Strategic Management Journal, Vol.

12, 95-1 I7 ( I 991)

/-----

TOWARDS A DYNAMIC THEORY OF STRATEGY

MICHAEL E. PORTER

T--

Graduate School of Business Administration, Harvard Universitv, Boston, Massachusetts, U.S.A.

This paper reviews the progress of the strategy field towards developing a truly dynamic
theory of strategy. It separates the theory of strategy into the causes of superior performance
at a given period in time (termed the cross-sectional problem) and the dynamic process by
which competitive positions are created (termed the longitudinal problem). The crosssectional problem is logically prior to a consideration of dynamics, and better understood.
The paper then reviews three promising streams of research that address the longitudinal
problem. These still fall short of exposing the true origins of competitive success. One
important category of these origins, the local environment in which a firm is based, is
described. Many questions remain unanswered, however, and the paper concludes with
challenges f o r future research.

INTRODUCTION
The reason why firms succeed or fail is perhaps
the central question in strategy. It has preoccupied
the strategy field since its inception four decades
ago. The causes of firm success or failure
encompass all the other questions that have been
raised in this collection of essays. It is inextricably
bound up in questions such as why firms differ,
how they behave, how they choose strategies,
and how they are managed. While much of the
work in the field has been implicitly domestic, it
has become increasingly apparent that any search
for the causes of firm success must confront the
reality of international competition, and the
striking differences in the performance of firms
in a given industry based in different nations.
Yet, the question of why firms succeed or fail
raises a still broader question. Any effort to
understand success must rest on an underlying

Key words: Strategy formulation, dynamic theories,

competitiveness
0143-2095/9 1/100095-23$11 .SO

0 1991 by John Wiley & Sons, Ltd.

theory of the firm and an associated theory of

strategy. While there has been considerable
progress in developing frameworks that explain
differing competitive success at any given point
in time, our understanding of the dynamic
processes by which firms perceive and ultimately
attain superior market positions is far less
developed. Worse yet, some recent research has
tended to fragment or dichotomize the important
parts of the problem rather than integrate them,
as I will discuss later.
My purpose in this essay is to sketch the
outlines of a dynamic theory of strategy. Drawing
on recent research, some parts of the outline can
be filled in. Many unanswered questions remain,
however, and I will try to highlight some of the
most important of them.
As a starting point for building a dynamic
theory of strategy, we must step back from
specific hypotheses or models and look broadly
at the literature in both strategy and economics.
I will begin by describing the traditional rationale
for company success that emerged in the early
literature on strategy. This reflected an orien-

96

M . E. Porter

tation of the strategy field that has differed

in important respects from that which has
characterized most research in economics, arguably the discipline with the most obvious connection to strategy. The strategy fields traditional
answer to why firms succeed or fail was also
based on a set of largely implicit, but crucial
assumptions about the nature of firms and the
environment in which they operate.
Although these assumptions grew out of a deep
understanding of practice, they raise profound
challenges for a theory of strategy. I will outline
some of the most important challenges and the
trade-offs they raise in both theory and empirical
testing. Taking these challenges as a starting
point, I will then describe my own answers to
the causes of superior firm performance at a
given point in time, which can be framed as a
chain of causality. This problem, which I term
the cross-sectional problem, is logically prior
to a consideration of dynamics and better
understood. A body of theory which links firm
characteristics to market outcomes must provide
the foundation for any fully dynamic theory of
strategy. Otherwise, dynamic processes that result
in superior performance cannot be discriminated
from those that create market positions or
company skills that are worthless.
I will then move to the dynamic process by
which positions are created, which I term
the longitudinal problem. To understand the
dynamics of strategy, we must move further back
in the causality chain. I will explore three recent
streams of research that begin to address i t : game
theoretic models, models of commitment under
uncertainty, and the so-called resource-based
view of the firm. While illuminating important
characteristics of the dynamic processes by which
advantage is created and sustained, however, this
research still falls short of exposing the true
origins of advantage, and I will discuss the
reasons why. One important category of these
origins, that has emerged from my recent work,
is the nature of the local environment in
which the firm is based. We observe striking
concentrations of successful firms in a particular
industry in particular locations, which suggests
that something about these locations is fundamental to creating and sustaining advantage. I will
summarize some of my findings about these
issues. Many questions remain unanswered in
our search for a dynamic theory of strategy,

however, and this essay will conclude with some

challenges for future research.
Determinants of firm success: The early answers

Any discussion of the determinants of firm

success must begin with a clear definition of what
success means. For purposes of this essay, I will
assume that firm success is manifested in attaining
a competitive position or series of competitive
positions that lead to superior and sustainable
financial performance. Competitive position is
measured, in this context, relative to the worlds
best rivals. Financial success derived from government intervention or from the closing of markets
is excluded. A successful firm may spend some
of the fruits of its competitive position on meeting
social objectives or enjoying slack. Why a firm
might do this, however, is treated as a separate
question.
To explain firm success, the early literature on
strategy defined three essential conditions. The
first is that a company develop and implement
an internally consistent set of goals and functional
policies that collectively defined its position in
the market. Strategy is seen as a way of
integrating the activities of the diverse functional
departments within a firm, including marketing,
production, research and development, procurement, finance, and the like. An explicit and
mutually reinforcing set of goals and functional
policies is needed to counter the centrifugal forces
that lead functional departments in separate
directions. Strategy, in modern language, is a
solution to the agency problem that arises because
senior management cannot participate in or
monitor all decisions and directly ensure the
consistency of the myriad of individual actions
and choices that make up a firms ongoing
activities. If an overarching strategy is well
understood throughout the organization, many
actions are obviously ruled out and individuals
can devise their own ways to contribute to the
strategy that management would be hard pressed
to replicate.
The second condition for success is that this
internally consistent set of goals and policies
aligns the firms strengths and weaknesses with

Sec Learned e t al. (1905).

See also Andrews (1971).

In the absence of a strategy. the narrow motivations and
logistics of each functional area will guide bchavior.

Towards a Dynamic Theory of Strategy

the external (industry) opportunities and threats.
Strategy is the act of aligning a company and its
environment. That environment, as well as the
firms own capabilities, are subject to change.
Thus, the task of strategy is to maintain a
dynamic, not a static balance.
The third condition for success is that a firms
strategy be centrally concerned with the creation
and exploitation of its so-called distinctive
competences. These are the unique strengths a
firm possesses, which are seen as central to
competitive success. The recent interest in the
notion of firm resources or competences is
interesting in light of this heritage.4 I will return
to this stream of work later.
The early strategy literature contained only
broad principles governing firm success. It is
instructive to understand why these authors,
coming as they did from a heritage that stressed
the administrative point of view and the study
of in-depth cases, chose to approach the question
in this way. There were two principal reasons.
The first was that their orientation, and that of
many in the strategy field, was to inform business
practice. A theory that sought to explain part of
a phenomena, but which left out important
elements that precluded the offering of credible
guidance for individual companies, was seen as
inadequate to the task.
A second reason for the early formulation was
the recognition, indeed the preoccupation, with
the fact that competition was complex and highly
situation-specific. The early scholars in the
strategy field, especially those at Harvard, recognized that firms were composed of numerous
functions and subfunctions, and that many diverse
aspects of a firm and its environment could be
important to success in particular cases. Indeed,
it was the act of achieving consistency of action
in the many parts of the firm that was seen as
crucial to competitive success. Scholars such as
Andrews saw each company as unique, with its
own history, personality, capabilities, and set of
current policies. Every industry was also unique,
with its own circumstances and critical success
factors. Finally, every period of time was seen
as unique, because both companies and their
environment were in a state of constant change.
~

This notion

is due originally to Selznick (1957).

See, for example, Wernerfelt (1984) and Prahalad and
Hamel (1990).

97

Yet firms were seen as possessing considerable

ability to build on their strengths and overcome
their weaknesses, latitude in influencing or
altering their environment, and the ability to
influence change over time, not merely respond
to it. Indeed, the recognition that industry
structure and other exogenous conditions affect
performance and constrain choices had to await
further work.
The challenges for a theory of strategy

The view of the world that guided the early

efforts to formulate a theory of strategy raises
profound challenges for research. The complexity,
situation specificity, and changing nature of the
firm and its environment strains conventional
approaches to theory building and hypothesis
testing. indeed, the early research offered no
theory for examining the firm and its competitive
environment at all; instead strategy formulation
took place through applying the broad principles
of consistency and fit to individual case studies.
Four principal issues emerge from the nature
of actual economic competition as one contemplates a theory of strategy:
Approach to theory building
First, there is a fundamental question about the
approach to theory building that will most
advance both knowledge and practice. The broad
alternatives are represented in Figure 1.

