
121, 316–28 global prioritization in context
commonly-used games for 10–16
depicting CPR problems 125, measures of irreplaceability
160, 317–19 11–12, 17
institutional solutions to CPR measures of vulnerability 12–13, 17
problems 319–21, 323 spatial patterns 14–16, 18–19
worked examples from semi-arid spatial units 13–14
Tanzania 321–8 impact of global prioritization 16–17
and international environmental nine major institutional templates
agreements 200, 201–2, 210 of 8–10
language and representation forms Global Conservation Fund 16
310–16 Global Environment Facility 16–17,
Gardner, R. 152 48, 49, 69, 277
Global Environment Monitoring Griffiths, C.J. 287

System (GEMS) 367, 368, 377, Griffiths, J.F. 284
396 Grima, A.P.L. 143
Global Plan of Action (GPA) 229 Grimes, P.E. 364, 373
global warming 6–7, 173, 359–60, 362, Grossman, G.M. 360, 364, 366, 367–8,
373, 408, 410, 423, 427 376, 383, 385
see also climate change ground parrot 189
globalization 202–3, 234, 241 group heterogeneity, effect on
goats 84, 243 collective action 150–58, 161
Godoy, R. 332, 335, 337 Grove, R. 30, 31, 299
golden shouldered parrot (Psephotus Guard, M. 261, 267
chrysopterygius) 179 Guatemala 337
González Bernáldez, F. 83 Guggenheim, S. 152, 157
Goodstein, E. 361, 362 Gulliver, P.H. 292
Gordon, H.S. 124–5 Gurr, T.R. 383
gorillas 35, 42 Gwynne, H. 287
Government of Tanzania (GOT) 50,
69 Haas, P.M. 200, 201, 202, 205, 206–7,
government regime, effect on income– 215, 216, 217–18, 219–20, 231–2,
pollution relationship 382–3 234, 235
government scientists’ policy stance on Habermas, J. 171
burning 178–80, 181, 182 habit forming cooperation 151
discourses adopted in support of Hackel, J.D. 30, 42
182, 183, 185, 186, 187, 188–9, Hadejia-Nguru Wetlands
192 Conservation Programme
Graddy, K. 383–4 (HNWCP) 335, 338, 350
Graham, R.T. 89 HADO (Hifadhi Ardhi Dodoma – soil
Grainger, A. 97 conservation in Dodoma) project
grazing 59, 60, 61, 132, 156, 262, 286, 301
293, 295, 333 Hadza tribes 291
environmental effects of 83, 84, Haigh, N. 107
86–7, 88–91, 92–4, 97, 285, Haiku poetry 187, 190
296–9, 300–301, 333 Hajer, M.A. 171, 172–4, 181, 182, 184,
Greece 221, 394 192, 193
Green Party 215 Halkos, G.E. 398–9
green radicalism discourse 175, 185, Halpern, B.S. 16
186 Hamersley, A. 284
greenhouse gas emissions 6, 359, 395 Hamilton, A.C. 284, 285
biofuels and 409–10, 411, 416, Hammig, M. 365, 381
417–18, 419, 423, 424, 425, 426, Hanemann, W.M. 101, 336
427 Hanley, N. 96, 101, 105, 106, 107
Kyoto protocol and 7, 214, 408–9, Hanna, S. 138, 139, 158
411–12 Hannah, L. 31, 37
see also carbon dioxide (CO2) Hansen, L.A. 48
emissions; CFC emissions Hardin, G. 1, 68
grey list 222, 223 Harmon, M.E. 336
Grice, T.C. 178 harmony 401
Griffin, J.R. 87 Harper, N. 16
Griffin, R.C. 136 Hartley, D. 47, 52
Griffiths, C. 364, 365, 370–71, 388 Hartnett, D.C. 89
Index 443
Hastings, A. 168, 172, 194 Hoffman, M.T. 90, 91

Hatton, J. 244, 262, 264, 270, 271, 274 Holling, C.S. 78, 79, 80, 87, 91, 93, 94
Hausman, J.A. 101 holm oak dehesas 83, 85
Hawaii 14 Holmes, R.T. 87
Hawkins, J. 266 Holtz-Eakin, D. 372–3
hazard reduction burning 178, 179 Homer-Dixon, T. 215
hazardous wastes 223, 226 Homewood, K. 290, 293, 299, 300, 301
Heal, G.M. 128 honey 57, 59, 61
Healey, P. 170, 171, 174 Horrill, A.D. 90
health care, access to 275–6 horses 88
health risks from biodiesel 420 Hough, J.L. 31
Heath, M.F. 83, 88 household preferences, role in income–
heather moorland 97, 106 pollution relationship 391–4
Hecht, S. 337, 338 Hovi, J. 211
Hechter, M. 142 Hughes, R. 31
hegemonic stability theory 201, 234 Hughes, T.P. 21
Heidebrink, G. 366, 371 Hulme, D. 31, 33, 42
Hein, D. 92 Hulme, P.D. 87, 94
Heller, M. 194 Human Development Index 244
Helm, C. 209, 210 Humphries, C.J. 12
Hemmer, W. 420 Hunter, M.L., Jr. 8
Henry, W. 102 hunting 42, 45, 52, 54–5, 61, 88, 98,
Hernandez, M.P.G. 89 176, 225, 291, 299, 333, 334
Hernroth, L. 244 Huntoon, P.W. 336
Herrera, J. 87 Hutchinson, A. 8
Herrera, S. 414 Hutchinson, G.E. 78
Hester, A.J. 89 hybrid private sector governance,
Heston, A. 368, 373, 379, 381, 382, transaction costs of 143–6
383, 390, 398 Hyden, G. 294
Hettige, H. 363, 375–6, 384–5 hydrocarbon emissions 392, 418
Heywood, V.H. 332, 356 Hymas, O. 52, 53
high-biodiversity wilderness areas 9,
10, 16, 18 IAEA (International Atomic Energy
Hill, C. 347 Authority) 227
Hill, I. 158 Ibo Island 242, 243
Hill, J. 409, 418 IBRD (International Bank for
Hill, R. 176, 177 Reconstruction and Development)
HilleRisLambers, R. 285 360, 361, 362
Hillier, J. 170 ICBP (International Council for Bird
Hilton, H.F.G. 379–80 Protection) 9
Himalayas 14 ICDPs see Integrated Conservation
historical materialism 202–3, 219–20, and Development Projects
234 (ICDPs)
HNWCP (Hadejia-Nguru Wetlands Iceland 397
Conservation Programme) 335, identity, link between language and
338, 350 169, 191
Hobbes, Thomas IDRISI 41
Leviathan 200 IIED (International Institute for
Hobohm, C. 14 Environment and Development)
Hoekstra, J.M. 9, 12, 13, 16 31, 55, 335, 338, 350
Ijim Ridge 56 information set 311

