2014

AfDB
www.afdb.org

E c o n o m i c B r i e f
CONTENTS Innovation and Productivity Empirical
Analysis for North African Countries
Abstract

Key Messages
Introduction p.2
• In North Africa, especially in Morocco and Egypt, the impact of qualified human resources on innovation incentives and
productivity levels is insignificant. This points to the under-utilization and inefficient allocation of capital in these countries.
I – Determinants of • The impact of exports on innovation is insignificant, mainly as a result of the rigid structure of comparative advantages
Innovation and Productivity in these countries and the concentration of exports in sectors with little value addition and limited technological potential,
particularly in the case of Morocco.
in Developing Countries p.5
• To foster innovation and stimulate productivity, special attention should be paid to certain aspects of the national innovation
system. The recommendations of this note are as follows:
II – Empirical Analysis for • Strengthen governance of the national innovation system;
• Stimulate the research and corporate environment with a view to decompartmentalizing the two spheres and
North African Countries p.8 ensuring more efficient and effective interfacing;
• Build support for the higher education system and vocational training in order to enhance human resource
III – Main Results p.11 competencies to obtain a better contribution to the production and innovation processes;
• Create more advantageous incentive programmes for innovative FDI with higher value-addition, which is
sufficiently well integrated upstream and downstream of the local economy; and
IV – Conclusions and • Establish mechanisms for contractualization, supported by the State in the area of science and technology
between the research centres, universities, potentially innovative local firms and foreign companies wishing
Recommendations p.13 to relocate their R&D activities, to strengthen cohesion between the different components of the national
innovation system around innovative projects.

Annex I : Description of
Variables p.15 Abstract:
Annex II : Econometric his article highlights the key determinants countries, innovation is far from being the result
Models p.16
T of innovation and their impact on the
performance of firms in three North African
of R&D and new technology creating activities
alone. It also occurs by the adoption and
Annex III : Estimation
countries (Algeria, Egypt and Morocco) on the adaptation of technologies created elsewhere
Results p.17
basis of World Bank survey data on the investment through learning and assimilation-related
Bibliographical climate. Initially, our econometric approach mechanisms requiring more highly qualified
References p.20 consists of estimating the impact of the traditional human capital and improvement of the
determinants of innovation by underscoring investment climate. We have also shown
the critical role played by human capital in the weakness of the effect of technological
technological ownership and absorption. We externalities generated by export and foreign
then estimate the relationship between innovation investment activities on innovation potential. The
Zondo Sakala and productivity taking into account certain rigid structure of comparative advantages and
Vice President ORVP
characteristics of the investment climate and the the concentration of exports and FDI in activities
[email protected]
quality of infrastructure and public services. with limited value addition which are poorly
integrated in the local economy generate few
Jacob Kolster The main results suggest that, in North African upstream-downstream externalities.
Director ORNA
[email protected]
+216 7110 2065 This report was prepared by Moez El Elj (Institut Supérieur de Gestion) under the supervision of Vincent Castel (Chief
Country Economist, ORNA) with the support of Kaouther Abderrahim Ben Salah (Young professional ORNA). This work was
conducted under the general guidance of Jacob Kolster (Director ORNA). The analysis and conclusions in this report reflect the
views of their authors and not those of the Group of the African Development Bank, its Board of Directors or the countries
they represent.

A f r i c a n D e v e l o p m e n t B a n k
 AfDB Economic Brief

2 0 1 4 • w w w . a f d b . o r g

Introduction

oday, the new international division of labour and competition from lack of competitiveness and weak innovation capacity in North African
T new emerging markets increasingly steers the production objectives
of the industrialized countries towards more sophisticated products
countries. Indeed, significant gaps remain in terms of R&D potential,
innovation capacity, the quality of research institutions, technological
with a high technological content and makes technological innovation cooperation and the quality of human capital (Tables 1 and 2). In this
capacity a key factor in international competition. Against this backdrop, region, the share of R&D in GDP is very low compared to the mean in
the economic systems are put under ever-increasing pressure resulting developed countries3 and it is worth noting that innovation, R&D and
from the complexity of cross-border networks of flows of knowledge, technological cooperation tend to be ad hoc, spur-of-the moment
ideas and technology. Joining the race for innovation by assuming activities with the result that few firms have an R&D department.
ownership of these flows of new technologies and knowledge then According to Arvanitis et al. (2007), the North African Region suffers
becomes a critical issue for firms and nations. Integration of these mainly from strong State intervention both in the financing and conduct
networks will determine the countries’ growth and development paths. of R&D. Research activities appear to be mainly confined to the public
Currently, global trends in this area unfortunately show that the circulation domain and are poorly interconnected to the productive sector. The
of these flows is bringing about a new configuration of innovating clusters private production sphere and research sphere are compartmentalized
in the world from which the North African Region seems to be excluded. and subject to different requirements. The private sector is not a seeker
Indeed, over the years, the North African countries seem to have lagged of public research given its low level of specialization in low value-added
considerably behind in this area while the lack of knowledge acquisition, activities, and public research specializes in subjects of little or no interest
absorption and use seem to be among the main constraints on human to the few innovating firms. Furthermore, these countries also suffer
development.1 However, since the 90s, the North African region has from administrative red-tape which constrains the interface between
been the focus of particular attention especially from the European the research and production environments. Yet, the very nature of R&D
Union with a view to strengthening cooperation and partnership ties and innovation (complexity, cost, risk, etc..) calls for flexible and
aimed at reducing conflicts in the region and creating a climate of stability pro-active interface structures.
and prosperity which will ensure that the North African countries catch
up in the medium to long term. In this respect, many initiatives have However, the weakness of R&D activities is not in itself a constraint on
been taken and various association and cooperation agreements have innovation in developing countries. Theoretical and empirical work
been signed as part of the Barcelona process. This process of integration (Goedhuys (2007b), Dabla-Norris E., Kersting E., and Verdier, G. (2010))
and its extension still represent an opportunity and challenge for the shows that the capacity to absorb knowledge and technology especially
countries on the southern shores of the Mediterranean insofar as these through foreign direct investment and its assimilation by qualified human
countries now compete directly with their neighbours of the North, which capital are rather drivers of innovation and increased productivity.
implies reconsideration of their comparative advantages and an industrial However, it appears that, for North African countries, the policy of
shift focused on the production of goods with higher value addition. opening up to trade and foreign capital, especially to European
However, despite the different association agreements and support, companies has not had the expected impact in terms of technological
modernization and upgrading programmes, North Africa’s share of transfers and innovation. The expected technological externalities and
2
European trade flows remains fairly low and trade is till characterized their impacts on the local economy seem to be highly limited. Indeed,
by rigid comparative advantages and over-specialization in the production although the EU is the main partner of North African countries, this
of low valued-added, low cost goods. Sectoral specialization with low region attracts a limited proportion of European Foreign Direct Investment
value addition and the rigidity of comparative advantages reflect the (FDI) (less than 1%)4 and most of it is concentrated in low value-added

1
Arab Human Development Report (2010)
2
EU imports from North African countries represent almost 3.19% of total European imports while EU exports to North African countries do not exceed 4.6% of total EU exports.
3
The share of R&D in GDP is 0.23% in Egypt, 0.64% in Morocco and 1.2% in Tunisia while the average ratio is about 1.8% in Europe.