MODELS

FRAMEWORKS

Figure 1. Approaches to theory building

On the one hand, one might approach the task

of developing a theory of strategy by creating a
wide range of situation-specific but rigorous (read
mathematical) models of limited complexity.
Each model abstracts the complexity of competition to isolate a few key variables whose
interactions are examined in depth. The normative significance of each model depends on the
fit between its assumptions and reality. No one
model embodies or even approaches embodying

98

M . E. Porter

all t h e variables of interest, and hence the

In frameworks, the equilibrium concept is
applicability of any models findings are almost imprecise. My own frameworks embody the
inevitably restricted to a small subgroup of firms notion of optimization, but no equilibrium in
or industries whose characteristics fit the models the normal sense of the word. Instead there is
assumptions.
a continually evolving environment in which a
This approach to theory building has been perpetual competitive interaction between rivals
characteristic of economics in the last few takes place. In addition, all the interactions
decades. It has spawned a wide array of among the many variables in the frameworks
interesting models in both industrial organization cannot be rigorously drawn. The frameworks,
and trade theory. These models provide clear however, seek to help the analyst to better
conclusions, but it is well known that they are t h i n k through the problem by understanding
highly sensitive to the assumptions underlying the firm and its environment and defining
them and to the concept of equilibrium that is and selecting among the strategic alternatives
employed. Another problem with this approach available, no matter what the industry and
is that it is hard to integrate the many models starting position.
These two approaches to theory building are
into a general framework for approaching any
situation, or even to make the findings of the not mutually exclusive. Indeed, they should
various models consistent. While few economists create a constructive tension with each other.
would assert that this body of research in and of Models are particularly valuable in ensuring
itself provides detailed advice for companies, logical consistency and exploring the subtle
these models, at their best, provide insights into interactions involving a limited number of varicomplex situations that are hard to understand ables. Models should challenge the variables
without them, which can inform the analysis of included in frameworks and assertions about
their link to outcomes. Frameworks, in turn,
a particular companys situation.
Given the goal of informing practice. the style should challenge models by highlighting omitted
of research in the strategy field, including my variables, the diversity of competitive situations,
own, has involved a very different approach. To the range of actual strategy choices, and the
make progress, it was necessary to go beyond extent to which important parameters are not
the broad principles in the early work and fixed but continually in flux. The need to inform
provide more structured and precise tools for practice has demanded that strategy researchers
understanding a firms competitive environment such as myself pursue the building of frameworks
and its relative position. Instead of models, rather than restrict research only to theories that
however, the approach was to build frameworks. can be formally modelled. As long as the building
A framework, such as the competitive forces of frameworks is based on in-depth empirical
approach to analyzing industry structure. research, it has the potential to not only inform
encompasses many variables and seeks to capture practice but to push the development of more
much of the complexity of actual competition. rigorous theory.
Frameworks identify the relevant variables and
the questions which the user must answer in
Chaiti of causalify
order to develop conclusions tailored to a
particular industry and company. In this sense, A second fundamental issue in creating a theory
they can be seen as almost expert systems. The of strategy is where to focus the chain of causality.
theory embodied in frameworks is contained in A stylized example will illustrate. We might
the choice of included variables, the way variables observe a successful firm and find that its
are organized, the interactions among the vari- profitability is due to a low relative cost position
ables, and the way in which alternative patterns compared to its rivals. But the firms cost position
of variables and company choices affect outcomes. is an outcome and not a cause. The question

Interestingly, the earlier work in industrial economics. in

the MasoniBain tradition. was much closer t o strategy
research in its effort to capture complexity.
S e e examples such as Porter (1985) and Ghemawat (1991).

Frameworks can also he challenged because their complexity

makes i t difficult to falsify arguments. Yet ascribing this
property to models is also problematic if they omit important
vii ria bles
I

Towurds u Dynamic Theory of Strategy

becomes: Why was the firm able to attain this
cost position? Some typical answers might be
that it is reaping economies of scale, or has
moved aggressively down the learning curve. But
again, the question becomes why? Some possible
answers might include entering the industry early,
or the firms ability to organize itself particularly
well for cost reduction. Once again, however,
the question becomes why? And we could
continue moving along such a chain of causality
even further.
Another way of framing the same set of issues
is as the problem of drawing the boundary
between exogenous and endogenous variables.
Should the environment be taken as given or
not? Is the firms scale an outcome or a cause?
And so on. The literature in both strategy and
economics addresses many different points in this
chain of causality. Indeed, many differences are
less conflicts than theory positioned at different
points in the chain, as we will see later.
Any theory of strategy must grapple with how
far back in the chain of causality to go. The
answer may well be different for different
purposes. A theory that aims very early in the
chain may be intractable or lack operationality.
Also, aspects of the firm that are variable in the
long run may be fixed or sticky in the short run.
Conversely, a theory oriented later in the chain
may be overly limiting and miss important
possibilities.

Time horizon

A third challenge for theory is the time period

over which to measure and understand competitive success. Should we be building theories for
explaining success over two or three years, over
decades, or over centuries? Clearly, the likelihood
of significant environmental change will differ,
as will the exogenous and endogenous variables.
A theory that aims at explaining success over 50
years will focus on very different variables, almost
inevitably more internal ones, than a theory that
addresses success over one or two decades. This
is because industry and competitive conditions
are likely to be wholly different over a half
century, placing greater emphasis on a firms
ability to transform itself. Time period relates
closely to position in the chain of causality. Over
long periods, theories aimed earlier in the chain
would seem more appropriate.

99

Empiricul testing
A final important issue is how to test theories
of strategy empirically. Empirical testing is vital
both for frameworks and models. Testing of
models is difficult given the need to match their
assumptions. Given the myriad of relevant
variables in frameworks and the complex interactions among them over time, rigorous statistical
testing of frameworks is also difficult, to say the
least. In my own research, I pursued crosssectional econometric studies in the 1970s but
ultimately gave up as t h e complexity of the
frameworks I was developing ran ahead of the
available cross-sectional data. I was forced to
turn to large numbers of in-depth case studies
to identify significant variables, explore the
relationships among them, and cope with industry
and firm specificity in strategy choices.
The need for more and better empirical testing
will be a chronic issue in dealing with this subject.
Academic journals have traditionally not accepted
or encouraged the deep examination of case
studies, but the nature of strategy requires it.
The greater use of case studies in both books
and articles will be necessary for real progress
at this stage in the fields development.

TOWARDS A THEORY OF STRATEGY

To explain the competitive success of firms,
we need a theory of strategy which links
environmental circumstances and firm behavior
to market outcomes. My own research would
suggest a chain of causality for doing so, outlined
in Figure 2.
The basic unit of analysis in a theory of strategy
must ultimately be a strategically distinct business
or industry. While firms can redeploy or share
resources, activities, and skills across different
businesses, the competitive value of such actions
can only be measured in terms of some set of
rivals delivering a discrete product or service to
some set of buyers. Meaningful approaches to
corporate-level strategy for diversified firms must
grow out of a deep understanding of how
companies prosper in individual businesses, and
the role of the corporate office and other sister
business units in the process.
At the broadest level, firm success is a function
of two areas: the attractiveness of the industry

100

M . E. Porter

Cross-Sectional
Success

Attractive industry
Structure
(5 Forces)

Attractive Relative
Position

Sustainable
Competitive
Advantage

Drivers

Structural determinants of
ditferences in the cost or
buyer value of activities or
CI~OUDS of activities

Longitudinal

Figure 2.

The determinants of success in distinct businesses

in which the firm competes and its relative

position in that industry. Firm profitability can
be decomposed into an industry effect and a
positioning effect. Some firm successes come
almost wholly from the industry i n which they
compete; most of their rivals are successful,
too! The distinction between industry structure
and relative position is important because,
among other things, the firm can choose
strategies that will improve one while harming
the other. Firms actions, by triggering imitation, can positively or negatively influence
the structure of an industry without leading to
competitive advantage. Ideally. however, a
firms actions trigger responses by rivals which
improve industry structure but simultaneously
allow the firm to gain competitive advantage
because rivals ability to imitate the chosen
mode of competition is incomplete.

Industry structure

I have presented a framework for diagnosing

industry structure, built around five competitive
forces that erode long-term industry average
profitability (see Figure 3). This framework has
been explored, contributed to, and tested by
many others. The industry structure framework
can be applied at the level of the industry, the
strategic group (or group of firms with similar
strategies) or even the individual firm. Its ultimate
function is to explain the sustainabifiry of profits
against bargaining and against direct and indirect
competition. Profit differences vis-u-vis direct
rivals, though, depend on positioning.
Industry structure is partly exogenous, and
partly subject to influence by firm actions. Hence
structure and firm position ultimately interrelate,
which makes separating them a simplification

Towards a Dynamic Theory of Straregy

101

v
THREAT OF

NEW ENTRANTS

PRODUCTS OR SERVICES

Figure 3. Five forces: Summary of key drivers

The five forces (p. 4): From Competitive strategy: Techniques f o r Analyzing Industries and Competitors by
Michael E. Porter. Copyright 01980 by The Free Press, a Division of Macmillan, Inc. Reprinted by permission
of the publisher.

though a useful one for analytical purposes. The

firms scope for influencing industry structure,
and ways of modeling it, are a fruitful area for
research. My focus here, however, is on relative
position because this is where many of the most
interesting questions for a dynamic theory of
strategy lie.
Relative position

Holding industry structure constant, a successful

firm is one with an attractive relative position.
An attractive position is, of course, an outcome
and not a cause. The question becomes why, or
how did the attractive position arise? The answer
must be that the firm possesses a sustainable
competitive advantage vis-u-vis its rivals. To
understand competitive advantage, however, we
must decompose it. Competitive advantages can
be divided into two basic types: lower cost than
rivals, or the ability to differentiate and command
a premium price that exceeds the extra cost of
doing so. Any superior performing firm has

achieved one type of advantage, the other, or

both. To say it another way, superior profitability
can only logically arise from commanding a
higher price than rivals or enjoying lower costs
(including, at a lower level in the causality chain,
asset costs).
Competitive advantage cannot be examined
independently of competitive scope. Scope
encompasses a number of dimensions including
the array of product and buyer segments served,
the geographic locations in which the firm
competes, its degree of vertical integration, and
the extent of related businesses in which the
firm has a coordinated strategy. Competitive
advantage is attained within some scope, and the
choice of scope is a central one in strategy. Scope
choices can also influence industry structure.
These principles make it clear that the essence
of strategy is choice. There is no one way to
* A firm that can command higher volume at a given price
takes its superior profitahility in the form of lower cost
provided costs are scale sensitive.