Iliffe, J. 288, 290, 292, 293, 300 infrastructure quality 381
Illius, A.W. 297 see also roads
implementation constraints 67 Inhaca Island 269, 270, 271
implementation costs 138, 141–5 Innes, J.L. 13
passim insects 17–20, 87, 91, 93, 104
imputed values for non-marketed institutional choice theory 319
goods 345, 350, 355 institutional effectiveness 206, 208,
income effect 121 233, 234, 235
income elasticity of demand for institutional failure 132, 159
environmental quality 366, 371, institutions and the management of
378, 391–4 environmental resources 4, 119–62
income and wealth inequality definitions of institutions 120, 204
and economic growth 367 effect of policies and institutions on
and participation in collective action income–pollution relationship
151–8, 161 363, 365–6, 369, 376–84, 399,
and pollution 374, 382 402
independent evolutionary history group heterogeneity and emergence
(IEH) 77 of local management
independent scientists’ policy stance on institutions 150–58, 161
burning 180, 181, 182 international agreements see
discourses adopted in support of international environmental
182, 184–5, 186, 187, 191–2 regime effectiveness
India new institutional economics (NIE)
biodiversity conservation priorities and 119–21, 159
in 14 open access versus common
biofuel cost in 421 property regimes 124–7, 131
biofuel demand from 415, 426 property rights transformation and
economic growth and pollution in 375 resource management 121–4
entitlement to products of the theoretical aspects of common
commons in 158, 161 property regimes 130–34, 160
exempt from reducing carbon theoretical aspects of open access
emissions 409 regimes 127–30, 159–60
transaction costs associated with transactions costs and natural
group-owned wells in 148–9 resource management 134–50,
Indian Ocean 243, 263, 284 159, 160–61
indirect use value 100, 331–2, 335–6 institutions and the structuring of
measurement of 102–4, 340, 343–4, discourses 173–4, 176
351–5 integrated coastal zone management
Indonesia 14, 375, 415, 416, 423, 426, (ICZM) programmes 271
427 Integrated Conservation and
industrialism 175, 186 Development Projects (ICDPs) 3,
INFO/RAC (Information and 30–70
Communication Regional advantages of 32
Activity Centre, previously ERS/ Bamenda Highlands of Cameroon
RAC) 229 56–64, 68, 69
information asymmetry 136, 138, 141, creation of 30–32
143 critiques of 32–4
information costs 135, 139, 146, 148, improving the model 34–45
149 issues arising from case studies 64–8
Index 445
Uluguru Mountains of Tanzania International Whaling Commission

45–56, 67, 68–9 209
interest groups 59, 61, 125 Intertropical Convergence Zone
intergenerational displacement of (ITCZ) 283–4
environmental costs 361 invertebrate fishery 245, 248, 252, 254,
Intergovernmental Panel on Climate 267–8, 269, 270, 272, 278
Change (IPCC) 408 invertebrates 17, 20
International Atomic Energy Ioris, E.M. 337
Authority (IAEA) 227 IPCC (Intergovernmental Panel on
International Bank for Reconstruction Climate Change) 408
and Development (IBRD) 360, Iraqw tribes 291
361, 362 Ireland 83, 89, 394
International Council for Bird irreplaceability
Protection (ICBP) 9 measures of 11–12, 17
International Energy Agency (IEA) relative to vulnerability 10–11,
374, 386 14–16, 18–19
international environmental regime irreversible loss 137, 142, 208, 363, 399
effectiveness 5, 198–237 irrigation systems 153, 155–6
defining and measuring 205–12, Isenmann, P. 78
234–5 Isla de Moçambique 271
new approach to 234–7 Islam, N. 396
domestic politics and 214–15 Islam 243, 245, 260, 272
institutional economics and 213 Isle of Man 260
Mediterranean Action Plan (MAP) Israel 93, 221
5, 216–33, 235–7 Italy 103, 221, 229, 394, 397
regime theory and 199–205, 219–20, itinerant fishers 271–3, 274, 276–7
234 IUCN (International Union for
security considerations and 215–16 Conservation of Nature and
transparency, openness and Natural Resources) 9, 11, 13, 16,
participation and 214 77, 301
verification of compliance and IUCN Species Survival Commission 20
213–14, 229–30 ivory trade 243
International Institute for
Environment and Development Jack, M. 32
(IIED) 31, 55, 335, 338, 350 jacks (Carangidae) 248, 251
International Maritime Organization Jackson, T. 107
224, 228 Jacobs, B.F. 285
International Monetary Fund (IMF) Jacobs, K. 172
401 Jacobson, K. 212
international political economy Jaggers, Keith 383
approach 202–3, 234 James, A. 8, 17
International Road Federation 383 James, F.C. 92
International Standard Industrial Jane, G.T. 89
Classification (ISIC) 384–5 Japan 161, 380, 394, 397
international trade, effect on income– Jeanrenaud, S. 31
pollution relationship 365, 367–8, Jeftic, L. 231
375, 384–8, 398, 400 Jeltsch, F. 89
International Union for Conservation Jennings, S. 242
of Nature and Natural Resources Jensen, O.B. 173
(IUCN) 9, 11, 13, 16, 77, 301 Jepson, P. 14, 17, 21, 22
Joffre, R. 85 Knetsch, J. 100

Johannes, R.E. 260 Knoll, A.H. 21
Johns, A.D.88 Knox, A. 137
Johnson, L.A. 102 Koh, L.P. 410
Johnson, O.E.G. 121 Köhler-Rollefson, I. 85
joint forest management (JFM) 158 Komen, M.H.C. 394
Jones, R. 176 Kondoa 301
Jorritsma, I.T.M. 88 Koop, G. 371
Juma, S.A. 267, 269 Koponen, J. 293
Just, R.E. 103 Korea 375, 380
Koziell, I. 32
K-strategist species 91–2 Kramer, R. 30
Kaare, S. 47, 52 Krasner, S.D. 203–5, 234
Kahn, M.E. 363, 392 Kremen, C. 43
Kalahari 89 Kress, W.J. 12
Kanbur, R. 153 Krueger, A.B. 360, 364, 366, 367–8,
Kant, S. 141, 151, 157 383, 385
Kareiva, P. 21, 22 Kruess, A. 91
Kaufmann, R. 365, 390–91 Krupnick, G.A. 12
Keefer, P. 377, 381 Krutilla, J.V. 99, 100
Keeley, J. 173 Kullenberg, G. 216
Kelly, C.K. 17 Kumilamba 242, 245–6
Kenya 43, 260, 261, 267, 272, 289 Kumm, K.I. 148
Keohane, R.O. 150, 200, 204 Kundhlande, G. 336, 340, 352, 355
Kerala 424 Kunich, J.C. 17
Kerario, E. 301 Küper, W. 17, 20
keystone species 78, 94 Kuperan, K. 149–50
Khanna, N. 393–4 Kütting, G. 200, 201, 203, 206, 208,
Kherallah, M. 120 217, 233, 234, 235
Khoisan tribes 291 Kuznets, S. 360, 367
Kienast, F. 89, 93 Kuznets curve 367
Kieran, J.A. 290 see also environmental Kuznets
Kikula, I.S. 294, 302 curve
Kilum-Ijim forest 56–64 passim, 68, 69 kwifon (ruling council) 60, 61
Kilum-Ijim Forest Project 63, 64, 69, Kyoto Protocol 7, 214, 408–9, 411–12
70
Kimwani dialect 243, 245, 247, 248, Laffer curve 401
249, 276 Lal, R. 336
Kingdon, J. 284, 285, 286 Lamoreux, J.F. 12, 14
Kingery, J.L. 89 land abandonment 75–6, 83–4
kinship networks 261–2 land disputes 262, 295
Kirby, D. 309, 332 land tenure 13, 36, 295, 298, 333, 380
Kirsten, J. 120 landscape-level ICDP design 34–6,
Kiss, A. 33, 68 49–50, 58–9
Kitzes, J. 408 Langdale-Brown, I. 285
Kjekshus, H. 293, 300, 302 language use see discourse analysis
Klaassen, G. 378 Lanza, A. 374–5
Klein, P.G. 135 large-scale mono-cultures 7, 419, 425
Klieman, D.G. 43 Larson, P.S. 31, 33
Knack, S. 150, 377, 381 last of the wild 9, 10, 18
Index 447
Late-glacial Interstadial (11000–12000 Luxembourg 397

