How does exchange rate policy affect manufactured exports in MENA countries?

Abstract

This paper shows that, during the 1970s and 1980s, MENA economies were characterized by a significant overvaluation of their currency. This overvaluation has had a cost in terms of competitiveness. To determine the degree of overvaluation of the MENA currencies, an indicator of misalignment was developed based on the estimation of an equilibrium exchange rate (Edwards, Exchange Rate Misalignment in Developing Countries, The Johns Hopkins University Press, Baltimore, 1988). The empirical work was based on a panel of 53 developing countries, ten of which are MENA economies. Although overvaluation decreased in the 1990s, probably due to flexibilization of the exchange rate regime in some MENA countries and to better macroeconomic management in others, misalignment remained higher than in other regions. This may be explained by the MENA countries' delay in adopting more flexible exchange rates, as well as in reforming their economies. In terms of competitiveness, the estimation of an export equation has shown that manufactured exports have been significantly affected by the overvaluation of the MENA currencies. Countries that already had a more diversified economy benefited more from the decreased overvaluation in the 1990s. These countries also saw a continuous rise in diversification of their manufactured exports, resulting from the significant decline in exchange rate misalignment.

Full text

Better Governance and Deeper Reforms
in the Middle East and North Africa
Mustapha Kamel Nabli
Breaking the Barriers
to Higher Economic Growth
43969

Breaking the Barriers
to Higher Economic Growth

Better Governance and Deeper Reforms
in the Middle East and North Africa
Mustapha Kamel Nabli
Breaking the Barriers
to Higher Economic Growth

© 2007 The International Bank for Reconstruction and Development / The World Bank
1818 H Street NW
Washington DC 20433
Telephone: 202-473-1000
Internet: www.worldbank.org
E-mail: feedback@worldbank.org
All rights reserved
2 3 4 5 11 10 09 08
This volume is a product of the staff of the International Bank for Reconstruction and
Development / The World Bank. The findings, interpretations, and conclusions expressed
in this volume do not necessarily reflect the views of the Executive Directors of The World
Bank or the governments they represent.
The World Bank does not guarantee the accuracy of the data included in this work. The
boundaries, colors, denominations, and other information shown on any map in this work
do not imply any judgement on the part of The World Bank concerning the legal status of
any territory or the endorsement or acceptance of such boundaries.
Rights and Permissions
The material in this publication is copyrighted. Copying and/or transmitting portions or
all of this work without permission may be a violation of applicable law. The International
Bank for Reconstruction and Development / The World Bank encourages dissemination of
its work and will normally grant permission to reproduce portions of the work promptly.
For permission to photocopy or reprint any part of this work, please send a request with
complete information to the Copyright Clearance Center Inc., 222 Rosewood Drive,
Danvers, MA 01923, USA; telephone: 978-750-8400; fax: 978-750-4470; Internet:
www.copyright.com.
All other queries on rights and licenses, including subsidiary rights, should be addressed
to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433,
USA; fax: 202-522-2422; e-mail: pubrights@worldbank.org.
Cover photo: © Henk Badenhorst/iStockphoto
Library of Congress Cataloging-in-Publication Data
Breaking the barriers to higher economic growth : better governance and deeper
reforms in the Middle East and North Africa / Mustapha Kamel Nabli ... [et al.].
p. cm.
ISBN 978-0-8213-7415-3
1. Middle East—Economic conditions. 2. Africa, North—Economic conditions.
3. Middle East—Economic policy. 4. Africa, North—Economic policy.
5. Economic development—Political aspects—Middle East. 6. Economic
development—Political aspects—Africa, North. 7. Macroeconomics. I. Nabli,
Mustapha K.
HC415.15.B74 2008
338.956—dc22
2007047530