2
A f r i c a n D e v e l o p m e n t B a n k
 Economic Brief
AfDB
2 0 1 4 • w w w . a f d b . o r g

sub-contracting activities. A significant proportion of FDI mainly concerns microeconomic data used in the World Bank Investment Climate
the off-shore regime which mostly targets the energy sectors or low Survey (WBICS)6 to estimate the impact of the main determinants
cost, unskilled labour intensive industrial activities. These activities are of innovation on the performance of firms in the region while
often poorly integrated into the local economy and consequently highlighting the role played by economic openness, the institutional
generate few upstream or downstream impacts. environment and qualifications of human capital.

Despite the importance of the innovation issue, few studies have The paper is organized as follows. The first section discusses empirical
explored the specificity of innovation and its impact on productivity in works on the specificities of innovation and its determinants as well
developing countries. (Bogliacino et al, 2009)5 For North African as its role in improving the performance of firms in developing
countries, only Morocco since 1999 and Tunisia in 2005 and in 2009 countries. The second section provides a brief description of the
have carried out innovation surveys in order to provide data that will survey data and estimated econometric models. The third section
help to identify the specificities of innovation in these countries (Arvanitis presents the main results of the econometric analysis of the
et al. (2010)). determinants of innovation and the relationship between innovation
and productivity in North African countries. Finally, the last section
This paper’s objective is to contribute to the analysis of the presents the main conclusions and recommendations underscoring
determinants of innovative decisions and examine their impact on the critical role played by institutional factors, the investment climate
the performance of firms in three North African countries which are and human capital in the innovation and productivity dynamic in the
partners of Europe (Algeria, Egypt and Morocco). We shall use the North African region.

4
ANIMA 2011
5
Since the 90s, only the countries of Latin America have carried out innovation surveys designed on the basis of the Bogota Manual which draws on the Oslo Manual which is the
benchmark for innovation surveys in the industrialized countries.
6
http://www.enterprisesurveys.org/

3
A f r i c a n D e v e l o p m e n t B a n k
 AfDB Economic Brief

2 0 1 4 • w w w . a f d b . o r g

Table 1: Global Competitiveness Index 2011

Competitive Index Higher Education and Technological Innovation

Training Readiness
North African Countries

Algeria 86 98 106 107

Egypt 81 97 87 83

Morocco 75 102 75 81

Tunisia 32 30 55 31

European Countries

Germany 5 19 10 8

France 15 17 12 19

Italy 48 47 43 50

Spain 42 31 30 46

Source: Global Competitiveness Report, World Economic Forum 2011

Table 2: Main Innovation Indicators 2011

Capacity for Quality of Company University- Availability of Tertiary Availability Firm-level

Innovation Scientific Spending Industry Scientists and Education of Latest Technology
Research on R&D Collaboration Engineers Enrollment Technologies
Institutions in R&D

North African Countries

Algeria 125 96 106 119 43 80 109 128

Egypt 109 110 74 120 25 78 91 58

Morocco 94 93 97 104 46 102 68 74

Tunisia 36 38 35 41 7 69 42 33

European Countries
France 8 19 13 44 12 39 16 23

Germany 1 6 4 9 27 52 17 14

Spain 42 43 47 46 47 17 32 49

Source: Global Competitiveness Report, World Economic Forum 2011

4
A f r i c a n D e v e l o p m e n t B a n k
 Economic Brief
AfDB
2 0 1 4 • w w w . a f d b . o r g

I. Determinants of Innovation and Productivity in Developing Countries

I.1. Innovation in Developing Countries the personnel employed, strengthens the firm’s innovation potential.
However, this effect is not always evident since the existence of
lthough the issue of innovation and its role in economic growth qualifications is not always synonymous with the optimization of skills.
A have been widely studied in the context of developed countries,
few works have been dedicated to the study of the specificities and
In some cases, under-used human capital in value-adding and
knowledge generating activities does not always effectively contribute
determinants of innovation in developing countries. These works point to innovation.7
out that the traditional determinants of innovation identified in the
abundant empirical literature on developed countries are not always Concerning technological externalities likely to be generated by economic
significant in developing countries, which implies a deeper investigation openness as a result of the trade of goods and services on the one
taking into consideration the specificities of these countries in the hand, and the promotion of foreign direct investment inflows on the
area of human capital accumulation and technology absorptive other, the findings of empirical studies differ and do not reach any
capacity. conclusion on the importance of spillover effects of openness on the
host country’s capacity for innovation. Indeed, although some works
Of the traditional determinants highlighted in the abundant theoretical stress the significant effect of foreign direct investment on innovation
and empirical literature, firm size and R&D intensity also appear to be at domestic firm-level in some developing countries (Almeida and
significant indicators in some developing countries. (Ayyagari et al. Fernandes, (2008), Yuriy et al. (2010)), others find an statistically
(2007), Pamukcu (2003), Almeida and Fernandes (2008) and Yuriy et insignificant effect of the participation of foreign capital on the capacity
al. (2010), El Elj (2012), Rahmouni et al. (2010)). Indeed, large firms for innovation (El Elj (2012), Pamukcu (2003), Karray et al. (2009)). These
benefiting from economies of scale and with a more advantageous findings are explained by the positioning of off-shore companies with
market position have easier access to financial resources to carry out foreign capital in low valued-adding sub-contracting activities and by
costly R&D activities. They are also better known and have more the lack of integration of these companies into the local environment
experience, which allows them to collaborate both with research centres both upstream and downstream. The foreign partnership seems to be
and with foreign firms. Furthermore, large firms have more resources directed more towards making productivity gains based on low costs
to purchase or use patented innovations in return for royalty payments. especially payroll costs than towards promoting innovation. In this
Licenses are a major source for the acquisition of new technologies context, technical externalities are virtually zero and the expected impact
especially in developing countries (Almeida et Fernandes, 2008) even of foreign direct investment and the potential for innovation at local firm
though, in some cases, restrictions imposed in license purchase level very low.
agreements impede the transfer of technologies (Pamukcu (2003) and
Koouba et al. (2010)). With regard to the export effect, the empirical works of Ayyagari et al.
(2007) and Yuriy et al. (2010) for some developing countries suggest
Furthermore, the empirical studies show the critical importance of that exporting firms innovate more than those focused on the local
managerial quality and qualification of human resources in building a market. Export firms more exposed to pressure from foreign competition
firm’s knowledge base as a source of technological and technical are encouraged to innovate to improve their structural competitiveness.
innovation and ownership. In this context, the works of Pamucku (2003), However, other works show that this effect is not always significant
Ayyagari et al. (2007) and Almeida and Fernandes (2008) on some insofar as it depends on a country’s sector specialization and the nature
developing countries have shown that the quality of human resources, of its comparative advantages. In the case of Turkish firms, Pamukcu
measured by the manager’s experience and the level of education of (2003) showed that the export effect on innovation is statistically