BP) 89 Lyamuya, V.E. 50
Latin America Lynch, J.F. 88, 92
economic growth and deforestation
in 370, 381, 388 Maasai tribes 291, 293, 300
forest values in 337–8 MacArthur, J.W. 92
Leach, M. 2, 30, 33, 43, 171, 299, 300, MacArthur, R.H. 79, 91, 92
301 Mace, G.M. 8, 12, 17, 21, 22
lead (Pb) pollution 368, 376, 379, 383, machairs 83
384 machambas (allotments) 245–6, 252,
lead states 215 253, 255, 256, 259, 265, 268, 269
Leader-Williams, N. 42 Machiavelli, Niccolò
Leakey, L.S.B. 289 The Prince 199–200
Lebanon 221 Macia, A. 244
Lees, L. 193 MacKinnon, K. 31
Leffler, K. 148 MacLeod, N.D. 296, 297
Levinson, A. 366, 379–80 Madagascar 14, 21, 22
Levy, M.A. 204 Maddox, G. 302, 303
Libecap, G. 120, 122, 123–4, 136, 158, Mafia Island 270
161 Main, A.F. 79
Libya 221 Maisels, F. 63
Liebman, M. 103 maize 418
Lin, J.Y. 119 Maji Maji 293
Lindberg, K. 75 Majone, G. 170
Lindén, O. 241, 271 Makonde people 243, 244
Linder, P. 92 Makua people 243, 244, 260, 262
line fishing 248, 251, 273 Malawi 43
Liu, J. 13 Malaysia 14, 389–90, 415, 416, 423, 426
livestock movements, restrictions on Maler, K.G. 158
90–91 Mali 156–7
livestock production function 339–40, malleefowl 189
343, 344, 345–8 Malta 221, 224, 228
Local Authority Forest Reserves 54 mammal herbivore species 86
Logan, V.S. 14 mammal species 17, 20, 63
Logical Framework Approach 39–40, managed landscapes, biodiversity
44, 52, 60, 65–6 conservation in 3–4, 75–108
Lolium perenne 85 ecosystem disturbance and
Londo, G. 83 biodiversity 87–91
Long, A. 8, 14 moderate ecosystem disturbance
Long, J. 103 and biodiversity 91–5
Loreau, M. 22 genetic and functional diversity 76–9
Loureiro, N.L. 251 importance of biodiversity in
Lovejoy, T.E. 79 managed landscapes 79–82
Lovett, J.C. 22, 48, 139, 143, 180, 284, landscape management and
285, 286, 299, 300, 301 biodiversity 82–4
Lowore, J. 32 landscape management and
Lughadha, E.N. 12, 14, 17, 21, 22 ecosystem properties 84–7
Luguru tribe 48, 49, 52 markets and 95–108
Lukumbuzya, K. 42 management agreements 105, 106
Lundin, C.G. 241, 271 mangrove crabs (Scylla serrata) 249, 252
mangrove whelk (Terebralia palustris) Marlow, D. 178, 180

249 Maro, P.S. 284, 294
Manx Heritage Foundation 260 Marrs, R.H. 90
manufacturing Martin, G. 338
share in the economy 366, 375, 378, Marvier, M. 21, 22
379–80, 385, 386, 387, 388, 396 Masaiganah, M. 261
toxic intensity of 384–5 Masih, S.K. 335
MAP see Mediterranean Action Plan Mason, M. 170
(MAP) Massinga, A. 244, 262, 264, 270, 271,
Maputo 244, 277 274
Maquis, R.J. 87 Massoud, M.A. 231
Marañon, M. 82, 85, 86 Mate, I.D. 83
marema trap fishing 246, 247, 251, Matthews, E. 269
254–5, 258, 259, 261, 263, 266–7, Mattson, W.J. 93
268, 269, 270, 273, 278 Matzke, G. 294
marginal benefits 129, 130 May, R.M. 79
marginal cost 129, 130 Mbaya, S. 42, 68
marginal propensity to emit (MPE) McAllister, D.E. 21
373, 397 McClanahan, T.R. 260, 261, 272
marginal rate of substitution 392 McClean, C.J. 13
marginal value 351 McConnell, K.E. 366, 391–2
Margolius, R. 31, 40, 43, 56 McConnell, M.L. 231
Margules, C.R. 10, 11, 12, 16 McDonald, M.J. 171
marine biodiversity 21, 222, 225, 228 McHenry, D.E., Jr. 294
marine parks 222, 225, 271 McKean, M. 152, 158, 161
marine pollution McManus, J.W. 242
definition of 221, 236–7 McNaughton, S.J. 86, 90, 103
measuring 212 McNeely, J.A. 97, 105
oil pollution 215, 218, 222, 224 MCSD (Mediterranean Commission
prevention in Mediterranean Sea on Sustainable Development) 227,
see Mediterranean Action Plan 229
(MAP) McShane, T.O. 31, 43, 67
market failure 96–8, 371, 402 Meadows, D.H. 175, 359, 360
correcting with command and Mearns, R. 2, 171, 299, 300, 301
control regulations 107–8 meat production 42
correcting with economic Mecufi Coastal Zone Management
instruments 104–6 Project 271
markets and biodiversity 95–108 Mecufi district 251, 255
indirect use values and ecosystem MED POL programme 217, 224,
function 102–4 227–30, 231
market failure 96–8 medicines 255, 275–6, 333, 335
correcting with command and Mediterranean Action Plan (MAP) 5,
control regulations 107–8 216–33
correcting with economic Barcelona Convention and 217, 219,
instruments 104–6 220–26, 229–30, 231, 232, 233
measuring diversity values 100–101 contracting parties 217, 221
policy failure 97 effectiveness of 231–3, 235–7
value of species and habitats 101–2 international environmental
valuing biodiversity 98–100 cooperation and the creation of
Marland, G. 374 218–20
Index 449
objectives of 217, 232 Mitchell, R.C. 101

origins, negotiation and formation Mittermeier, R.A. 9, 11, 12, 13, 14, 16,
217–18 20, 21
structure and components 220–31 mixed equilibrium 316
Mediterranean Commission on mixed strategy 316, 319
Sustainable Development Mlay, W.I.F. 294
(MCSD) 227, 229 Moçimba da Praia 277
Mediterranean region Molinas, J.R. 142, 155, 162
biodiversity conservation priorities Molinia caerulea 94
in 14 Monaco 221
biodiversity in managed landscapes Monbiot, G. 415, 423
in 82–3, 84, 85, 93 monitoring costs 136, 138, 142, 143,
keystone species in 78 144, 145, 150, 212, 327
see also Mediterranean Action Plan Montgomery, C.A. 102
Mediterranean Trust Fund 231 Montreal Protocol (1987) 107, 209,
megadiversity countries 9, 10, 13, 18 214, 377
Meinzen-Dick, R. 137 Moomaw, W.R. 396–8
Menard, C. 119, 141 moose 88
mercury 368, 384 Moran, D. 13
Meshack, C.K. 139 Morgan, R.K. 87
Mesoamerica 14 Morgenstern, O. 310
meta-analysis 375 Morgenthau, H. 215
metaphors 176 Morocco 221
in Cape York seminar transcript Morris, R.F. 87
187, 188 Morrow, P.A. 89
methane emissions 372 Morse, C. 401
Metrick, A. 77 Mortimer, N.D. 418, 419
Mexico 155–6, 337, 367–8, 375, 385 Mortimore, M. 284, 287, 288
MICOA (Ministry for the Mortished, C. 415, 423
Coordination of Environmental Mott, J.J. 176
Affairs, Mozambique) 271 Mount Cameroon 62, 63
microbes 17 Mount Oku 56
Middle East 423 mountain beech (Nothofagus solandri)
Midgley, G.F. 21 89
Midlarsky, M.I. 380 Mozambique
Miles, E.L. 209 civil war in 244, 262, 264
Mill, G.A. 101 integrated coastal zone management
Millennium Development Goals 7, in 271
282, 409 see also seagrass fishery on Quirimba
Millennium Ecosystem Assessment 8 Island, socioeconomic aspects
Miller, H.G. 104 of
Miller, K.R. 75 Mozambique Ministry for the
Miller, T.E. 92 Coordination of Environmental
MINFOF (Cameroon Ministry of Affairs (MICOA) 271
Forestry and Wildlife) 58, 60–61, multiple Nash equilibria 316, 318
63 multispecies game systems 87
mining 125, 128 multispecies herbivore systems 85–7
miombo woodland 285–6, 293 Munasinghe, M. 158, 401–2
Mitchell, F.J.G. 88, 89, 93 Mung’ong’o, C.G. 301
Mitchell, R.B. 206, 211–12, 234, 235 municipal waste 369, 372
Munro, J.L. 259 niche relations 92