Preface vii
Acknowledgments xi
Part I Growth, Reform, and Governance 1
1 Long-Term Economic Development
Challenges and Prospects for the Arab Countries 3
Mustapha K. Nabli
2 Reform Complementarities and Economic
Growth in the Middle East and North Africa 29
Mustapha Kamel Nabli and Marie-Ange Véganzonès-Varoudakis
3 After Argentina: Was MENA Right to Be Cautious? 77
Mustapha K. Nabli
4 Restarting Arab Economic Reform 89
Mustapha K. Nabli
5 Democracy for Better Governance and Higher
Economic Growth in the MENA Region? 103
Mustapha K. Nabli and Carlos Silva-Jáuregui
6 The Political Economy of Industrial Policy in the
Middle East and North Africa 135
Mustapha K. Nabli, Jennifer Keller, Claudia Nassif,
and Carlos Silva-Jáuregui
Part II Labor Markets And Human Capital 167
7 The Macroeconomics of Labor Market
Outcomes in MENA 169
Jennifer Keller and Mustapha K. Nabli
v
Contents

8 Challenges and Opportunities for
the 21st Century 203
Mustapha Nabli
9 Labor Market Reforms, Growth, and
Unemployment in Labor-Exporting Countries
in the Middle East and North Africa 211
Pierre-Richard Agénor, Mustapha K. Nabli, Tarik Yousef,
and Henning Tarp Jensen
10 Economic Reforms and People Mobility for
a More Effective EU-MED Partnership 267
Ishac Diwan, Mustapha Nabli, Adama Coulibaly, and
Sara Johansson de Silva
Part III Trade, Competitiveness, and Investment 297
11 Cruise Control, Shock Absorbers, and
Traffic Lights 299
Mustapha K. Nabli
12 Trade, Foreign Direct Investment, and
Development in the Middle East and
North Africa 305
Farrukh Iqbal and Mustapha Kamel Nabli
13 Making Trade Work for Jobs 329
Dipak Dasgupta, Mustapha Kamel Nabli, Christopher Pissarides,
and Aristomene Varoudakis
14 Exchange Rate Management within the
Middle East and North Africa Region 355
Mustapha Nabli, Jennifer Keller, and Marie-Ange Véganzonès
15 How Does Exchange Rate Policy Affect
Manufactured Exports in MENA Countries? 381
Mustapha Kamel Nabli and Marie-Ange Véganzonès-Varoudakis
16 Public Infrastructure and Private Investment
in the Middle East and North Africa 399
Pierre-Richard Agénor, Mustapha K. Nabli, and Tarik M. Yousef
17 Governance, Institutions, and
Private Investment 423
Ahmet Faruk Aysan, Mustapha Kamel Nabli, and Marie-Ange
Véganzonès-Varoudakis
Index 465
Contentsvi

vii
The world’s attention to the countries of the Middle East and North Africa1
(MENA) region has often been dominated and held captive by headline issues.
Much of that attention has centered on instability and uncertainty in the region,
supported through coverage of both recent and long-standing conflicts. More
recently intermixed with this perspective on the region is the world’s growing
attention to rising oil prices, from which another picture of the region emerges—
one of strategically well-positioned oil producing economies with rapidly
expanding wealth (and influence).
But little of the international coverage devoted to the region considers the
broad range of development challenges facing this diverse group of countries.
Providing quality jobs to a rapidly increasing labor force and reducing poverty,
promoting the private sector, expanding gender equity, improving education
access and quality, and effectively managing scarce water resources and finite oil
wealth are only a few of the challenges facing the countries of this region. Because
the MENA region’s development challenges are multiple and complex, meeting
them requires ambitious and coherent policy and institutional reforms. Through
knowledge sharing and policy-relevant research, the World Bank’s Middle East
and North Africa region’s Office of the Chief Economist has sought over the last
few years to deepen understanding of these issues and help promote effective
strategies for confronting these challenges. This book presents some of the work
undertaken in this context.
The greatest challenge on which this book centers is that of implementing a
comprehensive reform agenda to “break the barriers” to higher economic growth
and create needed job opportunities. It is an agenda that requires realigning the
MENA economies, to move from public sector-dominated to private sector-
driven economies, from closed economies to more open economies, and from oil-
dominated and volatile economies to more stable and diversified economies. The
central message from this body of work is that deeper economic reforms are need-
Preface