102

M . E. Porter

centrally between activities that directly produce,

market, and deliver the product and those that
create or source inputs or factors (including
planning and management) required to do so.
Support activities, then, are integral to the
process by which assets internal to the firm are
acquired and accumulated.
Discrete activities are part of an interdependent
system in which the cost or effectiveness of one
activity can be affected by the way others are
performed. I term these linkages. The cost of
after-sale service, for example, is influenced by
how product design, inspection, and installation
are performed. Such linkages can extend outside
the firm to encompass the activities of suppliers,
Activities
channels, and buyers. The concept of linkages
If an attractive relative position results from begins to operationalize the notion of internal
possessing competitive advantage within some consistency.
Activities involve human resources, purchased
scope, the question once again becomes why
does that happen? In order to address it, we inputs, and a technology for performing them,
must decompose cost, differentiation, and scope. broadly defined to include organizational rouThis requires a theory which provides an elemen- tines. Activities also use and create information.
tal look at what firms do. My own approach to Performing an activity requires tangible and
such a theory, and to the sources of competitive intangible assets that are infernal to the firm,
advantage, centers around acfiviries. (Porter, such as physical and often financial assets (e.g.
1985). A firm is a collection of discrete, but working capital) as well as intangible assets
interrelated economic activities such as products embodied in human resources and technology.
being assembled, salespeople making sales visits, Performing an activity, or a group of linked
and orders being processed. A firms strategy activities. also creates assets in the form of skills,
defines its configuration of activities and how organizational routines, and knowledge. While
they interrelate. Competitive advantage results the tangible assets normally depreciate, the
from a firms ability to perform the required intangible assets involved in performing activities
activities at a collectively lower cost than rivals, can cumulate over time (provided the environor perform some activities in unique ways that ment remains relatively stable). These become
create buyer value and hence allow the firm to an important part of corporate balance sheets,
command a premium price. The required mix as many writers have stressed.
Performing activities can also create assets
and configuration of activities, in t u r n , is altered
exterrzal to the firm. Some are tangible assets
by competitive scope.
The basic unit of competitive advantage, such as contracts. Most, however, are intangible
then, is the discrete activity. The economics of assets such as brand images, relationships, and
performing discrete activities determines a firms networks. These external assets then feed back
relative cost, not attributes of the firm as a to influence the cost or effectiveness of performing
whole. Similarly, it is discrete activities that activities on an ongoing basis. A strong brand
create buyer value and hence differentiation.
reputation because of cumulative past advertising,
The activities in a firm can be schematically for example, can lower the cost of current
arrayed in what I term the value chain and the advertising or make a given rate of spending
value system (see Figure 4). The term value more effective. Without reinvestment, however,
refers to customer value. from which the potential both the external and internal intangible assets
profit ultimately derives. A firms strategy is
manifested in the way in which it configures and
Scc Porter and Millar (1985).
links the many activities in its value chain relative 1 0 see. for exarnplc. Itnrni (1987) and Baldwin and Clark
to competitors. The value chain distinguishes (1991).

position within an industry, but many positions

involving different choices of the type of advantage sought and the scope of the advantage.
Several positions can be attractive in absolute
terms, and a variety of positions may be relatively
the most attractive depending on the firms
starting position. Choice is essential, however.
because there are logical inconsistencies in
pursuing several types of advantage or different
scopes simultaneously. Also, the firm must stake
out a distinct position from its rivals. lmitation
almost ensures a lack of competitive advantage
and hence mediocre performance.

Towards a Dynamic Theory of Strategy

103

\ \

FIRM INFRASTRUCTURE

HUMAN RESOURCE MANL~GEMENT

TECHLOLOGY DEVELO~MENT

VALUE
CHAIN

INBOUND
LOGISTICS

PROCUREMENT

l-lOPERATIONS

OUTBOUND
LOGlSTlCS

MARKETING
AND SALES

VALUE
SYSTEM
CHAINS
UPSTREAM
VALUE

THE
FIRMS
VALUE
CHAIN

CHAINS

DOWNSTREAM
VALUE

Figure 4. Value chain and value system

The Value Chain (p, 37) and The Value System (p. 35): From Competitive Advantage: Creating atid
Sustaining Superior Performance by Michael E. Porter. Copyright @ 1985 by Michael E. Porter. Reprinted
by permission of The Free Press, a Division of Macmillan, Inc.

attached to activities or groups of activities

depreciate. Maintaining or enhancing these assets
demands reinvestment through performing activities. Both the external and internal assets are
not valuable in and of themselves, however, but
because they fit industry structure and a particular
strategy. Activities performed poorly, or inconsistently with buyer needs, can create liabilities
not assets. At the same time, technological and
other industry changes can nullify assets or turn
them into liabilities.
The value chain provides a template for
understanding cost position, because activities
are the elemental unit of cost behavior. The
move to activity-based costing is a manifestation
of this perspective. The value chain also
I

See Porter (1985. Chapter 3 ) .

See Johnson and Kaplan (1087), Cooper and Kaplan (19XX.

1991).

provides a means to systematically understand the

sources of buyer value and hence differentiation.
Buyer value is created when a firm lowers its
buyers cost or enhances its buyers performance.
This, in turn, is the result of the ways a firms
product as well as its other activities affect the
value chain of the buyer. Firms must not only
create value, but signal that they will do so,
through their sales forces and other activities.
Households and individual consumers have value
chains, just as do industrial or institutional
buyers. By understanding how households perform activities related to a product (e.g. procurement, storage, use, disposal, etc.), the sources
of differentiation can be better understood.
Finally, the value chain provides a tool for
analyzing the added costs that differentiating may
require. Only differentiation that results in a
price premium exceeding the extra costs of
delivering it results in superior performance.

104

M . E. Porter

example, timing may have allowed the firm to

begin advertising early and hence to develop a
If competitive advantage grows out of discrete reputation uncluttered by the competing claims of
activities, however, we once again confront the rivals. The reputation from cumulative advertising
question, why? Why are some firms able to then allows the firm to spend less on current
perform particular activities at lower cost or in advertising or to spend at a comparable rate to
ways that create superior value than others? My rivals but command a premium price. Alternaanswer to this question is the concept of drivers. tively, greater current company sales volume may
These are stuctural determinants of differences lead to efficiencies in advertising that allow the
among competitors in the cost or buyer of firm to enjoy a superior reputation while spending
activities or group of activities. The most at a rate comparable to its rivals. Only by moving
important drivers of competitive advantage in an to the level of underlying drivers can the true
activity include its scale, cumulative learning in sources of competitive advantage be identified.
the activity, linkages between the activity and Tying advantage to specific activities/drivers is
others, the ability to share the activity with other necessary to operationalize the notion in practice.
business units, the pattern of capacity utilization
The value chain also provides the basic
in the activity over the relevant cycle, the architecture for analyzing international strategy
activitys location, the timing of investment and diversification, both fundamentally questions
choices in the activity, the extent of vertical of competitive scope. The central issue in
integration in performing the activity, institutional international strategy involves the spread of
factors affecting how the activity is performed activities to other countries (configuration) and
such as government regulation, and the firms the
integration
of
dispersed
activities
policy choices about how to configure the activity (coordination) (Porter, 1986). In corporate-level
independent of other drivers. The same set strategy for diversified firms, the central issue is
of drivers determines both relative cost and how firms can share activities across businesses,
differentiation. The mix and significance of or share proprietary skills in how to perform
individual drivers varies by activity, by firm, and particular activities though the value chains of
business units are distinct (Porter, 1987).
by industry.
Moving to the level of drivers also sheds light
on the important question of sustainability. The
sustainability of competitive advantage vis-u-vis THE ORIGINS OF COMPETITIVE
rivals depends on the number of competitive ADVANTAGE
advantages in the value chain and, especially,
the particular drivers underlying each one. The This set of frameworks aims to build a careful
durability of an advantage based on learning, for link between the underlying choices a firm
example, depends on the ability to keep the makes in terms of its industry, positioning, and
learning proprietary, while the sustainability of configuration of activities and market outcomes.
advantages due to timing of factor positions The proper choices depend on a firms existing
position, which can be evaluated systematically
depends on factor market imperfections.
Drivers constitute the underlying sources of via its value chain and drivers. The best strategy
competitive advantage, and make competitive also depends on the capabilities and likely
advantage operational. For example, brand repu- behavior of rivals, which can also be assessed
tation is a typical competitive advantage identified through their value chains and drivers. Finally,
by managers. But brand reputation may be a strategy depends on a sophisticated understanding
source of cost advantage (less need for marketing) of industry structure.
Firms inherit positions that constrain and shape
in some cases and a source of differentiation
(and a premium price) in others. The substantive their choices, but do not determine them. They
implications are very different depending on have considerable latitude in reconfiguring the
which it is. Yet, brand reputation is an outcome, value chain with which they compete, expanding
not a cause. The real question is how and why or contracting their competitive scope, and
brand reputation is an advantage. To understand influencing important dimensions of their industry
this, one must move to the level of drivers. For environment. Strategy is not a race to occupy
Drivers