murex shells (Chicoreus ramosus) 249, Nicholson, S.E. 284
268 Nicholson, W. 350
Murphree, M. 42 nickel 368, 384
Mussa, D.T. 284 Nilotic tribes 291, 292
Mwalyosi, R.B.B. 287 nitrates 372, 384, 387
Mwani people 243, 262, 273 nitrogen mineralization 85
mycorrhiza 103–4 nitrogen oxides (NOx) emissions 363,
Myers, N. 8, 9, 12, 16, 21, 107, 336, 370, 372, 376, 387, 391, 393, 395,
351 418, 419
Myerson, G. 193 nodes 311, 312
Nolan, T. 86
Nabil, M.K. 119 non-convexities in production
Nacala 271, 272 functions 154–5
Nampula province 250, 260, 261, 271, non-cooperative extensive form games
272, 273, 276 311–14
Nardus stricta 94 non-cooperative game theory 125, 310
Nash, J. 328–9 key concepts used in 311–16
Nash equilibrium 210, 316, 317, 318, non-cooperative strategic form games
319 314–16
National Carbon Intensity (NCI) 373 non-equilibrium systems 90, 94–5
National Parks 16, 33, 34, 60, 68, 100, non-equilibrium theory, ecological and
134, 180, 299, 300, 301 social paradigms based on 5–6,
national security 200, 215–16, 234, 409 297–9, 303
nationalization 124, 127, 381 policy discourse associated with 6,
Nations, J.D. 337, 338 300–301, 303, 304
natural relationships, assumptions non-exclusivity in consumption 127,
relating to 176 128, 131, 133, 134, 160, 316–17,
in Cape York seminar transcript 362
187, 188–9, 190, 191 see also exclusivity in consumption
Nature Forest Reserves 54, 68 non-governmental organizations
NatureServe DSS 41 (NGOs) 31, 32, 40, 48, 58, 63, 65,
Naveh, Z. 83 171, 214, 228, 323
Naylor, R. 410 non-linear dynamical analysis 397
Nee, S. 17 non-rivalry in consumption 362
negotiating costs see bargaining costs see also rivalry in consumption
neoliberal institutionalism 203–5, 234 non-use values 99, 100
neorealism 200–202, 219, 234 Norris, K. 16
Nepal 127, 129, 149, 156 North, D.C. 122, 131, 135, 159, 213
net energy balance 417, 419, 420, 421 North American Free Trade
Netherlands 88–90, 375, 394, 395, 397 Agreement (NAFTA) 367
Neumann, R.P. 30, 31, 299 North–South divide 202, 219–20, 233
New Guinea 14 northern spotted owl 102
New Forest, England 88 Norway 394
new institutional economics (NIE) Nossif, R. 174
119–21, 159 Novotny, V. 17
New Zealand 89, 93 Nugent, J.B. 119
Newmark, W.D. 31 nurse effects 87
Ngazi, Z. 255, 269, 270 Nwafor, O.M.I. 418–19
Niamir-Fuller, M. 332 Nyerere, J. 300
Index 451
O’Brien, E.M. 283, 284, 285 Oslo-Potsdam solution 201, 202,

O’Connor, C. 13 209–11
O’Connor, T.G. 297 Ostrom, E. 124, 125, 132, 150, 151,
O’Neill, R.V. 80 152, 156, 158, 170, 310, 316, 317,
Oak Ridge National Laboratory 318, 319, 320, 321, 329
(ORNL) 373, 374, 395, 396 outboard engines 265–6
Oakerson, R.J. 127 Ovadia, O. 14
Oates, J.F. 30, 33, 43 Overseas Development Administration
Oba, G. 298, 299, 302 (ODA) 287
Ockwell, D.G. 2, 169, 180, 194, 283, ownership by capture 128–9, 134
302, 304, 321, 332 oysters (Pinctada nigra) 248, 252, 267,
Octel’s Worldwide Gasoline Survey 268, 270
379 ozone depletion 199, 362, 376, 418, 419
octopus (Octopus vulgaris) 248, 252, ozone pollution 376, 393
268, 273
ODA (Overseas Development paddocks 90
Administration) 287 Page, T. 401
Odling-Smee, L. 22 Pakistan 103, 393
OECD Environmental Data Pakistan Integrated Household Survey
Compendium 378, 380, 395 393
oil pollution 215, 218, 222, 224 palm oil 415, 419, 421, 423, 426, 427
oil price shocks 397, 398 Panayotou, T. 336, 360, 363, 364, 365,
oil reserves 128, 129 366, 370, 374, 377–8, 388–9, 396,
oil supplies, security of 423 402
oilseed rape 414, 415, 416, 418, 419, PAP/RAC (Priority Actions
420, 421, 423, 425, 426 Programme Regional Activity
Oiterong, E. 269 Centre) 228, 231
Oliveira, J.A.P. 422 Paraguay 155
Olson, D.M. 9, 12, 13, 14, 16, 21 parasitism rates 91
Olson, M. 125, 154 Pareto improvement 341
Olsson, O. 120, 159 Pareto optimality 136
OMM (Organizaçao da Mulher Parmesan, C. 13
Moçambicana – Mozambican Participatory Environment
Women’s Organization) 245 Management Programme 35
ontology of a discourse 176 Participatory Rural Analysis (PRA)
Oommachan, M. 335 338
Opdam, P.F.M. 88 pastoral lifestyle, causes and
open access regimes consequences of changes in 333–4
theoretical aspects of 127–30, pastoral paradigm 298–9, 301, 302–3
159–60 pastoralism, definition of 332
versus common property regimes pastoralists’ policy stance on burning
124–7, 131 178, 181, 182
openness 214 discourses adopted in support of
opportunistic behaviour 138 182, 183–4, 185, 186, 187,
opportunity costs 105, 106, 139, 140, 189–90, 192
143, 149, 312, 337, 393, 421 Paterson, M. 200, 201, 202, 203, 204,
option value 98 216
Opuntia 79 Patzek, T.W. 421
Orkney 97, 106 Pauly, D. 259
Orme, C.D.L. 12, 17 Pautsch, G.R. 336
Pavasovic, A. 231 plant products, valuation of 335, 337,

Payments for Environmental Services 339
43, 55 plant species, conservation of 9, 10,
payoff, expected 312–28 passim 12, 17, 18, 20, 35, 48, 55, 63, 82–3,
Pearce, D.W. 99 87, 92
peasant cooperatives 155 Plassmann, F. 393–4
peat bogs 101 Platteau, J.P. 121, 128, 150, 151, 154,
Peberdy, J.R. 284 155, 157, 161, 316
pedunculate oak (Quercus robur) 89–90 Poa trivialis 86
Pelletier-Fleury, N. 138 poaching 31
Pemba 246, 247, 248, 249, 250, 251, Pócs, T. 48
255, 257, 260, 264, 265, 267, 270, Poitrat, E. 417, 419, 420, 421
271, 275, 277 Poland 88
Penn World Tables 368, 371, 373, 374, Polasky, S. 75
377, 379, 381, 386, 396, 397, 398 policy failure 97, 402
perennial herbaceous species 85 Polity III database 383
Perkin, S. 31 polluter pays principle 221
Perlez, J. 427 pollution
Perman, R. 95, 98, 99 economic growth and 6, 359–403,
Permit to Light Fire system 178, 179 408
Perrings, C. 95, 103, 366 international agreements on 199,
Peru 337 211, 215
pervasive power 193, 194 see also Mediterranean Action
pesticides 104, 418, 419, 420 Plan (MAP)
Peterken, G.F. 88 measuring of 212
Peters, C.M. 335, 337, 338 in open access regimes 129
Pezzey, J. 95 and rainfall 284
Pfaff, A.S.P. 392–3 transaction costs associated with 148
pH 389 see also under specific types of
Phalaris aquatica 85 pollution, e.g. carbon dioxide
Philippines 14, 149–50, 375 (CO2) emissions; water
Phillips, O. 337 pollution
phosphorous 387 pollution activity 379, 380
phylogenetic diversity 21 pollution dumping 221–3, 229, 379,
Picea abies 88 380, 384, 387–8, 400
pied flycatcher (Ficedula hypoleuca) 83 pollution haven hypothesis (PHH) 387
Pignatti, S. 82, 83 pollution intensity 375–6, 379, 380,
pigs 88 384–5
Pimbert, M.C. 30, 31 Polunin, N.V.C. 242
Pimentel, D. 421 Pomeroy, C. 269
Pimm, S.L. 8, 11, 13, 80 Pope, R.D. 103
Pimmental, D. 75, 82 population density, effect on income–
pine 92 pollution relationship 364, 370,
see also scots pine (Pinus silvestris) 371, 377, 381, 388–91
Pinedo-Vasquez, M. 337, 338 Populus 88
pink ear emperor fish (Lethrinus Portney, P.R. 101
lentjan) 247 Portugal 394
pinna shells (Pinna muricata) 248, 252, Posey, D.A. 34
267, 268, 270 Posner, R.A. 120
pioneer species 78, 92 Possingham, H.P. 12
Index 453
Poundstone, W. 329 property rights 2, 4, 119, 371, 383