ed to attain higher growth rates and that better governance is key in achieving the
required progress on reform. In Part I of this book, we explore two major themes
related to the economic reform and growth nexus. First, we attempt to understand
the factors which caused the growth collapse of the 1980s and the failure of reforms
undertaken since then to spur a growth revival until the early 2000s. Second, we
delve into the domain of the political economy of reform, in order to understand
the reasons behind MENA’s slow progress with implementing economic reforms.
The core of this work points to the central role that public governance has played
in hindering the region’s progress with implementing the extensive reform agenda.
In “Long-Term Economic Development Challenges and Prospects for the Arab
Countries” I provide a broad overview of the development challenges in the region,
examining the constraints of the old models of development and defining the set
of transitions that constitute the contours of a new development model. In
“Reform Complementarities and Economic Growth in the Middle East and North
Africa,” (with Marie-Ange Veganzonès-Varoudakis) a quantitative analysis of the
relationship between economic reforms and economic growth in the Middle East
and North Africa region is undertaken, in an effort to understand whether the ane-
mic growth over the 1990s was the result of insufficient economic reforms or a
poor growth dividend from reform. “After Argentina: Was MENA Right to Be
Cautious?” looks at Argentina’s experience with reform and its subsequent eco-
nomic crisis and draws lessons for MENA on the strategic directions with respect
to speed, consistency and scope of reforms so as to avoid such crises. In “Restarting
Arab Economic Reform,” the difficulties with implementing deeper economic
reform are explored, and the reasons behind the stalled reform agenda in the region
and ways in which to move it forward are discussed. In “Democracy for Better
Governance and Higher Economic Growth in the MENA Region?” (with Carlos
Silva-Jáuregui) we review the literature on the relationship between economic
growth and democracy, and explore the implications for MENA, in order to under-
stand whether democratic development could help or hinder a stronger economic
performance in the MENA region, and under what conditions democracy leads to
better governance. Finally, in “The Political Economy of Industrial Policy in the
Middle East and North Africa,” (with Jennifer Keller, Claudia Nassif, and Carlos
Silva-Jáuregui) the specific topic of the region’s experience with industrial policy is
explored, to understand the political economy factors behind the persistence of a
traditional model of government intervention, and to discuss the likely direction of
new models of “industrial policy” in the region.
A central motivation for the economic growth agenda in the MENA region is
the challenge it has been facing over the last two decades—and which it will con-
tinue to face over the next two decades—in terms of job creation. This challenge is
unprecedented in the world and throughout history, with the region experiencing
rates of growth of the labor force of 3–4 percent a year over several decades. The
members of MENA’s labor force, of which a growing proportion are women, are
better educated than their parents and increasingly aware of economic opportuni-
Prefaceviii