7
Karray et al. (2009), Koouba et al. (2010) and El Elj (2012) pointed out the unexpected impact of the quality of human resources on innovation potential in Tunisia.

5
A f r i c a n D e v e l o p m e n t B a n k
 AfDB Economic Brief

2 0 1 4 • w w w . a f d b . o r g

insignificant. The same result was underscored in the case of Tunisian the effects of organizational and marketing innovations are only
firms (El Elj (2012), Rahmouni et al. (2010)). It appears, therefore, that significant for Argentina and Colombia.
exporters operating mainly in medium to low technological intensity
sectors and working as sub-contractors of European firms specialized Other work carried out by Goedhuys (2007b) and Goedhuys et al.
8
in low value-added activities are affected by constraints relating to cost (2008) on the basis of the World Bank Investment Climate Surveys
minimization and continuing improvement of productivity gains and not (ICS) stressed that R&D and innovation activities as they are usually
by constraints of innovation and continuing improvement of products measured may not be considered as explanatory factors of firms’
and processes. productivity in developing countries.

I.2. Innovation and Productivity in Developing These works reflect the importance of reconsidering theoretical and
Countries empirical investigations on the specificity of innovation and on the
nature of the relationship between productivity and innovation in
The issue of the relationship between productivity and innovation has developing countries. In these countries, R&D intensity is generally
been widely studied in most developed countries to stress the very low and technological ownership through the purchase or use of
importance of R&D activities (through cooperation agreements with patents or through technological cooperation with research centres
research centres and universities, suppliers and clients and in some and laboratories is not always effective. Also, it must be noted that
cases competitors) in improving productivity levels both for the firm and innovations in developing countries are usually minor and incremental
for the economy as a whole. In these developed countries, the and not often patented. The patentability of inventions is not often a
relationship between R&D, innovation and productivity was tested on common practice for innovating firms in these countries.9
the basis of the European Community Innovation Survey (CIS) with
reference to the Crépon, Duguet and Mairesse (1998) (CDM) basic In this context, in order to thoroughly analyze the specificity of
theoretical model. The empirical results confirm the significant relationship innovation in emerging and developing countries and its impact on
between R&D, innovation and productivity in both the industrial and productivity, Dabla-Norris et al. (2010) opt for a broader definition of
service sectors (Griffith et al (2006), Lopes and Godinho (2005), Mairesse innovation than the one usually retained for developing countries.
and Mohnen (2010), Polder et al. (2010), Hall (2011)). Indeed, there is Beyond the radical innovations of products and processes, the authors
a close correlation between innovation as measured by patents issued consider an aggregate indicator for innovation which comprises a
and the intensity of R&D and other technological ownership variables broader series of either major or minor innovating activities (products
(cooperation, license, etc..). It was also demonstrated that there is a or processes) or of minor organizational innovations. Their empirical
strong causal relationship between productivity and innovation in which tests on 14000 firms in 63 emerging and developing countries show
the quality of human capital plays a major role. the significance of the causal relationship between productivity and
innovation in the broad sense. Furthermore, they demonstrate the
However, an estimation of the CDM model in certain emerging and relative importance of absorptive capacity and learning mechanisms
developing countries did not produce the same results. For example, as determining factors of innovation and increased productivity of
in the case of Chile, Benavente (2006) shows that neither R&D firms in developing countries. Technological ownership is reflected in
spending, nor the share of sales of innovating products or services the purchase and exploitation of patented inventions, cooperation
have any significant impact on productivity levels of firms, while with innovating firms, the practice of reverse engineering, etc…and
Chudnovsky et al. (2006) find that such relationships are significant requires confirmed qualification of human resources (manager’s
for Argentina. Furthermore, in another model tested for six Latin experience, promotion of training) and continuous technological
American countries (Argentina, Chile, Colombia, Costa Rica, Panama intelligence.
and Uruguay), Crespi and Zuñiga (2010) produced divergent results
depending on the type of innovation. They show that there is a In addition to ownership capacity and the qualification of human
significant positive effect on innovations in products and processes resources, it also appears that the investment climate and institutional
on a firm’s productivity level in the six countries of the sample, while framework in general influence innovation incentives and thereby affect

8
In Tunisia and Morocco), exports are produced mainly from foreign capital used in low-cost, low value- added off-shore production activities (ANIMA 2011).
9
www.wipo.int/ipstats/fr/statistics/patents/

6
A f r i c a n D e v e l o p m e n t B a n k
 Economic Brief
AfDB
2 0 1 4 • w w w . a f d b . o r g

the firm’s productivity level. Regulatory inflexibility, bureaucracy and on access to financing, anti-competitive practices, deployment of
weak interface between the administration and firm are all obstacles informal activities, ineffectiveness of the intellectual property system,
which impede creativity, innovation and increased productivity crime, corruption and disturbances also discourage investment in
(Goedhuys (2007a)). Dabla-Norris et al. (2010) also show that constraints general and reduce the incentive to innovate in particular.