Towards a Dynamic Theory of Strategy

one desirable position, but a more textured
problem in which many positions can be chosen
or created. Success requires the choice of
a relatively attractive position given industry
structure, the firms circumstances and the
positions of competitors. It also requires bringing
all the firms activities into consistency with the
chosen position.
While these frameworks have pushed a considerable distance backward along the chain of
causality, the focus thus far has been on what
might be termed the cross-sectional problem.
What makes some industries, and some positions
within them, more attractive than others? What
makes particular competitors advantaged or
disadvantaged? What specific activities and
drivers underlie the superior positions?
But in answering these questions, we again
confront the question of causality. Why were
particular firms able to get into the advantaged
positions and sustain/or fail to sustain them? This
is what might be termed the longitudinal problem,
which requires crossing the dotted line on Figure

2.13

The frameworks for addressing the crosssectional problem are agnostic as to the process
by which the superior positions were attained,
and largely unaffected by it. Whether the strategy
was consciously chosen, happenstance, the result
of incremental steps, or driven by one major
decision does not itself affect the attractiveness
of the position independently of the activities
and drivers on which it rests. Similarly, the
past process by which firms accumulated their
strengths and capabilities is not, in and of itself,
decisive, The cross-sectional frameworks address
the choice of strategy given whatever array of
capabilities the firm and its rivals possess at a
point in time and can feasibly develop in the
future. The effort by some to dichotomize process
~
are
and substance is simply i n ~ 0 r r e c t . lBoth
necessary and important to understand.
The cross-sectional problem is also logically
prior. Without a rather specific understanding of
what underpins a desirable position, it is virtually
impossible to deal analytically with the process

I avoid the terms static and dynamic intentionally, because

both the cross-sectional and longitudinal problems have both
static and dynamic components.
l4 See, for example, Mintzberg (1990).

105

of getting there. Stratzgy becomes an aimless

process in which luck dztermines the winners.
Assuming an understanding of the crosssectional problem, however, the longitudinal
problem takes on prime importance. Why do
some firms achieve favorable positions vis-u-vis
the drivers in the value chain? Why do some
firms gain scale advantages? Why do some firms
move early, or late, whichever leads to advantage?
Why do some firms conceive of and implement
superior configurations of activities or spot
entirely new and desirable competitive positions?
Logically, there are two answers. The first is
initial conditions. Firms may have pre-existing
reputations, skills, and in-place activities as a
result of their history. These initial conditions
may reside within an individual firm or, as I will
discuss later, in the environment in which the
firm is based. Initial conditions clearly influence
feasible choices as well as constrain them. l 5
The second reason that firms might achieve
favorable positions is through pure managerial
choices, or choices independent of initial conditions, putting aside for the moment the process
by which the choices were made. These managerial choices, which are made under uncertainty
about the future, define the firms concept for
competing (positioning), its configuration of
activities, and the supporting investments in
assets and skills. Pure managerial choices lead
to the assembly or creation of the particular skills
and resources required to carry out the new
strategy.
Numerous case studies illustrate vividly that
highly successful firms often arise out of creative
acts where there were few initial strengths. WalMart decided to locate in small- and mediumsized towns and configure its logistical system in
a particular way because it had a better idea,
not because of any compelling pre-existing
strengths. If anything, its choices were shaped
more by what it did not possess than what it did.
The same could be said about Federal Express,
Apple Computer, Crown Cork and Seal, and
many other companies. American Airlines
developed its MIS systems almost by accident.
Its frequent flyer program was partly a function
of the existence of its MIS system, but other

I 5 Initial conditions can also be set at different points in time.

See below.

106

M . E. Porter

airlines had these as well. American's management was simply more creative.
Many strategies clearly reflect some combination of initial conditions and creative choice.
The balance between the influence of initial
conditions and acts of pure managerial choice
varies by company and industry. Yet there may
well be a tendency, for a variety of reasons to
be discussed later, to overstate the role of initial
conditions.
Lying behind all initial conditions internal to
the firm were earlier managerial choices. The
skills and market position a firm has built today
are the result of past choices about how to
configure activities and what skills to create or
acquire. Some of these choices, as Ghemawat's
(1991) work among others had emphasized,
involve hard-to-reverse commitments down certain paths (path dependency). Earlier choices,
which have led to the current pool of internal
skills and assets, are a reflection of the external
environment surrounding the firm at the time.
The earlier one pushes back in the chain of
causality, the more it seems that successive
managerial choices and initial conditions externul
to the firm govern outcomes.
The importance of managerial choice is also
highlighted by the cross-sectional problem. Whatever configuration of activities and skills a firm
has inherited may or may not be competitively
valuable. Simply having pools of skills. knowledge, or other resources is not in and of itself
a guarantee of success. They must be the
right ones. If managers can understand their
competitive environment and the sources of
competitive advantages, they can better search
creatively for favorable positions that are different
from competitors', assemble the needed skills
and assets, configure the value chain appropriately, and put in place supportive organizational
routines and a culture which reinforces the
required internal behavior. The most successful
firms are notable in employing imagination to
define a new position, or find uew value in
whatever starting position they have.
Towards a dynamic theory

How, then, do we make progress towards a truly

dynamic theory of strategy? Scholars, in both
strategy, organizational behavior, and economics,
sensing this as the frontier question, have made

some headway. There are three promising lines

of enquiry that have been explored in recent
years. Each addresses important questions,
though focusing on a somewhat different aspect
of the problem.

Game theoretical rnodeIs

The first line of inquiry is the proliferation of
game theoretic models of competitive interaction,
referred to earlier, which seek to understand the
equilibrium consequences of patterns of choices
by competitors over a variety of strategic variables
such as capacity and R&D. Since this literature
is reviewed elsewhere in this volume," the
treatment here can be brief. The central concern
of these models is to understand the conditions
that lead to mutually consistent equilibria and
the nature of these equilibria. Each model is
restricted to one or a few variables, and the
environment (technology, products, preferences,
etc.) is assumed fixed except for the variables
examined. Given this structure, timing plays a
central role in determing outcomes. With a frame
of reference in which these assumptions are
plausible, Shapiro (1989) terms this literature a
theory of business strategy.
These models have helped us understand better
the logical consequences of choices over some
important strategy variables. In particular, these
models highlight the importance of information
and beliefs about competitive reaction and
the conditions required for a set of internally
consistent choices among rivals.
Yet, this line of work stops short of a dynamic
theory of strategy. By concentrating sequentially
on small numbers of variables. the models fail
to capture the simultaneous choices over many
variables that characterize most industries. The
models force a homogeneity of strategies. Yet it
is the trade-offs and interactions involved in
configuring the entire set of activities in the value
chain that define distinct competitive positions.
Finally, the models hold fixed many variables
that we know are changing. Ironically, these
models explore the dynamics of a largely static
world. (The papers by Saloner, Camerer and
Postrel in this volume raise additional useful
questions.)
"' Editor's Note: See the articles by Garth Saloner, Colin
Camerer. and Steven Postrel.

Towards a Dynamic Theory of Strategy

Commitment and uncertainty
Another body of work is beginning to emerge on
the problem of making irreversible commitments
under uncertainty. Ghemawats recent book
(1991) is a notable example. The notion here is
that strategy is manifested in a relatively few
investment decisions that are hard to reverse,
and which tend to define choices in other areas
of the firm. These commitments must be made
under uncertainty. Ghemawat highlights the
importance of such choices, and argues that they
should consume much of the attention in strategy
analyses. He posits that analysis of such decisions
must begin with cross-sectional frameworks. In
choosing among feasible positions, however,
Ghemawat stresses the need to carefully examine
their sustainability and the influence of uncertainty in choosing among them. He brings a
broader perspective to bear on sustainability than
is present in the game theory models.
Related to Ghemawats research is work that
seeks to define ways of understanding the
uncertainties a firm faces, and the alternative
ways it can be addressed in strategy choices. The
scenario technique for organizing and bounding
uncertainty has received much attention. More
recently, taxonomies have begun to emerge which
attempt to categorize the ways in which firms can
respond to uncertainty. In In addition, Teisberg
(1991b) begins to explore the biases and heuristics
in decisionmaking in complex and uncertain
circumstances that distort strategy choices, drawing on work in behavioral decision analysis and
cognitive psychology.
This emerging stream of work emphasizes the
lumpiness of strategy choices and the importance
of uncertainty in making them. It sheds important
light on how to approach discrete investment
decisions from a rich strategic perspective. This
comes at the price, however, of a focus on large,
discrete, sequential investments rather than the
simultaneous set of choices throughout the value
chain that define a firms competitive position.
Like the game theoretic models, the environment
is taken as relatively stable (though uncertain)
so that commitments have long-lived consequences and the possibilities for reconfiguring
ISec Waek (198Sa,b) and Schwartz (1991).
See Wernerfelt and Karnani (1987), Porter (1985, Chapter
13), Teisberg (19Yla), and Collis (l9Yla).