ties worldwide, and they have greater expectations for good jobs. In Part II, the
extent of this development challenge is explored, along with the links between
labor market outcomes and various reforms. The first paper on “The
Macroeconomics of Labor Market Outcomes in MENA: How Growth Has Failed
to Keep Pace with a Burgeoning Labor Market” (with Jennifer Keller) reviews the
trends and developments in labor markets in MENA over the 1990s and analyzes
the links between these disappointing outcomes and the weak economic growth
experience (which is the subject of Part I) and private investment (which is the
subject of Part III). In “Challenges and Opportunities for the 21st Century:
Higher Education in the Middle East and North Africa,” the labor/education rela-
tionship is addressed, looking at ways in which the MENA region can best
improve the social and private returns from its higher education systems in an
increasingly globalized world. In “Labor Market Reforms, Growth and
Unemployment in Labor-Exporting Countries in the Middle East and North
Africa,” (with Pierre-Richard Agénor, Henning Tarp Jensen, and Tarik Yousef) a
quantitative assessment is made of the impact of labor market reforms in a typi-
cal labor-exporting MENA country on growth, real wages, and unemployment.
Finally,in “Economic Reforms and People Mobility for a More Effective EU-MED
Partnership,” (with Ishac Diwan, Adama Coulibaly, and Sara Johansson de Silva)
the developments in labor markets in the Southern Mediterranean countries are
linked to the potential for managed migration and people mobility with the
European Union (EU), and the role it may play in supporting the reform agenda
in these countries.
The overall thrust of the analysis in Parts I and II points to one critical mes-
sage: In order to break the barriers to higher growth and meet the employment
challenge, the power of the private sector in MENA countries needs to be
unleashed—through greater openness and integration with the world economy
and greater competitiveness. Part III of the book looks in more detail at a number
of issues relating to trade reform, the enhancement of competitiveness, and the
growth of private investment in the region. In “Cruise Control, Shock Absorbers,
and Traffic Lights: The Macroeconomic Road to Arab Competitiveness,”I discuss
the key macroeconomic policy elements needed for a supportive business envi-
ronment in MENA, a business environment that facilitates long-term planning,
promotes investment and knowledge transfer, and supports the efficient alloca-
tion of resources throughout the economy. In “Trade, Foreign Direct Investment,
and Development in the Middle East and North Africa,” (with Farrukh Iqbal) we
discuss how the limited progress in trade reform and weak foreign direct invest-
ment, in interaction with a weak domestic investment climate, contributed to the
region’s weak growth performance. In “Making Trade Work for Jobs: International
Evidence and lessons for MENA,” (with Dipak Dasgupta, Christopher Pissarides,
and Aristomene Varoudakis) the relationship between international trade and
employment in manufacturing is examined, to analyze the prospects for stepping
up employment growth in the region through trade and investment policy
ixPreface

reform. In “Exchange Rate Management within the Middle East and North Africa
Region: The Cost to Manufacturing Competitiveness,” (with Jennifer Keller and
Marie-Ange Veganzonès-Varoudakis) the repercussions from exchange rate mis-
alignment in the region are discussed, and the reasons behind the region’s contin-
ued reliance on fixed exchange rates analyzed. Further exploration of the costs of
exchange rate misalignment is presented in “How Does Exchange Rate Policy
Affect Manufactured Exports in MENA Countries?” (with Marie-Ange
Veganzonès-Varoudakis). In “Public Infrastructure and Private Investment in the
Middle East and North Africa,” (with Pierre-Richard Agénor and Tarik Yousef) we
assess quantitatively the impact of public infrastructure on private capital forma-
tion in three MENA countries: Egypt, Jordan, and Tunisia. Finally, Part III con-
cludes with a paper on “Governance, Institutions, and Private Investment: An
Application to the Middle East and North Africa,” (with Ahmet Faruk Aysan and
Marie-Ange Veganzonès-Varoudakis) which assesses the critical roles that poor
governance and weak institutions have played in determining the region’s low lev-
els of private investment.
While major economic and political developments have an impact on the
short-term policies and priorities for the MENA region, the core of the develop-
ment agenda for the next two or three decades remains largely intact. The chal-
lenges of achieving significantly higher economic growth, of reducing poverty, of
improving the quality of education and skills, of creating the hundreds of millions
of good jobs for a rapidly growing male and female labor force, of building strong
businesses and a healthy private sector, remain through the vicissitudes of oil price
fluctuations, conflict, and political turmoil. It is hoped that the collective thinking
represented in this book contributes in some modest way to highlighting and
meeting these challenges.
Mustapha K. Nabli
1. The World Bank formally defines the Middle East and North Africa region as including the follow-
ing countries: Egypt, Lebanon, Jordan, Tunisia, Morocco, Djibouti, and West Bank and Gaza (classi-
fied as resource-poor labor-abundant economies); Iran, Iraq, Algeria, Syria and Yemen (classified as
resource-rich labor-abundant economies); and Saudi Arabia, United Arab Emirates, Kuwait, Qatar,
Bahrain, Oman, and Libya (classified as resource-rich labor-importing economies).
x Preface