7
A f r i c a n D e v e l o p m e n t B a n k
 AfDB Economic Brief

2 0 1 4 • w w w . a f d b . o r g

II. Empirical Analysis for North African Countries

ur analysis concerns firms in three North African countries for product line over the previous two years and "0" if not.11
O which we have collected the data of the World Bank Investment
Climate Survey (WBICS). The sample comprises 1712 firms in 9 industrial The explanatory variables selected inform on the firm’s size and age,
10
sectors (Table 2) distributed among the three countries. total and export turnover, the proportion of foreign capital, the level
of qualifications of personnel, R&D intensity12 and other variables
II.1. Data and Descriptive Analysis measuring innovation, the use of ICTs and the perception of the
business climate, regulations, etc. As shown in Table 3 North African
In our analysis we have constructed a number of indicators (Annex I) for countries are characterized by the predominance of small- and
the analysis of innovation and productivity. Despite some differences in the averaged-sized family firms respectively representing 91%, 57% and
structure of questionnaires from one country to another and the non- 62% of the sample in Algeria, Egypt and Morocco.
availability of certain data for some countries, we have selected a harmonized
series of variables in a concern to ensure comparability of results. Concerning the variables which express the internal innovation effort,
we have selected the usual indicator approximated using R&D intensity
In the WBICS survey, the innovation indicator is measured by a (R&D/turnover). The data highlight the weakness of this ratio in North
qualitative dichotomous variable "INNOV" which takes the value of African countries as is the case in all developing countries. However,
"1" if the firm has introduced a new product line or improved an existing although this ratio is very low, it is fairly high for innovating firms.

Table 3: Descriptive Statistics

Algeria Egypt Morocco

Sample Size 291 962 459

Size (% total personnel)

Small (<20 employees) 42 34 19

Average (between 20 and 99 employees) 49 23 43

Large (over 100 employees) 9 43 38

Distribution by Sector (%)

Agro-food 31 15 25

Electronics 2 6 1

Machinery and Equipment 30 2 1

Metals - 16 -

Non-metal and plastic products - 23 1

Ready-to-wear 12 11 24

10
The data on services are not of good quality. Several fields are missing.
11
In the absence of such an indicator for Moroccan firms, we have retained R&D as a proxy innovation variable.
12
Except for Morocco.

8
A f r i c a n D e v e l o p m e n t B a n k
 Economic Brief
AfDB
2 0 1 4 • w w w . a f d b . o r g

Chemical and Pharmaceutical Products 7 10 20

Textiles 13 17 11

Other Manufacturing Sectors 5 - 17

Average Personnel Employed Total

All firms 39 326 140

Innovating Firms 39 617 182

Non-innovating firms 40 183 127

Proportion of qualified human resources in employment

All firms 46 45 42

Innovating firms 48 42 40

Non-innovating firms 45 46 42

Manager’s Average Number of Years of Experience

All firms 19.6 17.2 21.9

Innovating firms 19.2 18.1 23.1

Non-innovating firms 19.9 16.8 21.5

Vocational Training (%)

All firms 19 23 25

Innovating firms 22 42 53

Non-innovating firms 17 13 16

Foreign Participation (%)

All firms 2 8 23

Innovating firms 4 11 33

Non-innovating firms 0.6 6 20

Exports (%)
All firms 5 36 46

Innovating firms 4.8 58 50

Non-innovating firms 4.8 25 45

R&D spending /Turnover (%)

All firms 0.55 0.43 -

Innovating firms 0.84 0.54 -

Non-innovating firms 0.42 0.07 -

Licenses (%)
All firms 13 9 13

Innovating firms 13 18 25

Non-innovating firms 13 5 9

Certification (%)
All firms 12 29 23

Innovating firms 14 51 49

Non-innovating firms 10 18 15

ICT (%)
All firms 68 47 80

Innovating firms 71 69 96

9
A f r i c a n D e v e l o p m e n t B a n k
 AfDB Economic Brief

2 0 1 4 • w w w . a f d b . o r g

To measure human capital qualifications we retained the quality of human we have selected two indicators to express a firm’s openness: on the
resources approximated using 3 indicators: manager’s experience one hand, export intensity (EXPO) given by the ratio of exports to
(MAN_EXP), proportion of qualified staff (SKILL) in the total number turnover and, on the other, the importance of foreign investment given
of personnel employed and an indicator for vocational training provided by the proportion of foreign capital (FOR) in the firm’s total capital.
by the firm for its employees (TRAIN). The bivariate analysis shows that
for Algeria, the proportion of qualified personnel is higher in innovating Finally, we have retained perception variables which reflect the obstacles
firms though the difference is not significant in relation to non-innovating encountered in the 3 countries. These indicators concern:
firms. However, for Egypt and Morocco, the result is absurd given that
the proportion of qualified human resources is higher in non-innovating • The quality of infrastructure and public services
firms. For the vocational training variables, the proportions are relatively (telecommunications, transport, administrative services, etc.);
higher for innovating firms in the three countries while, with regard to • The business climate (anti-competitive practices, informal sector,
manager’s experience, the superiority of innovating firms is noted for corruption, etc.);
Morocco but not for Algeria. • The availability of skills and qualifications; and
• Access to financing
Furthermore, in order to assess the absorptive and learning capacities,
we retained the quality certification indicator (CERT) and the technological II.2. Empirical Model and Estimation Method
ownership indicator as measured by the purchase of patent exploitation
licenses (PEL). The data show that the proportion of firms using certification In order to estimate the impact of innovation on firms’ performance in
is fairly high for innovating firms in the 3 countries while for licenses, the the three North African countries we have adopted a two-stage model.
superiority of innovating firms is noted for Morocco and Egypt but not for Initially, (Annex II), we test13 the significance of the traditional determinants
Algeria. Other variables such as the use of ICT have been introduced in of a firm’s innovation (size, R&D, quality of human capital) and its learning
order to control the firm’s ability to assume ownership of technologies and ownership capacity (license, certification, technological) by
and knowledge by using new tools and communication techniques. demonstrating the expected impacts of economic openness through
exports and participation of foreign capital.
In addition to ownership indicators, taking into account the level of
economic openness as a factor likely to affect incentives to innovation Secondly, we envisage measuring the impact of innovation and its main
and productivity in North African countries is relevant for this analysis. significant determinants on firms’ productivity (Annex II). In a first model,
Indeed, over the past two decades, North African countries (especially we estimate the direct impact of innovation on productivity taking into
Tunisia – Morocco and Egypt and to a lesser extent Algeria) have account the variable indicators of the business climate and the quality
implemented trade policies to encourage free trade with Europe and of public services on the one hand, and economic openness indicators,
strategies to attract foreign direct investment with a view to fostering on the other. By contrast, in the second model productivity is directly
competitiveness and innovation. The objective was to encourage explained by variables which characterize the firm (R&D, training, size,
the local productive system to improve its price and non-price learning, etc..) while considering the same control variables (business
competitiveness on the domestic and foreign markets under pressure environment, quality of infrastructure and public services and participation
of competition from European partners. Local firms would also benefit of foreign capital). The objective is to test the direct impact of the different
from possible technological externalities from the activities of foreign factors related to the adaptation and assimilation of knowledge and
investors in the host countries. Thus, for the requirements of this analysis, technologies on firms’ performance.14

13
The innovation equation is estimated by the maximum likelihood method using the Logit model.
14
This second model avoids the problem of endogeneity which could arise when estimating the relationship between productivity and innovation (efficient firms could be more inno-
vating). The model is estimated using the instrumental variable method in order to avoid the estimation bias caused by endogeneity.