poverty 42, 47, 244, 332, 333, 359 Coase theorem and 120
alleviation of 22, 32, 34, 65, 67, 158, collective action approach focusing
282, 303–4, 360, 409 on 119–20
Power, M.E. 78 definition of 120
power empirical studies of transactions
discourse and 169–70, 172–5, 193, costs 146–50, 160–61
194 problems associated with 140–41
inequality and pollution 381–2 four different types of property
realism and 200, 219 regimes for common pool
Powlson, D.S. 421 resources 126
Pratt, D.J. 285 group heterogeneity and emergence
Pratt, J. 287 of local management
precautionary principle 107, 221 institutions 150–58, 161
preferred strategy 312 motivations for contracting 121,
pregnancy 260–61 123–4
Prendergast, J.R. 12 open access versus common
Pressey, R.L. 10, 11, 12, 14, 16 property regimes 124–7, 131
pressure groups 215 property rights transformation and
Pretty, J.M. 31 resource management 121–4
Price, A.R.G. 21 theoretical aspects of common
Price, G. 329 property regimes 130–34, 160
Priority Actions Programme (PAP) theoretical aspects of open access
217, 230, 231 regimes 127–30, 159–60
Priority Actions Programme Regional transactions costs and selection
Activity Centre (PAP/RAC) 228, of appropriate governance
231 structure 135, 138–9, 141–6, 159
prisoner’s dilemma game 125, 160, 201, protected area coverage 13, 75
317, 319–20, 322–3 Prunus africana trees 57, 59, 61
private cost 96, 129 public goods 96–7, 127, 130, 134, 362
private property 96, 121, 126–7, 129, public relevance of transactions 137,
133 144, 159
privatization 124, 138–9, 333, 334 public research and development
problem-solving discourse 175, 185, (PRD) 394
187 public sector governance, transactions
problem trees 40, 60 costs of 142–6, 149–50
production function approach to pufferfish and tobies (Tetraodontidae)
valuing ecosystems 341–5 260
direct use values measured by Punjab, Pakistan 102
339–40, 342–3, 345–50 Puppan, D. 418, 419
indirect use values measured by purchasing power parity (PPP) 363,
102–4, 340, 343–4, 351–5 367, 374, 377, 381, 382, 386, 394,
limitations to use of 344–5 396, 397
productivity, ecosystem 80–81, 84–7, pure Nash equilibrium 316
89–90, 93–4, 97, 103–4, 106 pure strategy 316
valuation of 331–56 Putnam, R. 162
profit function 348, 349, 350, 353, 354 Putuhena, W.M. 336
profit maximization 97, 317, 332, 341,
345, 361 QPWS (Queensland Parks and Wildlife
‘property’, meaning of 126 Service) 178, 180, 184, 190
quasi-option value 98 Rawat, M.S. 336

Queensland Department of Natural realism 199–200, 219, 234
Resources, Mines and Energy red deer (Cervus elaphus) 89–90
(DNR) 178, 179, 180 Redford, K.H. 10, 338
Queensland Environmental Protection redstart 83
Agency (EPA) 178, 179–80 Rees, J. 171
Queensland Parks and Wildlife Service ‘regime’, meaning of 126, 204
(QPWS) 178, 180, 184, 190 regime theory 199–205, 219–20, 234
Queensland Rural Fire Service (RFS) Regional Action Plans
178, 179 components of 220
Queensland State Government 178, see also Mediterranean Action Plan
180 (MAP)
Quercus 88 Regional Activity Centres (RACs) 227,
questionnaire surveys 346–8, 352–3 228–9
Quinn, C.H. 287, 289, 290, 301, 302, Regional Marine Pollution Emergency
321, 332 Response Centre for the
Quirimba Island, Mozambique see Mediterranean Sea (REMPEC)
seagrass fishery on Quirimba 224, 228
Island, socioeconomic aspects of regulatory design 138
Quissanga 261 Reid, W.V. 21, 75
Reinhammar, L.-G. 83
r-strategist species 91, 92 Reinhardt, G. 418, 419, 420
Rada, F. 336 religious rhetoric 190
radical uncertainty 171, 193 Relva, M.A. 89
Raftopoulos, E. 220, 226, 227, 230, 231 REMPEC (Regional Marine Pollution
rainfall Emergency Response Centre for
decision problem dependent on the Mediterranean Sea) 224, 228
pattern of 311–14 Renewable Energy Road Map 417
in semi-arid regions 90, 91, 283–8, repeated games 125, 201, 320
292, 294, 297–8, 301, 321, 322, Repetto, R. 97
332, 336 Republic of Mozambique State
calculating indirect use value of Secretariat of Fisheries 263, 264,
351–5 272
Ramanathan, V. 284–5 reputation 170
Rambal, S. 85 resilience, ecosystem 80, 87, 90, 94, 97,
Randall, A. 101 103
random coefficients method 398 resource context, description of 138
random effects model 398 respondent bias 101
Random Roadside Test 392 RFS (Queensland Rural Fire Service)
rapeseed 414, 415, 416, 418, 419, 420, 178, 179
421, 423, 425, 426 rhetorical devices 176
rapeseed methyl ester (RME) 418, in Cape York seminar transcript
419 187–92 passim, 193
Ratcliffe, P.R. 92 Richards, M. 120, 146–8, 149
rational choice theory 201–2 Richardson, T. 173
rationality, economic 98, 169–70 Ricketts, T.H. 16, 23
bounded 135 Righelato, R. 410
Ravallion, M. 374 rights, emphasis on 65, 301
Raven, R.H. 75 rinderpest 292–3, 300
Rawat, J.S. 336 Ringia, O. 48, 52
Index 455
Rio Declaration on Environment and San Salvador Island, Philippines