xi
This book includes many of the papers and speeches I have published and pre-
sented over the last few years in various fora as Chief Economist for the Middle
East and North Africa (MENA) region at the World Bank. The Office of the Chief
Economist is responsible for monitoring, analyzing, and reporting on the MENA
region’s economic developments, presenting new research, and enriching the
knowledge and policy debate among development practitioners, both within the
MENA region and worldwide.
My greatest debt and gratitude goes first to Jennifer Keller, who in addition to
co-authoring a few of the papers, has diligently and efficiently put together this
volume. She also helped me draft and provided background work for many of the
other papers and speeches included. The book is also the result of the collective
effort of my colleagues who co-authored many of the papers, including Pierre-
Richard Agénor, Ahmed Faruk Aysan, Adama Coulibaly, Dipak Dasgupta, Ishac
Diwan, Farrukh Iqbal, Henning Tarp Jensen, Claudia Nassif, Christopher
Pissarides, Sara Johansson de Silva, Carlos Silva-Jáuregui, Aristomene Varoudakis,
Marie-Ange Veganzonès-Varoudakis, and Tarik Yousef. Finally, I am grateful to
the wonderful research assistance of Paul Dyer, Manuel Felix, Adama Coulibaly,
Nihal Bayraktar, and Claudia Nassif, which has enabled this work.
Acknowledgments

Part I
Growth, Reform, and Governance

3
Long-Term Economic Development
Challenges and Prospects for
the Arab Countries
Mustapha K. Nabli
1
It is a challenge to talk about an Arab economy and to explore its prospects.
The countries that compose the Arab world are a diverse set in terms of size,
geography, level of income, natural resource endowments, economic struc-
ture, human capital and skills, social structures, economic policies and insti-
tutions, and more. Yet, the economic similarities among the countries in the
Arab world abound. The region has been linked by a common resource base.
Oil provided the basis for rapid economic and social development through-
out the region—not only for oil-producing economies but for resource-poor
Arab economies as well, through labor remittances and aid flows. The com-
mon resource base also includes the shared lack of water resources, with water
per capita in the Arab countries the lowest in the world. The region has been
linked by policy, with similar models of economic development adopted by
Arab countries at the time of their independence. Almost all adopted, since
the 1950s and 1960s, models of development based upon state-led planning,
with social policies designed for redistribution and equity.
With these strong similarities in terms of economic policy, natural
resources, and the shared production base for economic growth and develop-
ment, it is not surprising, then, that many of the development challenges fac-
ing the Arab countries today are also similar. Almost all have confronted stag-
Presented at Second Arab Thought Foundation Conference “The Prospects of the Arab Future”;
Beirut, Lebanon; December 4–6, 2003. Originally titled “Long-Term Economic Development
Challenges and Prospects for Arab Countries.”

Breaking the Barriers to Higher Economic Growth4
nant growth since the decline in oil prices, and despite some measured eco-
nomic reforms in most of the economies, growth has remained weak through-
out the region. All have been affected by regional conflicts and instability, either
directly or through association, deterring investment. And almost all of the
Arab states are facing one of the most pressing development problems to date:
a burgeoning problem of unemployment, the result of both shrinking
prospects for the main modes of employment creation in the past; labor migra-
tion and public sector employment; and a rapidly expanding labor force.
The Arab region faces a growing realization that the development paths of
the past are no longer capable of achieving national objectives. The problem
of insufficient job creation in the Arab region is mounting, and without fun-
damental transitions in the Arab economies to ensure greater and sustainable
job creation, the employment challenge will worsen rapidly and dramatically.
Over the next pages, we will articulate the broad course of action needed in
the Arab economies.1First we describe the development challenges facing the
Arab world, in particular the challenge of employment creation. Next, we
examine the constraints of the old development model in the region in terms
of meeting these development challenges. Then we propose a set of transitions
that constitute the contours of a new development model. Finally, we outline
the fundamental changes needed for making this transition, including
improved governance, higher-quality education, and greater gender equality.
The concluding section makes a few observations about feasibility and issues
of implementation.
A Central Challenge in the 21st Century: Employment
The Arab region faces many economic challenges as it enters the 21st century.
There is the challenge of water scarcity, and its implications for the availabili-
ty of this vital resource to its citizens and for development prospects. There is
the environmental challenge and the sustainability of the use of natural
resources, including soil degradation and deforestation, desertification, the
preservation of sea coasts, and deep-sea resources. There are issues of poverty
and exclusion. There is the issue of population growth, which has slowed dra-
matically in many countries, but remains high for many. But perhaps there is
no more pressing economic challenge for the Arab world today than that of
employment, on which this paper will focus.
A generation disappointed…One of the striking characteristics about the
Arab world2is the youth of its population. Two-thirds of the population is
under the age of 30, making it the second youngest region of the world behind
Sub-Saharan Africa. As a comparison, in Europe those under 30 constitute
only a third of the population. In East Asia and the Pacific, only half. The Arab
world is a world of burgeoning youth, increasingly educated, with higher
expectations than the generation before. Yet, at the same time, this group,