107

the value chain are limited. This approach tends

to stress the value of flexibility in dealing with
change rather than the capacity to rapidly improve
and innovate to nullify or overcome it. By
focusing on discrete choices, the discretion a
firm has to shape its environment, respond to
environmental changes, or define entirely new
positions is implicitly limited or not operationalized by most treatments. )

The resource-based view

A third body of research in search of the origins
of Competitive advantage is the so-called resourcebased view of the firm. Closely related to the
resource-based view is the notion of core
competences and treatments that stress intangible
assets. Since this literature is more prominent
and more extensive than that on commitmentiuncertainty, it deserves a more detailed treatment.
Of the three literatures, the resource-based
view is the most introspective and centered on
the firm itself. The argument is that the origins
of competitive advantage are valuable resources
(or competences) that firms possess, which are
often intangible assets such as skills, reputation,
and the like. These resources are seen as relatively
immobile, and as strengths to be nurtured and
which should guide the choice of strategy. The
implicit focus of much of this literature is on the
underpinnings of successful diversification. It
is, of course, essential when diversifying to
understand a firms distinctive strengths
(remember Andrews).
The resource-based view has been proposed
as an alternative theory of strategy. What is
I Teisbergs (199la) essay, by making the influencing of
industry structure a way of dealing with uncertainty. is an
exception.
? Conversations with Cynthia Montgomery have stimulated
and informed my interest in this literature. Perhaps the
pioneer of this school is Penrose (1963). An early paper was
Werncrfelt (1983). For other references, see the bibliographies
in Peteraf (1990) and Collis (1991b). Recent papers include
Barney (1991) and Grant (1991).
Some writers in the resource school draw stylized comparisons with industrial organization (10)-based theories that
confuse rather than clarify. For example, Peterafs survey
(1990) asserts that 10-based models focus only on the
heterogeneity of markets while denying the heterogeneity o f
firms and the existence of differential competitive positions,
to be based only on monopoly rents. to lead only to strategies
of collusion. and to be restricted to formulating strategy at
the business unit level. This view is puzz\ing unless one is
talking about thc 10-based models of the 1970s. before
research aimed at bridging 10 and firm strategy began.

108

M . E . Porter

really unique about a firm, so the argument goes,

is its bundle of resources. It is factor market
impediments, then, rather than product market
circumstances that define success. The role of
internal resources is an important insight for
economic modelers, though less novel a notion
for strategy researchers.
The promise of the resource view for the
strategy field is the effort to address the
longitudinal problem, or the conditions that allow
firms to achieve and sustain favorable competitive
positions over time. As with the other literatures,
however, more work remains to be done. At
its worst, the resource-based view is circular.
Successful firms are successful because they have
unique resources. They should nurture these
resources to be successful. But what is a unique
resource? What makes it valuable? Why was a
firm able to create or acquire it? Why does the
original owner or current holder of the resource
not bid the value away? What allows a resource
to retain its value in the future? There is once
again a chain of causality, that this literature is
just beginning to unravel.
Some authors have begun to deal with these
questions by seeking to specify the conditions
under which resources are valuable. Valuable
resources are those that are superior in use, hard
to imitate, difficult to substitute for, and more
valuable within the firm than outside. Yet
valuable resources, in order to yield profits to
the firm, have been acquired for less than their
intrinsic value due to imperfections in input
markets, which Barney (1986) argues are usually
due to informational asymmetries (read better
managerial choices) or luck.
Yet, the resource-based view cannot be an
alternative theory of strategy. It cannot be
separated from the cross-sectional determinants
of competitive advantage or, for that matter,
from the conception of a firm as a collection of
activities. Stress on resources must complement,
not substitute for, stress on market positions.3
Resources are not valuable in and of themselves, but because they allow firms to perform
activities that create advantages in particular
markets. Resources are only meaningful in the

context of performing certain activities to

achieve certain competitive advantages. The
competitive value of resources can be enhanced
or eliminated by changes in technology, competitor behavior, o r buyer needs which an
inward focus on resources will overlook. More
reliable Japanese products, for example,
degraded the value of Xeroxs copier service
organization. The immobility of resources,
then, is as likely to be a risk as a source of
strength. For every firm with resources that
convey advantage, there will be another (and
perhaps many others) whose bundle resources
impeded change or proved t o be a liability in
light of environmental changes.
Competitive advantage derives from more than
just resources. Scale, sharing across activities, an
optimal degree of integration, and other drivers
have independent influences unless resources
are defined so broadly as to strain credibility. It
is the collective advantage gained from all sources
that determines relative performance.
The conditions which make a resource valuable
bear a strong resemblance to industry structure.
Bargaining power of suppliers refers to input
markets, substitutability to the threat of substitution, and imitability to barriers to entry/
mobility. The bargaining power of buyers, and
the dissipation of resource rents through rivalry
via price cutting or competition from alternative
resource bundles, represent additional threats to
the profitability of firms.
The connection between resources and activities is even more fundamental, however, because
resources represent an inherently intermediate
position in the chain of causality. Resources arise
either from performing activities over time,
acquiring them from outside, or some combination of the two. Both reflect prior managerial
choices. Performing an activity or group of linked
activities over time creates internal skills and
routines which accumulate. It also can create
external assets. A firms reputation, for example,
is a function of the history of its marketing and
customer service activities among other things.
Both internal and external assets depreciate,
however, unless they are reinvigorated through
continuing to perform activities. The rate of
depreciation appears to vary widely across differ22 In this respect, the paper by Prahalad and Hamel (1990)
ent types of assets, and can be rapid. Firms, then,
is perhaps the most inward looking and the most troubling.
have accumulated differing resources because of
2 3 Colliss (1991b) recent paper concludes on this point, which
emerges from his detailed case study of ball bearings.
differing strategies and configuration of activities.
~

Towards a Dynamic Theory of Strategy

Resources and activities are, in a sense, duals of
each other.Z4
Resources, then, are intermediate between
activities and advantage. An explicit link between
resources and activities, along with the clear
distinction between internal and external
resources that was drawn earlier, is necessary to
carefully define a resource in the first place.
Some firm attributes termed resources are
activities-such as sales forces or R&D organizations. A second and more appropriate category
of resources is skills, organizational routines, or
other assets attached to particular activities or
groups of interrelated activities.
The concept of activity drivers allows more
precision in defining how resources were created.
Some skills and routines emerge because of
learning over time. This learning is a reflection
of past strategy choices which have defined how
activities are configured. Other resources were
obtained through well-timed factor purchases
(timing). Still others are the result of the ability
to share across units. In turn, the resource view
adds an important dimension to the concepts of
activities and drivers. Underlying the ability to
link activities or share them across business units,
for example, are organizational skills and routines
that represent important assets.
A final category of resources is external assets
such as reputation and relationships.2s These are
normally created through performing activities
over time. Recognizing these assets, and their
link to the ongoing cost or differentiation of
activities, is another vaiuable contribution of the
resource view. The existence of such assets is
implicit in the concept of drivers but not well
developed.
All this still leaves unanswered the question,
however, of the origins of competitive advantage.
Why can valuable resources be created and
sustained? Interestingly, the requirement of
imperfect factor markets points strongly in the
direction of managerial choice, and goes against
24 Since the great preponderance of resources are created
either by past activities or managerial choices to assemble
outside resources in new activity configurations, my own view
is that activities are logically prior. Yet it is clear that
causality becomes blurred as accumulated resources affcct
ihe cost o r uniqueness of activities.
25 Defining a market position as a resource is inappropriate,
because it confuses the longitudinal problem with the crosssectional problem and obscures the mechanism by which
advantage i s created.

109

the primacy of prior resources (initial conditions)

in determining competitive advantage.
Resources whose value is obvious are bid up in
value. Hence the presence of resourcesiactivities
within the firm that are rent-yielding is likely to
reflect past managerial choices to assemble
resources in unique ways, combine particular
resources in a consistent way with many others,
pursue new undiscovered market positions, or
create resources internally. This allows resources
to be acquired cheaply and avoids the bargaining
away of their value to employees. Few resources
begin as inherently scarce. Their scarcity is
created through choice. Current managerial
choices, in turn, allow the innovative assembly
of new resources and the rendering obsolete of
prior ones.
The resource-based view will have the greatest
significance in environments where change is
incremental, the number of strategic variables
and combinations is limited, so that a few scarce
resources can govern outcomes, and the time
period is short to intermediate term so that
managerial choices can replicate or offset resource
stocks. The greatest value of the resource
view will be in assessing opportunities for
diversification, provided the resource and activity
views are integrated.zh A resource-based view of
diversification that defines resources broadly,
however, runs the risk of justifying the sort of
unrelated diversification that was so disastrous
in the 1970s and 1980s.