10
A f r i c a n D e v e l o p m e n t B a n k
 Economic Brief
AfDB
2 0 1 4 • w w w . a f d b . o r g

III. Main Results

III.1. The Determinants of Innovation which target both the European markets and other international
markets in which the pressures of competition respond more to the
he estimation of the innovation equation shows that the model is requirements of structural non-price competitiveness. Firms are
T not significant for Algeria whereas it is for Morocco and Egypt
(Annex III, Table 1) with similarities concerning the effect of the main
required to adopt increasingly sophisticated technologies in order to
meet the requirements of foreign markets. In the case of Morocco,
determinants on innovation. As in the case of developed countries, the the statistically insignificant effect of exports could be explained by
size of the firm appears to play a key role in the innovation process15 the specificity of the country’s industrial fabric and the channeling of
while R&D intensity does not seem to be a determining variable of exports towards low value adding18 and low-cost activities.
innovation in these countries. The innovations are minor and incremental
and do not require intensive R&D activities.16 Finally, the results highlight the non-significance of the participation of
foreign capital as an incentive for innovation for Egyptian and Moroccan
A surprising result stresses the non-significance of human resource firms. This result has been pointed out in several studies conducted
qualifications as measured by the proportion of employees with a in other developing countries19 and reflects the specialization of foreign
high level of education unlike the prediction of some other empirical direct investment targeting North African countries in activities with
works concerning developing countries. 17 This result could be low technological intensity. As we have already pointed out, both in
explained by the failure to optimize human capital in innovating Morocco and Tunisia, FDI are concentrated in low value –adding, sub-
activities. Indeed, it seems that human capital is channeled more contracting activities. These low cost and unskilled labour intensive
towards the performance of routine and ordinary tasks and less industrial activities are often driven by active foreign capital in the off-
towards creative and innovative activities. shore regime which is poorly integrated in the local economy and has
positioned itself as a first or second rank sub-contractor of European
On the other hand, the impact of vocational training on the dynamics principals.
of innovation is significant, which reflects the importance of upgrading
skills to build a firm’s capacity to adopt and absorb new know-how III.2. Innovation and Productivity
and knowledge which are vital to maintaining the momentum of
innovation even though it is minor. Furthermore, the other technological To measure the relationship between productivity and innovation (Annex
ownership indicators (licenses, certifications, use of ICT) have a II) we have retained20 in the productivity equation the traditional variables
significant impact and positively affect the firm’s innovation behavior. (the stock of physical capital per worker K and the level of use of
capacities21 U) in addition to the innovation variable, investment climate
With regard to the export effect on the incentive to innovate, the results indicators, the quality of public services and obstacles to the recruitment
show that Egyptian firms take advantage of openness to stimulate of qualified labour.
their innovations whereas for Morocco, the export effect on innovation
is not significant. This result could be explained by the types of The results of the estimation, (Annex III, Table 2), stress the positive
comparative advantages of the two countries. In Egypt, the effect of innovation in Egypt and Morocco whereas the relationship is
specialization of exports is increasingly focused on diversified products not significant for Algeria. The results also show the importance of

15
See Pamukcu, 2003, Ayyagari et al, 2007, Almeida and Fernandes, 2008, Yuriy et al, 2010.
16
Karray al. (2009), Koouba et al. (2010) and El Elj (2012) pointed out the same unexpected effect of the quality of human resources on the potential for innovation in Tunisia.
17
Pamucku (2003), Ayyagari et al. (2007) and Almeida and Fernandes (2008).
18
For the Tunisian case see El Elj (2012) and Rahmouni et al. (2010).
19
For the Tunisian case see El Elj, (2012), and Karray et al. (2009), and for the Turkish case see Pamucku (2003).
20
Labour productivity is one of the measures of a firm’s performance. Obviously, it would be more relevant to adopt other indicators such as value added or total factor. However,
such information is not available for most of the firms in our sample.
21
In order to control differences in capacity among firms.

11
A f r i c a n D e v e l o p m e n t B a n k
 AfDB Economic Brief

2 0 1 4 • w w w . a f d b . o r g

obstacles relating to the availability of qualifications as impediments of firms regarding the availability of qualifications is significantly negative
to the enhancement of productivity. The same applies to the poor (for Morocco and Egypt). These results reveal the inefficacy of the
performance of public services and infrastructure in the case of Morocco human resource allocation and call for the establishment of more
and informal competition, illegal practices and over-regulation in the effective education and training policies.
case of Algeria.
Finally, foreign participation in a firm’s capital is reflected in higher
These results are confirmed by the estimate of model 2 which brings productivity only in Egypt while for the innovation equation estimation,
all the determinants directly into play in the productivity equation this variable was not significant. This result confirms that foreign direct
(Annex III, Table 3). As in the case of innovation it seems that human investment could generate productivity gains without improving
resource qualifications pose a serious problem to firms in North African innovation potential. FDI in North Africa increasingly reflects a
countries. Indeed, for the three countries, the SKILL variable which positioning in response to the need to enhance productivity by reducing
measures employees’ level of qualifications is not significant whereas costs rather than by improving structural competitiveness which is
the effect of the SKILL-OBS variable which measures the perception synonymous with the drive to innovate and continually improve.