Part I: Growth, Reform, and Governance 5
which represents the new face of the Arab region, faces growing disappoint-
ment in the job market.
Labor market outcomes in the Arab region have steadily worsened over the
last two decades. With the unemployment rate increasing since the mid-
1980s, unemployment now averages over 15 percent of the labor force, by offi-
cial figures. Actual unemployment is probably much higher.
And unemployment falls disproportionately on the Arab region’s youth.
First-time job seekers make up more than 90 percent of all unemployed in
Egypt, and almost two-thirds of the unemployed in Yemen and the United
Arab Emirates.
These are workers coming into the labor force with substantially higher
levels of education than the generation before them: 20 years ago, the average
level of education in the Arab world was about two years; today, that level is
around fve years.3But unemployment has increasingly prevented those edu-
cational advancements from realizing economic returns in the labor market.
The Arab labor market is not only young, it is increasingly feminized, and
the poor labor market outcomes that characterize the Arab region particular-
ly affect women. While Arab women have the lowest labor force participation
rates in the world, their engagement in the labor force has grown considerably
over the last decades. But as their numbers increase, they are finding increas-
ingly fewer job opportunities.
Unemployment rates average 30 percent higher for women than their male
counterparts. The gender gap in unemployment is particularly great in coun-
Figure 1.1. Current Unemployment Rates in Arab Economies
United Arab Emirates
Kuwait
Saudi Arabia
Lebanon
Egypt
Oman
Syria
Yemen
Qatar
Bahrain
Jordan
Tunisia
Morocco
West Bank and Gaza
Algeria
unemployment (%)
35
30
25
20
15
10
1999 2003
1999 1997 2000 1996 2001 1999 2002 2001 2000
2001
2001
2002
2001
5
0
Source: Compiled by World Bank staff from ILO and country sources.
Note: Data include most recent estimates available for each country; rates in the GCC countries are for nationals only.

Breaking the Barriers to Higher Economic Growth6
tries like Bahrain, Syria, Egypt, and Saudi Arabia, where the unemployment
rate among women is two to three times that of males. Along with many other
factors that have altogether discouraged the involvement of Arab women in
the labor force, the poor prospects they face in the labor market have
undoubtedly deterred their participation.
Poor labor market outcomes are not limited only to unemployment, but
also to the poor wage prospects for the employed. Worker productivity, the
basis for real wage growth, has increased only marginally over the last decade,
Figure 1.2. First-Time Job Seekers as a Proportion of Total Unemployed
Egypt Jordan Morocco Yemen United
Arab
Emirates
Kuwait
percent
100
90
80
70
60
50
40
30
20
10
0
2000
2000
1995
2001
2001
1999
Sources: ILO 2003a, except for Kuwait: Population Census 2001; and for Jordan: AREUS 2001.
Figure 1.3. Distribution of Unemployed by Level of Education
0
10
20
30
40
50
60
70
80
Bahrain Oman Jordan Egypt Morocco Tunisia Algeria Djibouti
none
primary
secondary
tertiary
percent of total unemployed
Sources: ILO 2002, except for Bahrain: ILO 2003a; for Egypt: ELMS, 1998; for Morocco: LSMS 1999.