THE ORIGINS OF THE ORIGINS

We are left still short of a dynamic theory of
strategy, though we are beginning to learn about
the subprocesses involved. In order to understand
why firms choose and successfully implement the
right strategies, and why their internal activities
and assets are what they are, at least four
important issues must be addressed.
First, a theory must deal simultaneously with
both the firm itself as well as the industry and
broader environment in which it operates. The
environment both constrains and influences outcomes, which the more introspective resource
view neglects. Second, a theory must allow
See Montgomery and Wernerfelt (19%) and Montgomery
and Hariharan (1991).

?''

110

M . E. Porter

centrally for exogenous change, in areas such as

buyer needs, technology, and input markets. If
there is little exogenous change, the choice of
strategy can be viewed as a once-and-for-all game
and the initial stock of (properly defined)
resources can be crucial. In a world where
exogenous change is rapid or relatively continuous, however, the analytical problem becomes
far more complicated. The value of past resources
is continually depreciated or even rendered
negative. The choice of strategy is a series of
ever-changing games in which the position in one
game can influence, but does not determine. the
position in the next one. Case after case illustrates
that the leaders in one generation of products
often fail to lead in the next.
Third, a theory must provide latitude to the
firm not only to choose among well-defined
options but to create new ones. The firm cannot
be seen only as optimizing within tight constraints,
but as having the ability to shift the constraints
through creative strategy choices. other innovative activity, and the assembly of skills and
other needed capabilities. There are alternative
strategies open. The extent to which the environment shapes initial conditions and choice. in
contrast to idiosyncratic, creative decision-making
process within the firm, is a fundamental question.
A final issue that cuts across the others i s the
role of historical accident or chance. There is a
growing belief that historical accidents influence
competitive outcomes. Some of what economists
term historical accidents may simply be good
strategy choices, or reflect so far unmeasured
aspects of the environment. There are often
reasons why firms are lucky, as I will stress
below. Be that as it may, the extent of randomness
in Competition, and the role of true luck. has an
important influence on how one develops a
theory of strategy.
Origins within the firm
How then, do we explain good strategic choices
and the ability to carry them out? One view is
that since the number of variables is substantial
and environmental change is continuous and
unpredictable. the problem is not selecting good
strategies but creating a flexible organization that
learns and is able to continually redefine its
strategy. The resource view, taken to an unhealthy
extreme. is sometimes argued as encompassing

this position. The critical resources are the

capacity for learning and adaptation.
The problem with this notion is its collision with
empirical reality. Most successful organizations
improve but do not change strategy very often.27
They gain advantage from new insights into
competition and from consistent refinement of
their ability to implement a stable overall
strategy (e.g. differentiation) though its details are
continually evolving and improving.
Another view of the origins of advantage is
that it lies in the ability to make good strategy
choices and implement them. While this can
happen by chance, the odds are elevated by
better information and careful analysis. Once a
choice is made, the successful organization i s one
that can bring all its activities into consistency
with the strategy and rapidly accumulate the
necessary activities and resources. New choices
are made as the environment changes or as
accumulating activities and resources open up
new options. But, it must be said, a prominent
role for choice and capacity for implementation
still begs the question of why some firms are
better at it than others.
The environment as the origin of advantage
Instead of solely within the firm, the true origin
of competitive advantage may be the proximate
or local environment in which a firm is based.
The proximate environment will define many of
the input (factor) markets the firm has to draw
on, the information that guides strategic choices,
and the incentives and pressures o n firms to both
innovate and accumulate skills or resources over
time. Competitive advantage, then, may reside
as much in the environment as in an individual
firm. The environment shapes how activities are
configured, which resources can be assembled
uniquely. and what commitments can be made
successfully.
This richer view of the role of the environment
has emerged from my study of the causes of
international competitive success in a large sample
of industries in 10 leading trading nations. This
line of work emerged from a puzzle. After having
written about global strategy, and the ability of

2-

Sce Porter (1090) and Ghcrna\\at (1991)

Towards a Dynamic Theory of Strategy

firms to transcend national markets, I observed
that competitive advantage in particular industries
was often strongly concentrated in one or two
countries, often with several if not many successful
home-based competitors. These local rivals pursue different strategies and push each other to
innovate and improve much more rapidly than
foreign rivals, which allows them to penetrate
and prosper in foreign markets. The concentration
of successful competitors was particularly pronounced if one examined strategically distinct
industry segments rather than broad aggregates,
and if one excludes cases where firms were not
truly successful but merely surviving or sheltered
by government intervention. While the focus of
the research was on the role of the national
environment, it was also clear that successful
firms were also geographically concentrated
within nations. The same theoretical framework
can be used to help explain the concentration of
success in nations, regions within nations, or
even cities. It also seems possible to extend it to
help explain why one particular firm outperforms
others.
The starting point for the theory is that
environmental change is relentless and firms,
through innovation, have considerable latitude
in both influencing their environment and
responding to it. Firms create and sustain
competitive advantage because of the capacity to
continuously improve, innovate, and upgrade
their competitive advantages over time. Upgrading is the process of shifting advantages throughout the value chain to more sophisticated
types, and employing higher levels of skill and
technology. Successful firms are those that
improve and innovate in ways that are valued
not only at home but elsewhere. Competitive
success is enhanced by moving early in each
product or process generation, provided that
movement is along a path that reflects evolving
technology and buyer needs, and that early
mover3 subsequently upgrade their positions
rather than rest on them. In this view, firms have
considerable discretion in relaxing external and
internal constraints.
These imperatives of competitive advantage,
however, collide with the organizational tendencim of firms. Firms value stability, and change
is difficult and unsettling. Strong external or
environmental influences are often essential in
overcoming these tendencies.

111

Environmental determinants of innovation and

upgrading
Four broad attributes of the proximate environment of a firm have the greatest influence on its
ability to innovate and upgrade, illustrated in
Figure 5 . These attributes, which I collectively
term the diamond, shape the information firms
have available to perceive opportunities, the pool
of inputs, skills and knowledge they can draw
on, the goals that condition investment, and the
pressures on firms to act. The environment is
important in providing the initial insight that
underpins competitive advantage, the inputs
needed to act on it, and to accumulate knowledge
and skills over time, and the forces needed to
keep progressing.
The most important factors of production are
highly specialized factors tailored to the needs
of particular industries. Generalized factor pools
are either readily available or easy to source
through global networks. Specialized local factor
pools support the most rapid accumulation of
skill and the greatest rate of innovation. Generic
technology is readily sourced from distant suppliers, but transfer of know-how benefits from
proximity. Specialized factors are almost always
created through private and social investments.

FIRM STRATEGY.
STRUCTURE AND
RIVALRY

FACTOR
CONDITIONS

\ + /

RELATED AND
SUPPORTING
INDUSTRIES

y
I

Figure 5 . Determinants of national competitive

advantage.
'Determinants of national Competitive advantage'
(p. 72): From 77w Competitiv~Advantage of Nations
by Michael E. Porter. Copyright 0 1990 by Michael
E. Porter. Reprinted by permission of The Free Press.
a Division of Macmillan. Inc.

212

M . E. Porter

The presence of unique institutional mechanisms

for creating them in particular industries is an
important determinant of competitive success.
Selective disadvantages in the more basic factors
(e.g. unskilled labor, natural resources) are,
paradoxically, often a source of advantage. They
break dependence on factor costs and trigger
innovation and upgrading.
Home demand is important more for its
character than its size. Home demand plays a
disproportionate role in influencing the perception of buyer needs and the capacity of firms
to improve products and services over time.
Sophisticated and/or especially demanding home
customers often stimulate competitive success, as
do home market needs that anticipate those
elsewhere.
Competitive advantage is also strongly influenced by the presence of home-based suppliers
and related industries in those products, components, machines, or services that are specialized
and/or integral to the process of innovation in
the industry. Inputs themselves are mobile, but
there are local externalities for the process of
innovation in interactions between the firm and
local input suppliers. Home-based suppliers and
related industries provide advantages in terms of
information, signalling, access to new technologies, and market pressures. In many industries,
the scarce technology is know-how, which can
be difficult to transfer without cultural and
physical proximity. Companies with home-based
suppliers have the opportunity to influence
their suppliers technical efforts, help establish
specifications to fit particular needs, serve for
test sites for R&D work, and maintain senior
management contact. All of these accelerate the
pace of innovation.
The final determinant of advantage is firm
strategy, structure, and rivalry, or the context
for competition in a region or nation. The
national and local environments have a strong
influence on management practices, forms of
organization, and the goals set by individuals and
companies. The presence of local rivalry also has
a profound influence on the rate of improvement,
innovation, and ultimate success in an industry.
Local rivals provide a greater stimulus to upgrading than foreign rivals. Proximity speeds information flow and improves incentives to compete.
The presence of domestic competitors negates
basic factor advantages and forces firms to

develop higher order and more sustainable

advantages. Actual rivalry provides a greater
stimulus than potential rivalry. Intense local
rivalry may hold down profits in the home market
but spurs advantages that allow attractive profits
(contingent on overall industry structure) in
global markets.
Local rivalry also feeds back to improve other
parts of the diamond. It overcomes monopsonybased impediments to the development of specialized suppliers, stimulates greater investments in
specialized factors such as university programs
and specialized infrastructure, helps to upgrade
local demand, and so on.
There is a role for true chance events and
historical accidents in the process by which
competitive advantage is created, an issue which
I raised earlier. However, historical accidents are
less common than upon first impression. What
appear to be accidents are really events driven
by conditions in the diamond. Also, the role of
accidents cannot be seen independently of more
stable aspects of the local or national environment. True accidents rarely result in competitive
industries unless other favorable conditions in
the diamond are present. Similarly, accidents
that simultaneously occur in different locations
result in a competitive firm in that location with
the most favorable diamond.
There are many cases where a company
founded in one location, through an act of pure
entrepreneurship, relocated its operations to
another location or even to another country
because that new location offered a better setting
in which to nurture or reap the rewards of that
innovation. The pilgrimage of aspiring actors and
actresses to Hollywood is simply one example of
how ideas and talent flow to the environment in
which they can command the highest returns.
The ability to command the highest returns
depends on the simultaneous presence of unusual
local demand, related industries, active rivals
bidding, and other aspects of the diamond.
A final influence on the environment for
competitive advantage is government. The role
of government policy is best understood by
looking at how it influences the diamond.
Government at all levels can improve or impede
national advantage through its investments in
factor creation, through its influence on the goals
of individuals and firms, through its role as a
buyer or influencer of buyer needs, through its