12
A f r i c a n D e v e l o p m e n t B a n k
 Economic Brief
AfDB
2 0 1 4 • w w w . a f d b . o r g

IV. Conclusions and Recommendations

n this article, we have tried to conduct a comparative analysis among To boost innovation and stimulate productivity, special attention should
I three North African countries in order to identify a profile of an
innovating firm and highlight the different internal and external factors
be paid to certain aspects of the national system of innovation. In
the present context, faced with competitive threats from a number of
influencing a firm’s ability to innovate. We also looked at the effects emerging countries, especially Asian countries, offering unbeatable
of innovation efforts on a firm’s performance while stressing the production costs in the traditional sectors, the North African countries
importance of qualifications and the investment climate in improving are now required to target a wider range of technologies requiring
productivity. Despite some divergences among the North African significant absorptive capacity and highly qualified labour to strengthen
countries, we have confirmed in this article that these countries share their non-price competitiveness in the traditional sectors and diversify
common characteristics regarding potential for, and obstacles to their productive activities towards other sectors with higher value addition.
innovation and their impact on productivity improvements.
First, it is necessary to create synergies among the different actors
We emphasized that innovation in North African countries is far from of the embryonic national system of innovation in order to initiate the
being the result of activities relating to R&D and the creation of new necessary reforms that will in time help to activate all the factors for
technologies alone. It is to be seen in the adoption and adaptation successful innovation. In this area, the reform options should aim to
of technologies created elsewhere. Indeed, in these countries where strengthen the governance of the national innovation system,
productive structures are dominated by small or medium-sized firms stimulate the research and corporate environments with a view to
with limited technological capacities and negligible R&D intensity, decompartmentalizing the two spheres by ensuring more efficient
technological adoption and adaptation are presented as effective and effective interfacing. Furthermore, it is urgent to develop the
alternative options for joining the innovation race. Innovation is not quality of programmes and projects of educational and research
always synonymous with the creation of disruptive technology institutions as well as to align them with the needs of firms, especially
or technology which introduces entirely new products. It may be in the new strategic areas. In this context, greater support should be
reflected by an up-market move and by increasing value addition for provided to the higher education system and vocational training with
existing products. However, the absorptive capacity depends, among a view to improving human resource qualifications to ensure a better
others, on learning mechanisms, human resource qualifications and contribution to the production and innovation processes.
managerial quality. However, our results show that in the North African
countries, especially in Morocco and in Egypt, the impact of qualified Next, the North African countries should reconsider the reforms and
human resources is statistically insignificant both on the incentive to strategies related to foreign direct investment. It is necessary to improve
innovate and at the productivity level which reveals underutilization the investment climate and support it by more advantageous incentives to
and an inefficient allocation of human capital in North Africa. innovating investors. New policies should target FDI in productive activities
with greater value addition which are sufficiently well integrated upstream
We have also identified other unexpected results concerning the effect and downstream of the local economy. Moreover, the strategy should
of economic openness on innovation potential. On the one hand, facilitate interface with the research environment and encourage incubator
innovation is statistically insignificant due mainly to the rigid structure of programmes in collaboration with public bodies and the local private sector.
comparative advantages in these countries and the concentration of
exports in sectors with low value added and limited technological potential Finally, strategies to attract international R&D relocated by multinational
especially in the case of Morocco. On the other hand, we have shown firms should boost the innovation systems of the North African countries.
a limited impact of foreign investment on the incentive to innovate Contractualization mechanisms, backed by the State, in the area of
since these investments target low-cost, low value-added activities science and technology between the research centres, universities,
focused on sub-contracting. Moreover, these investments are generally potentially innovative local firms and foreign firms wishing to relocate their
disconnected and poorly integrated into the local productive system R&D activities should strengthen the cohesion between the different
and consequently generate few upstream-downstream externalities. components of the national innovation system around innovating projects.

13
A f r i c a n D e v e l o p m e n t B a n k
 Economic Brief
AfDB
2 0 1 4 • w w w . a f d b . o r g

Annexe I : Description of Variables

Table 3: Descriptive Statistics

Variable Description
Dependent Variables

Dichotomous variable which takes the value of 1 if the firm is innovating and 0 if the firm is not
INNOV
innovating

Turnover divided by the average number of full-time wage-earners during the last financial year in
Log(CA /L)
logarithmic terms

Independent Variables
The difference between the year under investigation and the
Age
firm’s year of birth.
Size Dichotomous variable which takes the value of 1 if the firm’s size is between 20 and 99 employees and
0 if not.
MEDUIM

LARGE Dichotomous variable which takes the value of 1 if the firm’s size is above or equal to 100 employees.

FOR (share of foreign capital) Share of the firm’s capital held by a foreign company.

EXP (Export Intensity) Percentage of sales dedicated to exports

SKILL Percentage of qualified personnel in relation to total full-time personnel

TRAIN Percentage of full-time personnel receiving vocational training

MANEXP Manager’s number of years’ experience in the firm’s sector of activity

LIC Dichotomous variable which takes the value of 1 if the firm uses a technology under foreign license

CERT Dichotomous variable which takes the value of 1 if the firm has international certification

E-mail Dichotomous variable which takes the value of 1 if the firm uses e-mail as a means of communication

Dichotomous variable which takes the value of 1 if the firm uses the Web site as a means of
Web
communication

LOAN Dichotomous variable which takes the value of 1 if the firm has obtained a credit or a loan

Log (K/L) The net value of total assets (machinery, equipment, land and buildings,...) per worker (in Log)

Capacity utilization rate is the actual use of the firm’s production capacity in relation to the maximum
U
capacity rate.

Obstacle variable which takes the value of 0 if the lack of training and qualified personnel are not
SKILL_OBS considered as constraints for the firm and takes the values of 1 to 4 depending on the degree of difficulty
of these constraints (1: minor obstacle and 4: very serious obstacle)

Obstacle variable which takes the value of 0 if the lack of public services (communication, electricity and
BUSS_SUP_SER transport) is not considered as a constraint for the firm and takes the values of 1 to 4 depending on the
degree of difficulty of that constraint (1: minor obstacle and 4: very serious obstacle).

Obstacle variable which takes the value of 0 if unfair competition from the informal sector, anti-competitive
practices and access to investment licenses and permits are not considered as constraints for the
INVEST_CLIM
firm and takes the values of 1 to 4 depending on the degree of difficulty of these constraints (1: minor
obstacle and 4: very serious obstacle).