Development (Agenda 21) 218, 149–50
229 sanctions for non-compliance 211–12,
rivalry in consumption 128, 131, 133, 225, 236
134, 160, 317 Sanderson, E.W. 9, 12, 13, 16, 20
see also non-rivalry in consumption Sands, P. 199, 217
river pollution 368, 369 sanitation 368, 369, 382, 383, 390
see also water pollution Sanjayan, M.A. 31, 33
RME (rapeseed methyl ester) 418, 419 Santos, T. 82
roads 383 Sarkar, S. 22
Roberts, C.M. 21 savannas 42, 86, 87, 90, 97, 176, 285,
Roberts, J.T. 364, 373 288
Robinson, J. 338 economic valuation of 331, 334–6,
Roca, J. 361 338–40
Rock, M. 385 scale effect 388, 396
rodents 61 scarids 248
Roderick, M.L. 285 Schemske, D.W. 92
Rodgers, W.A. 31, 37, 290, 293, 299, Schindler, D.W. 78, 79
300, 301 Schlager, E. 150
Rodrigues, A.S.L. 17, 20 Schmalensee, R. 373–4
Roe, D. 32, 43 Schmida, A. 92
roe deer (Capreolus capreolus) 88, Schmidt, K. 10
89–90 Schmidt, L. 419
Rohde, R. 297, 303 Scholes, R.J. 87, 94
Roland, M.A. 91 Schrijver, N. 31
Romania 409 Schroeder, M. 91
Room, J. 139 Schubert, C. 414
Rose, D.B. 176 Science 3
Rosenzweig, M.L. 75 science assessment 208
Rothman, D.S. 363, 386–7 Scoones, I. 173, 296, 297, 299, 334,
Rotty, R.M. 374 338–9, 343, 350
Rousseau, J.J. 329 scorpionfish (Scorpaenidae) 260
routines 174 Scotland 83, 90, 93, 101
Royal Society for the Protection of scots pine (Pinus silvestris) 88–9, 90,
Birds (RSPB) 48, 69 93
rubber-vine 179 Scott, A. 124–5
Rucker, R. 148 Scrase, I. 169
Ruddle, K.R. 241, 260, 262 SCUBA equipment 249, 251
rule of law 377, 383 sea cucumber (Holothuriidae) 241, 249,
rules for decision-making 120, 204 250–51, 274
Runge, C.F. 125 Sea of Marmara 221
Russell-Smith, J. 179 Seabright, P. 151
Russia 14, 423 seagrass fishery on Quirimba Island,
Ryan, L. 419 socioeconomic aspects of 5,
Rydin, Y. 2, 168, 169, 170, 172, 173, 241–79
174, 175, 193, 194, 283, 304 history of the Quirimbas 242–4
reasons for study 241–2
safe minimum standards approach study area 242
107–8 study conclusions 278–9
Salafsky, N. 31, 40, 43, 56 study methods 244–5
study results and discussion 245–78 Semesi, A.K. 267

alternative sources of income semi-arid regions
268–70 climatic and vegetative
changes in fishing patterns and characteristics of 90–91, 283–8,
effect on economy 278 294, 321–2, 332
division of fisheries between the definitions of 287–8
sexes 266–8, 269 ecological and social paradigms and
economic gain from fishing 257–9, management of 5–6, 295–304
278 economic valuation of different
employment opportunities 264–6 forms of land-use 331–56
fish prices 250, 251–3, 257, 259, game theory applied to CPR
263–4, 270 management in Tanzania 321–8
fishing traditions 260–63 role of people and livelihoods in
integrated coastal zone 90–91, 288–95
management in northern Sen, A. 329
Mozambique 271 Serbia and Montenegro 221
investment by fishers 254–7, 278 Sere, C. 287
itinerant fishers 271–3, 274, 276–7 Serengeti 86, 90, 285
management implications and Serra King, H.A. 269
problems 275–7 Seychelles 266
provincial and recent historical Shafik, N. 360, 363, 368–70, 385, 393
context 263–4 shark fin fisheries 274
Quirimba during the study period Sharma, M.P. 421, 422
(1996–97) 245–6 sheep 86–7, 89, 94, 97, 106
salaries 255, 256, 257, 258, 263, Shelanski, H.A. 135
264, 278–9 Sherman, P.B. 332
social structure of fishing fleet 266 Shetland 97, 106
socio-cultural aspects of fish use Shi, H. 13
259–60 Shields, D. 158
threats to sustainable resource use Shorter, A. 289, 290, 291
273–5 shrub species 89
total value of fishery 263 Sillitoe, P. 242
types of marine resource use Silva, J. 90
246–53 Silva-Pando, F.J. 89
seagrass parrot fish (Leptoscarus silver birch (Betula pendula) 89–90, 92
vaigiensis) 247, 250, 251 Simberloff, D. 94
seahorses 251 Simpson, I.A. 97, 106
seamoth fish (Pegasidae) 260 Sisk, T.D. 13, 17
search costs 135, 138 SITES 41
Sechrest, W. 21 sites of special scientific interest (SSSI)
sectoral composition of the economy 107–8
366, 375–6, 378–80, 385, 386, 387, Skjaerseth, J.B. 216–17, 232–3, 235
388, 396 Slatter, S.M. 178
seed dispersal 78 slave trade 243
seine net fishing 246, 247, 254, 255–8, Sloan, R.J. 87
260, 261, 263–4, 266–7, 269, 270, Slovenia 221
272, 278–9 Smale, M. 103
Selden, T.M. 360, 364, 366, 371–3 small animal husbandry 39
Seligman, N.G. 284, 287 Smith, A. 401
Selous Game Reserve 293–4 Smith, C.D. 294, 303
Index 457
Smith, G. 170 Species of European Conservation

Smith, J.M. 329 Concern (SPECS) 88
Smith, R.J. 13, 121 Spector, S. 21
Smith, T.B. 21 Spinage, C.A. 30
Smith, V.L. 125 Spracklen, D.V. 410
smoke 367–8, 382, 391 Sprinz, D.F. 209, 210
snappers (Lutjanidae) 251 spruce 92
Snidal, D. 310 spruce bark beetle (I. typographus) 91
Snow, C.P. 1 spruce budworm 93, 104
social capital 143, 144–6 squid 252
social constructivist approach 173–4, Sri Lanka 375
175 stability, ecosystem 79–80, 87, 90–91,
social cost 96, 129 94–5, 102–3
social norms 120, 122, 131, 204 stag hunt game 201, 329
social paradigms 298–9 see also assurance game
socially optimal level 317 Staines, B.W. 90
Sofala Province 271 stakeholder involvement in ICDP
soil erosion 298, 301, 321–3, 336, 344, design 36–7, 40, 41, 50, 52–3
380, 390 standards, environmental 209, 214,
soil type 343 217, 218, 362, 376, 415, 427
Solow, A. 77 safe minimum standards approach
Somalia-Maasai region 285, 288, 291, 107–8
298 Stanhill, G. 285
Song, D. 360, 364, 366, 371–2 start-up costs 139, 154–5
Songorwa, A.N. 32, 34 state and transition models (STM) 91
soot 418 state capability 144–6
Soulé, M.E. 77 state property 124, 126, 127
Sousa, A.G. 243 see also nationalization; public
South Africa 14 sector governance, transaction
South Asia 157–8 costs of
South East Asia 269, 270 state security 200, 215–16, 234, 409
South Pacific 260, 269 Stattersfield, A.J. 9, 12, 13, 14, 16
Southwood, T.R.E. 17 Stavins, R. 214, 362
soya beans 410, 418, 419, 421, 423, Steenblik, R. 410
425, 426 Steenkamp, J. 331
SPA/RAC (Specially Protected Areas Stern, D.I. 398, 400
Regional Activity Centre) 228 Sternberg, M. 93
Spain 82, 89, 221, 228, 394 Stevenson, G.G. 124, 128, 129, 130,
SPAMI List 225 132–4, 160
Spanish Pyrenees 84 Stevis, D. 202
Spash, C. 101, 105 Stockholm Ministerial Conference 217
spatial decision support system (SDSS) Stocking, G.C. 176
software 41 Stocking, M. 31, 301
spear-fishing 248, 249, 265 stocking rates 97, 106, 298–9, 301, 339
Spears, J. 152, 157 Stoll, J.R. 102
specialist opinion, irreplaceability story-lines 173, 174, 176
measures derived from 12, 17 in Cape York seminar transcript
Specially Protected Areas Regional 181–5, 186, 188, 192
Activity Centre (SPA/RAC) 228 Stott, P. 301
species extinctions 107, 137 Strict Nature Reserves 33, 34, 68
strong sustainability 402 swimming crab (Portunus pelagicus)