Towards a Dynamic Theory of Strategy

competition policies, and through its role in
related and supporting industries, among other
ways. Government plays an important part in
shaping the pressures, incentives, and capabilities
of the nations firms.
Governments proper role is as a catalyst and
challenger. It is to encourage, or even push,
companies to raise their aspirations and move to
higher levels of competitive performance, even
though this process may be unpleasant and
difficult. Government plays a role that is
inherently partial, and that succeeds only when
working in tandem with favorable underlying
conditions in the diamond. Government policies
that succeed are those that create an environment
in which companies can gain competitive advantage rather than those that involve government
directly in the process. (It is an indirect, rather
than direct, role).

The diamond as a dynamic system

These aspects of the local environment constitute
a dynamic system. This character of the environment bears centrally on the firm processes that
give rise to advantage. The effect of one
determinant depends on the state of others. The
presence of sophisticated and demanding buyers,
for example, will not result in advanced products
or production processes unless the quality of
human resources enables firms to respond to
buyer needs. There must also be a climate
that supports sustained investment to fund the
development of these products and processes.
Similarly, having selective disadvantages in basic
factors (e.g. higher labor, energy, or raw material
costs) will not spur innovation and upgrading
unless there is an environment of vigorous
competition among firms to trigger innovation in
people, products, and processes. It is this
contingent relationship that explains why, for
example, a selective factor disadvantage stimulates innovation in one country while hastening
decline in another.
The parts of the diamond are also mutually
reinforcing. The development of specialized
supporting industries, for example, tends to
increase the supply of specialized factors. Vigorous domestic rivalry stimulates the development
of unique pools of specialized factors. This is
particularly likely if the rivals are all located in
one city or region. The University of California

113

at Davis, for example, has become the worlds

leading center for wine-making research, working
closely with the California wine industry. Active
local rivalry also upgrades home demand through
educating buyers and providing choice, and
promotes the formation of related and supported
industries. Japans world-leading group of semiconductor producers, for example, has spawned
world-leading Japanese semiconductor equipment
manufacturers.28
The effects can work in all directions, and
causality can become blurred over time. Sometimes world-class suppliers become new entrants
in the industry they have been supplying. Or
highly sophisticated buyers may enter a supplier
industry, particularly when they have relevant
skills and view the upstream industry as strategic.
In the case of the Japanese robotics industry, for
example, Matsushita and Kawasaki originally
designed robots for internal use before selling
them to others.
The diamond also bears centrally on a nations
ability to attract factors of production, rather
than merely serve as a location for them. This
represents a final form of mutual reinforcement.
Mobile factors, particularly ideas and highly
skilled individuals, are becoming increasingly
important to international competitiveness.
Mobile factors tend to be drawn to the location
where they can achieve the greatest productivity,
because that is where they can obtain the highest
returns. The theory outlined above, because it
focuses on the determinants of productivity, also
explains the attraction of mobile factors. The
same features that make a nation an attractive
home base also help it attract mobile factors.
The national and local environment for competing in a particular industry itself evolves in a
dynamic process. The environment is created
over time through the mutual reinforcement of

zx The mutual reinforcement of the determinants suggests a

particularly important role for local rivalry, healthy new
business formation, and the responsiveness of local institutions
to signals from industry. Local rivalry stimulates improvements
in all the other determinants. New business formation.
whether through start-up or internal development, is a sine
yua non of developing related and supporting industries as
well as healthy rivalry. Competitive advantage also depends
on the capacity of the local education system, infrastructure
providers, and government institutions to respond to the
specialized needs of particular industries. Institutional responsiveness allows the proper types of skills, resources, and
infrastructure t o be created.

114

M . E. Porter

the determinants. It begins with an inheritance

drawn from other industries and history. and
exists amid a set of local institutions, values, and
attitudes. The mutual reinforcement of the
determinants, which reflect in part the actions of
firms themselves, build up the environment over
time. As this process proceeds, causality becomes
blurred.
In industries with modest levels of skill and
technology, firms can gain advantage solely on
the basis of factor advantages such as cheap labor
or abundant raw materials. Such advantages are
notoriously unstable, however, in a world of
globalization, technological change, and rapid
substitution. Competitive advantage in more
sophisticated industries and industry segments.
on the other hand, rarely results from strength
in a single determinant. Sustained success in
these industries and segments usually requires
the interaction of favorable conditions in several
of the determinants and at least parity in the
others. This is because advantages in various
parts of the diamond are self-reinforcing.
It is thus the juxtaposition of advantages
throughout the diamond in one location that
leads to competitive success. far more than the
presence of any single advantage no matter how
compelling. The firm's home base, which consists
of that group of activities most centrally involved
in the process of innovation and learning, must
be located in the same place to allow the
internal coordination and contact with the local
environment that is necessary for rapid progress.
The ability of a firm to progress rapidly and
appropriately by integrating research and production spread widely among several locations.
combining machines sourced from many disparate
distant suppliers, and so on is limited. The firm
that concentrates its core activities at a favorable
home base, while competing nationally and
globally, will normally progress more rapidly.
Activities outside the home base are focused on
sourcing low-cost basic factors and securing
market access.
These same arguments explain why we observe
clusters of competitive industries in one location.
Clusters involve supplier industries, customer
industries, and related industries that are all
competitive. Such clusters are characteristic of
every advanced economy-American
entertainment, German chemicals, Japanese electronics,
Danish foods. Clusters grow and transform

themselves through spinoffs and diversification

of firms into upstream, downstream, and related
industries and activities. The fields where several
clusters overlap are often fertile grounds for new
business formation. In Japan, for example, the
interstices between electronics and new materials
are spawning new competitive strengths in fields
as diverse as robotics and displays.
Another implication of this theory is the
importance of geographic concentration of successful firms and clusters within particular cities
or regions. National clusters are often themselves
geographically concentrated. Geographic concentration elevates and magnifies the interaction of
the four determinants, improves information
flow and signaling, makes innovation-enchancing
interactions with customers and specialized suppliers less costly, and provides a check on
opportunistic behavior, among other benefits.2"
Firms lose competitive advantage either
because of emerging weaknesses in their local
environment or due to rigidities or other internal
problems that external circumstances cannot
overcome.3" For example, a major shift in
technology may require an entirely new set of
specialized suppliers that are not present, or local
demand characteristics may evolve in ways that
distract from instead of forehadow international
needs. However, firms sometimes fail not because
their environment is unfavorable but because of
organizational or managerial rigidities that block
improvement and change. The environment can
provide important pressures to advance, but firms
differ in their responsiveness to them.

Environmental influences on the dynamics of

strategy
The environment, via the diamond, affects both
a firm's initial conditions and its managerial
choices. The diamond, through its influence on
information and incentives, shapes the content
of strategies. It influences the ability of firms to
carry out particular types of strategies, hence
limiting choice. Choices that may look accidental

For a detailed treatment and empirical tests, see Enright

(1990). See. also. Krugman (1991).
''I I discuss these issues more fully elsewhere. See Porter
(IY9O).
?''