15
A f r i c a n D e v e l o p m e n t B a n k
 AfDB Economic Brief

2 0 1 4 • w w w . a f d b . o r g

Annex II : Econometric Models

II.1. The innovation equation b- Direct Estimation of the Effects of Innovation Determinants on
Productivity:
The innovation equation is expressed as follows:
Model 2: Yi = α K i + η U i + γ K Χ K i + δ I i + λ For + ε i
INNOVi = β0 + β1 Age + β2Taille + β3 Exp + β4 For + β5Loan + βK ΧK

Where X k, is the vector for k technological and learning variables Where,

for each firm i, presented in the previous sections. This equation is
estimated separately for each country using the binomial logit model. Y is the productivity of labour measured by turnover per employee
As we have already mentioned for Moroccan firms, innovation is (in log).
approximated by the R&D effort. K is the stock of physical capital per employee (in log)
U is capacity utilization rate
II.2. Productivity Equations INNOV is the innovation indicator (0 or 1)
Xk denotes all the explanatory variables measuring the effect of
The productivity equations based on a Cobb-Douglas type approximation technological and learning factors.
of the production function are presented in the following form: I denotes all the proxies of investment climate perception (anti-
competitive practices, corruption, etc..) and the quality of infrastructure
a- Estimation of the Effect of Innovation on Productivity: and public services.
FOR measures the share of foreign capital
Model 1: Yi = α K i + η U i + γ INNOVi + δ I i + λ For + ε i εi is the error term representing random disruptions

16
A f r i c a n D e v e l o p m e n t B a n k
 Economic Brief
AfDB
2 0 1 4 • w w w . a f d b . o r g

Annex III : Estimation Results

Table 1: Determinants of Innovation : Results of Logit Model

Algérie Égypte Maroc

-0.699 -1.815 -4.074
Constant
(-1.665)* (-6.154)*** (-5.701)***
-0.008 -0.006 0.009
AGE
(-0.761) (-1.296) (1.143)
0.227 0.878 1.244
MEDUIM
(0.828) (3.760) *** (2.379) **
-0.207 1.043 1.117
LARGE
(-0.410) (4.341) *** (1.989) **
0.032 -0.001 0.000
FOR
(1.688)* (-0.208) (0.085)
0.035 0.007 -0.002
EXP
(0.957) (2.179) ** (-0.589)
0.007 -0.004 -0.002
SKILL
(1.341) (-1.052) (-0.472)
-0.000 0.001 0.012
MAN_EXP
(-0.048) (0.194) (1.047)
-0.002 0.011 0.015
TRAIN_INT
(-0.291) (2.524) ** (2.814) ***
0.401 0.629 0.748
CERT
(0.929) (3.302) *** (2.504) **
-0.185 0.539 0.568
LIC
(-0.484) (2.048) ** (1.665)*
0.029 0.102 0.760
E_MAIL
(0.102) (0.392) (1.561)
0.027 0.435 0.555
WEB
(0.089) (1.694)* (1.963) **
0.091 0.065 0.486
LOAN
(0.346) (0.760) (1.800)*
R² Mc Fadden 0.03 0.16 0.21

% of correct predictions 60.13 73.08 80.84

LRT statistic (Likelihood Ratio Test) 11.58 (N.S) 189.03*** 104.97***

* Significant at 10%, ** Significant at 5%, *** Significant at 1%

• LRT (Likelihood Ratio Test) : to verify the overall significance of • Pseudo-R² Adjustment Quality: Mc Fadden’s R² measures the
our model, we applied the likelihood ratio test estimating on the adjustment quality of the model. Mc Fadden’s R² values reflect a
one hand the constrained model and on the other, the unconstrained good adjustment quality, especially for Egypt and Morocco.
model. The LRT statistic quite simply corresponds to the deviation • Multicollinearity Test (Variance Inflator Factor VIF): in order to
of log likelihood: The values of the LRT statistic are statistically verify possible collinearity between the explanatory variables we
significant for Egypt and Morocco. Whereas this value is non- calculated the Variance Inflation Factor, (VIF). A value below 10
significant in the case of Algeria which allows acceptance of the means that the collinearity between the variables is tolerable. In all
null hypothesis suggesting that all the estimated parameters are the estimates, collinearity seems unlikely to bias the model since
null except for the constant. the VIF value is fairly low.

17
A f r i c a n D e v e l o p m e n t B a n k
 AfDB Economic Brief

2 0 1 4 • w w w . a f d b . o r g

Table 2: Results of the Estimation of the Productivity Equation (Model 1)

Algeria Egypt Morocco

0.012 1.460 1.845
INNOV
(0.08) (4.22) *** (4.81)***
0.043 0.254 0.227
K
(2.72)*** (10.49) *** (7.16) **
0.009 -0.002 0.008
U
(2.15) ** (-0.77) (2.42)**
0.0001 0.010 -0.001
FOR
(0.02) (3.89) *** (-0.81)
-0.030 -0.116 -0.132
SKILL_OBS
(-0.47) (-3.53) *** (-2.40) **
-0.201 -0.003 -0.254
BUSS_SUP_SER
(-1.34) (-0.04) (-1.70)*
-0.217 0.118 0.033
INVEST_CLIM
(-2.64) ** (1.60) (0.47)
10.29 6.128 7.048
Constant
(25.17) *** (21.93) *** (16.75) ***

R² 0.069 0.117 0.125

F 3.03*** 28.83*** 21.61***

N (number of observations) 291 962 459

* Significant at 10%, ** Significant at 5%, *** Significant at 1%

Instrumental Variables (IV) Approach: • The F statistic of the first stage of the estimation validates the re
levance of our instruments. The consistency of our instruments is
• The Durbin-Wu-Hausman test led to the rejection of the null confirmed especially in the case of Egypt and Morocco. For Al
hypothesis suggesting that the INNOV variable is exogenous and geria, the OLS estimator remains efficient and consistent and the
confirms the consistency of the IV estimators endogeneity of the INNOV variable does not appear to affect the
• The Sargan Test results in the acceptance of the null hypothesis results of the estimations.
(the instruments are not correlated with the residues).

18
A f r i c a n D e v e l o p m e n t B a n k
 Economic Brief
AfDB
2 0 1 4 • w w w . a f d b . o r g

Table 3: Estimation Results of the Productivity Equation (Model 2)

Algeria Egypt Morocco

0.036 0.266 0.258
K
(2.36) *** (11.52) *** (7.42) ***
0.006 0.002 0.009
U
(1.91) ** (0.92) (2.92) ***
-0.002 0.011 -0.001
FOR
(-0.25) (4.01) *** (-0.76)
-0.003 -0.002 -0.001
SKILL
(-0.87) (-0.92) (-0.67)
-0.005 0.003 0.007
TRAININT
(-1.17) (1.17) (2.82) ***
-0.006 -0.011 0.003
MANEXP
(-1.13) (-2.39) ** (0.56)
0.192 0.207 0.798
CERT
(0.80) (1.52) (5.09) ***
0.376 0.141 0.061
LIC
(1.72)* (0.74) (0.35)
0.502 0.287 0.133
EMAIL
(3.00) *** (1.65) (0.90)
-0.098 0.037 -0.250
WEB
(-0.56) (0.21) (-1.7)
-0.232 -0.103 -0.139
SKILL_OBS
(-1.57) (-3.26) *** (-2.93)**
-0.017 -0.024 -0.267
BUSS_SUP_SER
(-0.27) (-0.33)*** (-1.74)**
-0.216 0.132 0.033
INVEST_CLIM
(-2.67) *** (1.85) ** (0.54)
10.283 6.247 6.865
Constant
(23.43) *** (20.82) *** (13.87) ***

R² 0.128 0.193 0.329

F 3.12*** 15.87*** 18.12***

N (number of observations) 291 962 459

* Significant at 10%, ** Significant at 5%, *** Significant at 1%

19
A f r i c a n D e v e l o p m e n t B a n k
 AfDB Economic Brief

2 0 1 4 • w w w . a f d b . o r g

Bibliographical References

Almeida R., Fernandes A. M. (2008), Openness and technological innovations in developing countries: evidence from firm-level surveys,
Journal of Development Studies 44:701–727.