structural adjustment programmes 249
401–2 Switzerland 397
structural diversity 92 Sykes, J.M. 90
structural realism 200 symmetric equilibrium 316
Stuart, S.N. 17 Syria 221
Sudan 85
sugarcane 414, 418, 423, 425, 426 Taffs, K.H. 11, 12, 16
Sugden, R. 125 Taghi Farvar, M. 126
Sullivan, S. 297, 301, 303 Taiwan 375
sulphur dioxide (SO2) emissions 211, Takekawa, J.Y. 87, 104
363, 367–8, 369, 370, 372, 376, TAMARIN 41
377–9, 382, 384, 385, 387, 388–9, Tang, S.Y. 153
390–91, 393, 395, 398, 399, 418 TANU (Tanzania African National
Summers, R. 367–8, 373, 379, 381, 382, Union) 300
383, 390, 398 Tanzania
Sumner, D.R. 103 climatic and vegetative
Sun, D. 93 characteristics of 284, 285–6,
sunflower oil 419, 421 288, 294, 332
supply gap 171–2 environmental management in
Supriatna, J. 22 colonial-influenced approaches to
Suri, V. 365, 385–6, 393 6, 293–4, 295, 299–300
suri traps 247 ecological and social paradigms
survivalism discourse 175, 185, 186 and 5–6, 295–304
suspended particulate matter (SPM) economic valuation of different
363, 367–8, 369, 370, 372, 376, forms of land use 331–56
382, 387, 389, 390, 391, 393, 395, game theory applied to CPR
396, 399 management 321–8
sustainability discourse 174, 175, 185, Integrated Conservation and
186 Development Project (ICDP)
sustainable development in Uluguru Mountains 45–56,
definition of 282 67, 68–9
environmental management for 5–6, Participatory Environment
282–304 Management Programme 35
poverty reduction and 282, 303–4 tax system and 43
principles of 408 fish consumption in 270
sustainable end points for ICDPs 42–3, historical development of societies
55–6, 61–2, 66 in 289–95
sustainable extraction 39 income sources in 270
Svendsen, J.O. 48 itinerant fishers from 272
Swanson, T. 334 Tanzania African National Union
Swatuk, L.A. 215 (TANU) 300
Sweden taxes 43, 105, 401, 413
agri-environmental programme in Taylor, P.D. 91
148 Taylor, P.J. 107
biodiversity in 88, 91, 92 technology shifts, effect on income–
biofuel usage in 412–13, 414, 422 pollution relationship 366,
economic growth and pollution in 369–70, 385, 394–5
394, 397 technology uptake 63–4, 227
Swift, D.M. 298 Telleria, J.L. 82, 83, 92
Index 459
temperate forests 14 and selection of appropriate

Terborgh, J. 92 governance structure 135,
terminal nodes 312 138–9, 141–6, 159
termites 17–20 transactions costs approach to
terrestrial vertebrate species 12, 17, institutional analysis 119
20, 48 transboundary effects 362
Thailand 380 transparency 146, 208, 214
theft 261–2 transport
Thomas, C.D. 13 biofuel use in 409, 411–27
Thomas, R.P. 131 livestock used for 335, 339, 340
threats to natural resources 35–6, Trapnell, C.G. 285
38–41, 44, 45, 47, 56, 66, 143, 200, travel cost method (TCM) 100
273–5 tree planting 39, 171, 301
three Cs approach to measuring tree species 79, 88, 89
effectiveness of international trespassing game 324–8
institutions 206–7 triage 11
Thucydides tropes 192
History of the Peloponnesian War trophic-level analysis 78
199 tropical rain forests
tiger beetles 17–20 biofuel crop production and 7, 410,
tigers 41 415, 423–4, 425, 426, 427
Tilia 88 burning of 176–7
Tilman, D. 80, 81 see also Amazonian rain forests
timber production/extraction 42, 53, trust 150, 151, 152, 261, 303
54–5, 100, 101, 102, 103–4, 125, trust funds 43, 55, 62, 69, 231
148, 275, 331, 335 Tscharntke, T. 91
Tinbergen, L. 87 tsetse fly 286, 288, 293–4, 295
Tole, L. 371 Tsongwain, D.V. 64
Toledo, V.M. 337 Tuan, N.A. 380–81, 389
Torrano, L. 84 Tubbs, C.R. 88
Torras, M. 365, 381–2 Tucker, A.W. 329
total coliform 368, 384 Tucker, G.M. 83, 88
total economic value (TEV) 98, 99 tulip shells (Pleuroploca trapezium)
tourism 39, 42, 43, 45, 184, 244, 269, 248, 268
270, 276, 303, 333 Tumucumaque National Park 16
toxic emissions 375 Tunis 225, 228
traffic volume 372 Tunisia 221, 228
tragedy of the commons 124, 125–6, Turkey 221
128, 132, 160 Turley, D. 419, 420, 421
transaction cost economics (TCE) 135 Turner, R.K. 99
transactions costs turtle (Chelonidae) 20, 249
and collective action 138–41
definitions of 135, 140 Uganda 35
and effectiveness of international Ujamaa villages 294
environmental regimes 213 UK Department for International
empirical studies of 146–50, 160–61 Development (DFID) 242, 287,
problems associated with 140–41 288
group heterogeneity and 150 Ukraine 423
insights from economic literature Uluguru Mountains, Tanzania 45–56,
134–8 67, 68–9
uncertainty 107, 108, 136–7, 138, 139, United Nations Environment

141, 159, 171, 174, 184, 188, 193, Programme (UNEP) 75, 76, 217,
220, 303, 312, 425 218, 224, 227, 228, 229, 230, 231,
radical uncertainty 171, 193 232, 287, 371, 381
Underdal, A. 206, 209 United Nations Framework
underinvestment 130, 160 Convention on Climate Change
UNDP (United Nations Development 202, 408
Programme) 32, 64, 244, 382 United Nations Industrial
UNEP (United Nations Environment Development Organization
Programme) 75, 76, 217, 218, 224, (UNIDO) 227
227, 228, 229, 230, 231, 232, 287, United Nations International
371, 381 Comparison Programme 386
UNEP/MAP 221, 230 United Nations Millennium
see also Mediterranean Action Plan Development Goals 7, 282, 409
(MAP) United Nations Statistical Yearbook
UNEP Regional Seas Programme 217 390
UNESCO (United Nations United States
Educational, Scientific and biodiversity conservation priorities
Cultural Organization) 227 in 14
UNIDO (United Nations Industrial biofuel use in 418, 421
Development Organization) 227 economic growth and pollution in
unique Nash equilibrium 316, 317, 319 375, 376, 380, 393, 394–5, 397
United Kingdom Kyoto Protocol not ratified by
biodiversity in managed landscapes 408–9
in 83, 87, 88 valuing biodiversity in 100, 102,
economic growth and pollution in 104
394, 395, 397 Unruh, G.C. 396–8
ecosystem disturbance and Uphoff, N. 150
productivity in 90 Uri, N.D. 336
Environmentally Sensitive Area US Environmental Protection Agency
(ESA) scheme in 105–6 (EPA) 376, 395
feedstock production for biofuels in US–Mexico Free Trade Agreement
415, 420 367
sites of special scientific interest use values 98, 100
(SSSI) in 107–8 direct use value 99–100, 335
valuing biodiversity in 100, 101 measurement of 101–2, 336–40,
United Nations 203, 374, 376, 384 342–3, 345–50, 355
United Nations Conference on indirect use value 100, 331–2, 335–6
Environment and Development measurement of 102–4, 340,
‘Earth Summit’ (Rio, 1992) 218, 343–4, 351–5
221 utility maximization 119
United Nations Conference on the
Human Environment (Stockholm, Valderrábano, J. 84
1972) 199, 359 Vallega, A. 231
United Nations Development valuation of biodiversity 98–100
Programme (UNDP) 32, 64, 244, indirect use values and ecosystem
382 function 102–4
United Nations Educational, Scientific measuring biodiversity values
and Cultural Organization 100–101
(UNESCO) 227 value of species and habitats 101–2
Index 461
valuation of different forms of land use Wakeford, R.C. 266