Towards a Dynamic Theory of Strategy

or internally driven are often partly or wholly
derived from t h e local diamond. Over time,
through its stress on markets for difficult-to-trade
inputs, the state of the diamond conditions the
rate of accumulation of resources. It also sets
the pressures on firms to improve and upgrade.
The diamond, then, begins to address a dynamic
theory of strategy early in the chain of causality.
Yet firms retain a central role. Firms must
understand and exploit their local environment
in order to achieve competitive advantage. There
are often sharp differences in the performance
of firms based in the same region or nation.
These differences are partly a function of
managerial choices, differential rates of resource
accumulation, or chance. The differences also
appear, however, to be partly a function of
the subenvironment of each particular firm-its
particular early customers, supplier relationships,
factor market access, etc.
An important role for the local environment
in competitive success does not eliminate the role
of strategy nor the need for competitive analysis.
Industry structure, positioning, activities,
resources, and commitments remain important.
Rather, the diamond highlights new issues for
strategy that are normally ignored, such as the
importance of developing and nurturing homebased suppliers, the importance of local specialized factor markets, and the balance between
home-based activities and those dispersed to
other locations as part of a national or global
strategy.
The local environment creates potential for
competitive success, but firms must sense and
respond to it. Firms also have a considerable
ability to influence their environment in ways
that reinforce or detract from their capacity to
accumulate skills and resources and to innovatethere is a feedback loop between firm actions
and the local diamond. Many if not most firms,
even in favorable environments, do not achieve
competitive advantage. Firms based in an unattractive environment, however, face profound
challenges in achieving competitive success. More
and more firms are relocating their home bases
accordingly.
Issues for further research

Recent research has begun to shed some light

on the chain of causality that constitutes a

115

dynamic theory of strategy, but many unanswered

questions remain. I would highlight four that
deserve special attention. First, we need to better
understand the balance between environmental
determinism and companylleader choice in shaping competitive outcomes. What is emerging is
the beginnings of a more sophisticated way of
understanding how the environment surrounding
a firm influences both firm choices and outcomes,
and of the internal processes of choice and of
skill and asset (resource) accumulation that
underpin competitive advantage. It is clear that
company actions still matter, and that firms in a
given environment achieve widely different levels
of success. Can we, by looking across different
firms in a given nation or city, isolate unique
subenvironments that explain these differing
levels of performance? Or, can we identify
patterns of commitments to activities and resource
accumulation that characterize superior performers?
Second, we need to better understand the
degree of stickiness or inertia in competitive
positions once a firm stops progressing, or, to
put it another way, the durability of early
mover advantages? How important is a burst of
innovation vs. the capacity to improve and
innovate continuously? How important are pure
rents from scarce factors vs. advantage resulting
from innovation or which raise the value of
factors? My research suggests that the latter
is characteristic of successful firms who have
sustained their competitive positions, but much
more investigation is necessary,
Third, we need to know how necessary or
helpful it is to push even further back in the
chain of causality. I have argued that resources
are an intermediate step in the chain, from which
we can learn. Yet an important theoretical issue
is where in the chain of causality to best cut into
the problem. An example will illustrate that even
a focus on the local environment of the firm does
not go to the ultimate origin of advantage. The
presence of a specialized skill in a region or
nation is often the result of skill pools inherited
from other industries as well as human resources
trained at pre-existing institutions. These institutions, however, often draw to some extent on
the general education system which itself is
affected by social values and history. Just how
far back to the ultimate source does one need
to go to best examine these questions? I chose

116

M . E. Porter

in my own research to model the phenomena at

the level of the diamond, while highlighting that
each of its components is the result of history
and other local conditions. The appropriateness
of this choice is a subject for research. It should
be said that understanding the ultimate origins
of advantage may not always be necessary for
thinking about how to improve future advantage.
Finally, there is the important challenge of
crafting empirical research to make further
progress in understanding these questions. Some
argue that models that exceed a certain level of
complexity can never be tested. Yet it is clear
that there are many aspects of both firms and
their environment which determine competitive
success. How can we collect and analyze data to
help us discriminate among explanations and
weigh the various factors? I concluded in my
most recent research that detailed longitudinal
case studies, covering long periods of time, were
necessary to study these phenomena. Moreover,
these case studies had to encompass a large
number and wide range of industries and national
contexts in order to develop confidence about
the appropriate variables and their influence on
outcomes. This style of research nudges strategy
research, and indeed industrial economics, into
the world of the historian. It also involves
enormous fixed costs. I am convinced that more
research of this type will be needed to address
the dynamics of strategy. It also raises the
question of whether there are other approaches
to empirical testing to address these issues, or
whether we must wait until theories have been
much better developed before we can highlight
the relatively few variables which can be measured
and rigorously examined statistically.

ACKNOWLEDGEMENTS
I am grateful to Pankaj Ghemawat, Cynthia
Montgomery, and others in the Competition and
Strategy group at Harvard for a long series of
discussions that have immensely benefitted my
thinking about these issues, and to Jay Barney,
Richard Rumelt, and Garth Saloner for their
insights while visiting. David Collis, Cynthia
Montgomery, Richard Rumelt, Elizabeth Teisberg and Dan Schendel provided helpful comments on this manuscript.

REFERENCES
Andrews, K. R. The Concept of Corporate Strategy,
Dow Jones-Irwin, Homewood, IL, 1971.
Baldwin, C. Y. and K. B. Clark. Capabilities and
capital investment: New perspectives on capital
budgeting, Harvard Business School Working
Paper #92-004, Rev. 7/15/1991.
Barney, J. B. Strategic factor markets: Expectations,
luck and business strategy, Management Science,
32. October 1986, pp. 1231-1241.
Barney, J. B. Firm resources and sustained competitive
advantage. Journal of Management, 17( l), 1991,
pp. 99-120.
Collis, D. J. The strategic management of uncertainty,
Harvard Business School Working Paper #89-019,
Rev. 311991a.
Collis, D. J. A resource-based analysis of global
competition: The case of the bearings industry,
Srrategic Management Journal, Special Issue, Summer 1991b, pp. 49-68.
Cooper, R. and R. S. Kaplan. Measure costs right:
Make the right decisions, Harvard Business Review,
September-October 1988, pp. 96103.
Cooper, R. and R. S. Kaplan. The Design of Cost
Management Systems. Prentice Hall, Englewood
Cliffs, NJ, 1991.
Enright, M. J. Geographic concentration and industrial
organization, unpublished doctoral dissertation,
Business Economics Program, Harvard University,
1990.
Ghemawat, P. Commitment: The Dynamic of Strategy.
Free Press, New York, 1991.
Grant, R. M. The resource-based theory of competitive advantage: Implications for strategy formulations, California Management Review, Spring
1991, pp. 119-135.
Itami, H. (with T. W. Roehl). Mobilizing Invisible
Assets, Harvard University Press, Cambridge, MA
and London, England, 1987.
Johnson, H. T. and R. S . Kaplan. Relevance Lost:
The Rise and Fall of Management Accounting,
Harvard Business School Press, Boston, MA, 1987.
Krugman, P. R. Geography and Trade, MIT Press,
Cambridge, MA, 1991.
Learned, E. P . , C. R. Christensen, K. R. Andrews
and W. D. Guth. Business Policy Text and Cases,
Richard D. Irwin, Homewood, IL, 1965.
Mintzberg, H. The Design School: Reconsidering the
basic premises of strategic management, Strategic
Mangagement Journal, 11(3), March-April 1990,
pp. 171-195.
Montgomery, C. A. and B. Wernerfelt. Diversification, Ricardian Rents, and Tobins q, Rand
Journal of Economics, 19(4), Winter 1988,
pp. 623-632.
Montgomery, C. A. and S. Hariharan. Diversified
expansion by large established firms, Journal of
Economic Behavior and Organization, 15, 1991,
pp. 71-89.
Penrose. E. T. The Theory of the Growth of rhe Firm,
Blackwell, Oxford, 1963.

Towards a Dynamic Theory of Strategy

Peteraf, M. A. The resource-based model: An
emerging paradigm for strategic management,
Discussion Paper 90-29, J. L. Kellogg Graduate
School of Management, Northwestern University,
August 1990.
Porter, M. E. Competitive Strategy: Techniques for
Analyzing Industries and Competitors, Free Press,
New York, 1980.
Porter, M. E. Competitive Advantage: Creating and
Susfaining Superior Performance, Free Press, New
York, 1985.
Porter, M . E. From competitive advantage to corporate strategy, Harvard Business Review, May-June,
1987, pp. 43-59.
Porter, M. E. The Competitive Advantage of Nations,
Free Press, New York, 1990.
Porter, M. E., (ed.) Competition in Global Industries,
Harvard Business School Press, Boston, MA, 1986.
Porter, M. E. and V. A. Millar. How information
gives you competitive advantage, Harvard Business
Review, July-August 1985, pp. 149-160.
Prahalad, C. K . and G. Hamel. The core competence
of the corporation, Harvard Business Review,
May-June 1990, pp. 71-91.

117

Schwartz, P. The Art of the Long View, Doubleday.

New York, 1991.
Selznick, P. Leadership in Administration: A Sociological lnferpretation,Harper & Row, New York, 1957.
Shapiro, C. The theory of business strategy, Rand
Journal of Economics, 20, Spring 1989, pp. 125137.
Teisberg, E. 0. Strategic response to uncertainty,
Harvard Business School case, #N9-391-192, Rev.
4/11/1991a.
Teisberg, E. 0. Why do good managers choose
poor strategies? Harvard Business School Case
#N%391-172, Rev. 3/5/1991b .
Wack, P. Scenarios: Uncharted waters ahead, Harvard Business Review, September-October 198Sa,
pp. 73-89.
Wack, P. Scenarios: Shooting the rapids, Harvurd
Business Review, November-December 1985b,
pp. 139-150.
Wernerfelt, B. A resource-based view of the firm.
Strategic Management Journal, 5 ( 2 ) , April-June
1984, pp. 171-180.
Wernerfelt, B. and A. Karnani. Competitive strategy
under uncertainty, Strategic Management Journal,
8(2), 1987, pp. 197-194.