ANIMA Investment Network (2010), The Mediterranean between growth and revolution: Foreign Direct Investments and Partnership in MED
countries in 2010, Study N° 21, March 2010.

Arab Human Development Report (2010).

Arvanitis R., Boisard A.S., Clemente R.R., and Pasimeni P. (2007), Toward a Euro Mediterranean Innovation Space: Ideas for research and
Policy Making.

Arvanitis R., Mhenni H., (2010), ‘Monitoring Research and Innovation Policies in the Mediterranean Region’, Science Technology & Society
15(2):233-69.

Ayadi R., De Wulf L., El Elj M., Ghoneim M., Gasiorek M., Gonzales J. L., Guerin S., Holmes P., Kassal H., Maliszewska M., Tekce M.,
(2009), ‘Economic Integration in the Euro Mediterranean Region’, CASE Network Report.

Ayyagari, M., Demirguc-Kunt, A., and Maksimovic V. (2007), ‘Firm Innovation in Emerging Markets’, World Bank Policy Research Working
paper 4157.

Benavente, J. M. (2006), The Role of Research and Innovation in Promoting Productivity in Chile, Economics of Innovation and New Technology,
15(4-5), 301-315.

Bogliacino, F., Perani, G., Pianta, M., and Supino, S. (2009), ‘Innovation in Developing Countries: The Evidence from Innovation Surveys’,
paper prepared for the FIRB conference Research and Entrepreneurship in the Knowledge-based Economy, Milano: University L. Bocconi.

Chudnovsky, D., López, A., and Pupato G. (2006), Innovation and Productivity in Developing Countries: A Study of Argentine Manufacturing
Firms Behavior (1992-2001), Research Policy, 35, 266-288.

Crépon, B., Duguet, E., and Mairesse, J. (1998), ‘Research, innovation and productivity: an econometric analysis at the firm level’, Economics
of Innovation and New Technology, vol. 7, pp. 115-158.

Crespi, G., and Zuniga, P. (2010), ‘Innovation and Productivity: Evidence from six Latin American countries’, IDB Working papers series - IDB-WP-218.

Dabla-Norris E., Kersting E., and Verdier, G. (2010), ‘Firm productivity, Innovation and financial development’, IMF Working paper.

El Elj M. (2012), Innovation in Tunisia Empirical Analysis for Industrial Sector, Journal of Innovation Economics, 2012/1 (N° 9).

Global Competitiveness Report (2010-2011), World Economic Forum.

http://www3.weforum.org/docs/WEF_GlobalCompetitivenessReport_2010-11.pdf

20
A f r i c a n D e v e l o p m e n t B a n k
 Economic Brief
AfDB
2 0 1 4 • w w w . a f d b . o r g

Griffith, R. et al. (2006), ‘Innovation and Productivity across Four European Countries’, Oxford Review of Economic Policy 22(4): 483-498.

Goedhuys, M. (2007a), ‘Learning, Product Innovation, and Firm Heterogeneity in Developing. Countries: Evidence from Tanzania, Industrial
and Corporate Change’ 16: 269-292.

Goedhuys, M. (2007b), ‘The Impact of Innovation Activities on Productivity and Firm Growth: Evidence from Brazil’, UNU-MERIT Working Paper -002.

Goedhuys, M., N. Janz, and Mohnen P. (2008), ‘Knowledge-based Productivity in Low-tech Industries: Evidence from Firms in Developing
Countries’, UNU-MERIT Working paper Series 007.

Hall, B. H. (2011), ‘Innovation and productivity’, UNU-MERIT Working papers, ISSN 1871-9872.

Karray, Z., and Kriaa, M. (2009), ‘Innovation and R&D investment of Tunisian firms: a two-regime model with selectivity correction’, Economic
Research Forum working paper No 484.

Koubaa, K., M’henni, H., and Gabsi, F. (2010), ‘Innovation determinants in emerging economies: an empirical study based on an innovation
survey data in Tunisia’, Int. J. Technological Learning, Innovation and Development, Vol. 3, No 3, pp.205–225.

Lopes, L., and Godinho, M. (2005), ‘Services Innovation and Economic Performance: An analysis at the firm level’, DRUID Working papers.

Mairesse, J., and Mohnen, P. (2010), ‘Using innovation surveys for econometric analysis’, UNU-MERIT Working Papers, 2010-023.

Pamukcu, T. (2003), ‘Trade Liberalization and Innovation Decisions of Firms: Lessons from Post-1980 Turkey, World Development’, Elsevier,
vol. 31(8), pages 1443-1458.

Polder, M., Van Leeuwen, V., Mohnen, M., and Raymond, W. (2010), ‘Product, Process and Organizational Innovation: Drivers, Complementarity
and Productivity Effects’, DRUID Working paper No 10-24.

Rahmouni, M., Ayadi, M., and Yıldızoglu, M. (2010), Characteristics of innovating firms in Tunisia: The essential role of external knowledge
sources, Structural Change and Economic Dynamics.

World Economic Forum (2010), Global Competitiveness Report 2010-2011.

Yuriy, G., Svejnar, J., and Terrell, K. (2010), Globalization and Innovation in Emerging Markets, American Economic Journal: Macroeconomic,
2(2): 194–226.

21
A f r i c a n D e v e l o p m e n t B a n k
AfDB Economic Brief

2 0 1 4 • w w w . a f d b . o r g

22
A f r i c a n D e v e l o p m e n t B a n k
 Economic Brief
AfDB
2 0 1 4 • w w w . a f d b . o r g

23
A f r i c a n D e v e l o p m e n t B a n k
 © 2010 - AfDB - Design, External Relations and Communication Unit/YAL

A
f
r
i
c
a
n
D
e
v
e
l
o
p
m
e
n
t
B
a
n
k