331–56 Walker, B.H. 78, 87, 90, 94, 103
ecosystem values 334–6 Walker, T.S. 103
importance of valuation of Wallace, M.B. 129, 130, 160
ecosystems 331–2 Walter, A. 284
past valuation studies 336–41 Waltz, K.N. 200
production function approach to Wamer, N.O. 92
341–5 Wang, N. 122, 123
direct use values 339–40, 342–3, Warming, J. 124
345–50 Warren, A. 297, 298
indirect use values 340, 343–4, Wasser, S.K. 48
351–5 water payments 43, 55
limitations to use of 344–5 water pollution
Tanzanian land-use issues 332–4 biofuel feedstock production and
Van der Elst, R. 242 419, 420
van Dorp, D. 88 economic growth and 368, 369,
Van Hees, A.F.M. 89–90 375–6, 381–2, 383–4, 387, 389,
Van Keulen, H. 284, 287 390, 391
van Rensburg, T.M. 101 see also river pollution
Vane-Wright, R.I. 77 water supplies 42, 48, 52, 54, 55, 59, 64,
variegated emperor fish (Lethrinus 65, 295, 336
variegatus) 247, 250, 251 clean 42, 360, 368, 369, 380, 382, 383
Varughese, J. 130, 150, 156, 160 see also rainfall
Vätn, A. 146 Watt, P. 172
Veblen, T.T. 89 Wattle Hills residents’ policy stance on
Veech, J.A. 13, 14 burning 180, 181, 182
vehicle emissions 379, 383, 392, 409, discourses adopted in support of
417–18, 419, 420, 426, 427 182, 184, 185, 186, 187, 190–91
Velded, T. 150, 156–7 WCED (World Commission on
verification of compliance 213–14, Environment and Development)
229–30 95, 104, 282
vertical integration 138 Weale, A. 214
Victor, D.G. 213–14 Weber, G.E. 89
Vienna Convention (1985) 209, 376 Wehrmeyer, W. 31
Vietnam 380 Weisbrod, B. 98
village elders 262 Weitzman, M.L. 77
Village Forest Reserves 49, 50, 53, 65 Welch, D. 90
villagization 293–4, 295 Wells, M. 31, 32, 43, 67
Vincent, J.R. 364, 389–90 wells 148–9
volatile organic compounds (VOCs) Wenner, C.G. 301
387, 395 Werner, U. 14
von Moltke, K. 207–8, 214 Weslien, J. 91
Von Neumann, J. 310 Wesseler, J. 421
vulnerability West, H.G. 243, 262
measures of 12–13, 17 Westoby, M. 91
relative to irreplaceability 10–11, Wettestad, J. 206
14–16, 18–19 Wheeler, W.B. 171
Whelan, C.J. 87
Wagenaar, H. 171, 172 Whigham, D.F. 88, 92
Wainwright, C. 31 White, A.T. 241
White, F. 285, 286, 292 World Commission on Environment

white stork 83 and Development (WCED) 95,
Whiteman, A. 100 104, 282
Whittaker, R.H. 82 World Health Organization (WHO)
Whittaker, R.J. 14, 22 227, 367, 383
Whitten, T. 8 World Petroleum Institute 377
Whittington, M.W. 242 World Resources Institute (WRI) 75,
whooping crane (Grus americana) 102 372, 378, 381, 383
Wikramanayake, E.D. 22 World Summit on Sustainable
Wilcox, B.A. 77 Development (WSSD) 32
Wildlife Conservation Society of World Trade Organization 426–7
Tanzania 48, 49, 51, 69 World Wildlife Fund (WWF) 9, 11, 13,
Williamson, O.E. 135, 136, 141, 147, 16, 55
161–2 WRI (World Resources Institute) 75,
willingness to accept (WTA) 372, 378, 381, 383
compensation 98, 100 WSSD (World Summit on Sustainable
willingness to pay (WTP) 98, 100, 101, Development) 32
151–2 WWF (World Wildlife Fund) 9, 11, 13,
Willis, K.G. 100, 101, 105 16, 55
Willson, M.F. 92 WWF-World Bank Management
Wilson, E.O. 75, 91, 92, 331 Effectiveness Tracking Tool 56
Wilson, J.D.K. 272 Wyckoff-Baird, B. 31
Wilson, K.A. 12, 13 Wynter, P. 270
Wilson, M.V. 92
Wilson, W.L. 88 Xai-Xai Integrated Coastal Area
Wily, L.A. 42, 68 Management project 271
wind 94
witch-doctors (curandeiros) 243, 262 Yellowstone National Park 95
Wittmer, H. 136, 137, 140, 141, 142, Yohe, G. 13
143, 144, 145, 146, 147 Young, O.R. 202, 204, 207, 210–11
women, fishing by 245, 248, 252, 254,
255, 266–8, 269, 270, 278 Zak, P.J. 150
wood warbler (Phylloscopus sibilatrix) Zambesian region 285–6, 288, 298
83 Zambezi river 336
World Bank 55, 373, 401 Zanzibar (Unguja) Island 270
World Bank Tables 378, 396 Zimbabwe 43, 87, 339–40
World Development Report 381, 385 Zohary, M. 82

A Handbook of Environmental Management PDF

Uploaded by

Copyright:

Available Formats

A Handbook of Environmental Management PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

A Handbook of Environmental Management PDF

Uploaded by

Copyright:

Available Formats

What are the main topics covered in the document?

What are the main topics covered in the document?

What are some of the approaches discussed for valuing different forms of land use?

What are some of the approaches discussed for valuing different forms of land use?

A HANDBOOK OF ENVIRONMENTAL

All rights reserved. No part of this publication may be reproduced, stored

Edward Elgar Publishing, Inc.

A catalogue record for this book

Library of Congress Control Number: 2009940631

ISBN 978 1 84064 207 0 (cased)

Printed and bound by MPG Books Group, UK

List of contributors vii

11 Economic valuation of different forms of land-use in semi-

Bhim Adhikari, Institute for Water, Environment and Health, United

David G. Ockwell, Department of Geography, University of Sussex, UK

Environmental management has risen from being the task of technical

An implicit and almost universal assumption of discussions published in pro-

The point of Hardin’s paper is that technical solutions alone cannot

application of different social science techniques in combination with

as wilderness were formerly managed for livestock, hunting or exten-

time it clearly highlights the insights that environmental managers might

based on ‘non-equilibrium theory’ lends itself to a new perspective that

are predicted to be enormous, changing the whole ecology of the planet. A

Human actions are causing a biodiversity crisis, with species extinction

Crisis ecoregions (Hoekstra et al., 2005)

which the percentage conversion is ≥2 times the percentage

Biodiversity hotspots (Myers et al., 2000)

Centres of plant diversity (WWF and IUCN 1994–97)

≥10% are endemic either to the site or the region; or islands

Megadiversity countries (Mittermeier et al., 1997)

Frontier forests (Bryant et al., 1997)

Last of the wild (Sanderson et al., 2002)

Figure 2.1 Global biodiversity conservation prioritizations

Figure 2.2 Growth of attention to global biodiversity conservation

al., 2000). Although attempts have been made to summarize conservation

Global prioritization in context

A proactive reactive B proactive reactive

Source: Revised version of figure originally published in Brooks et al. (2006).

Figure 2.3 Global biodiversity conservation priority templates placed

provides a means of scheduling responses within such an overall frame-

Mittermeier et al., 2003b; Hoekstra et al., 2005). Species–area relation-

Source: Revised version of figure originally published in Brooks et al. (2006).

Figure 2.4 Mapping the overlay of approaches prioritizing reactive and

despite this, a noticeable pattern emerges from the overlay of different

Impact of global prioritization

mechanism addressing biodiversity conservation, is currently exploring a

Challenges facing global prioritization

Source: Revised version of figure originally published in Brooks et al. (2006).

Figure 2.5 Incorporating primary biodiversity data in global conservation

much higher levels of congruence (Mittermeier et al., 2004). Similarly,

Aquatic systems feature poorly in existing conservation templates,

areas of greatest importance in generating biodiversity are those of long-

From global to local priorities

important bird areas (BirdLife International, 2004). There is an obvious

Dinerstein, E. and E.D. Wikramanayake (1993), ‘Beyond “hotspots”: how to prioritize

Mittermeier, R.A., P. Robles Gil, M. Hoffmann, J. Pilgrim, T. Brooks, C.G. Mittermeier,

Where have ICDPs come from?

wildlife and forestry resources (Neumann, 1996; Schrijver, 1997). Areas of

because it always contains a core area protected for conservation reasons