Part I: Growth, Reform, and Governance 7
remaining far below that in East Asia and the Pacific, South Asia, and Latin
America and the Caribbean. As a result, real wages have stagnated or declined
in half the Arab countries since the 1980s, and only increased weakly in most
of the other Arab countries.
Unprecedented labor market challenges to come. Beyond the poor job
prospects that have characterized the Arab region over the past two decades,
there is also the knowledge that labor market pressures will only build over the
Figure 1.4. Female Labor Supply in Arab States, 1950–2020
1950s
percent
1960s 1970s 1980s 1990s 2000s 2010s
60 6.0
5.0
4.0
participation rate
(end of decade, left axis)
gender parity labor force index
(end of decade, left axis)
average annual labor force growth
(right axis)
3.0
2.0
1.0
0
50
40
30
20
10
0
Sources: ILO 1996; UN 2002.
Figure 1.5. Unemployment Rates by Gender in Arab Countries
0
5
10
15
20
25
30
35
40
female
male
Algeria 2000
West Bank and Gaza 2001
Jordan 2000
Morocco 1999
Tunisia 1997
Bahrain 2001
Qatar 2002
Yemen 1999
Syria 2001
Egypt 2000
Lebanon 1997
Saudi Arabia 1999
United Arab Emirates 1999
Kuwait 2003
percent
Sources: 1. Algeria, Jordan, Syria, Egypt, West Bank and Gaza, Yemen: ILO 2003b; 2. Tunisia, INS 2001; 3. Bahrain, Qatar, Saudi Arabia, Kuwait, UAE:
Girgis, Hadad-Zervose, and Coulibaly 2003; 4. Morocco, LSMS 1999; 5. Yemen: NPPS 1999.

Breaking the Barriers to Higher Economic Growth8
next two decades. A legacy of high population growth rates in the Arab region
between 1950 and1990, which peaked at 3.2 percent a year in 1985, has trans-
lated into some of the most intense pressures on labor markets observed any-
where in the world in the post-WWII period. The current labor force is grow-
ing by a rate of 3.3 percent per year. That translates into some 37 million new
workers entering the labor force between 2000 and 2010, swelling the total
labor force by close to 40 percent in a span of only 10 years. Between 2000 and
2020, the labor force in the Arab states will have expanded by some 75 percent.
In the span of 20 years, 74 million new jobs will need to be created, just to
absorb the growing labor force.
Job creation must be even higher, if the region is to address the current
unemployment problem. If the Arab region is to accomplish the more ambi-
tious goal of absorbing these unemployed workers, in addition to the new
entrants to the labor force, close to 90 million new jobs will need to be creat-
ed by the end of the next decade, well over double the number of jobs that
currently exist in the Arab world.
Under any comparison, the Arab region faces an unprecedented challenge
in development. These labor market pressures would be a development chal-
lenge for any country. But for the Arab region, the need for accelerated job
creation comes just as two of the Arab region’s major sources of employment
creation over the past four decades, public sector employment and labor
migration, are providing fewer and fewer jobs.
Figure 1.6. Labor Force Growth in Developing Regions, 1970–2010
1970–80
1990–2000
1980–90
2000–10
East Asia South Asia Latin America
and the
Caribbean
Sub-Saharan
Africa
Arab Countries
percent
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
.5
0
Source: ILO 1996.
Note: Arab countries include members of the League of Arab States except Comoros, Djibouti, Mauritania, Somalia, and Sudan.

Part I: Growth, Reform, and Governance 9
Constraints of the Old Development Model
Employment creation in the Arab region has been greatly influenced by the
historical model of development adopted throughout the Arab region. In part
to correct for a legacy of inequities and poverty during the first half of the 20th
century, at the time of independence the political actors of the Arab region
adopted models of development based on strong governments, central plan-
ning of economic and social priorities, and wide-scale policies for redistribu-
tion and equity. This strategy included nationalization of many private assets,
state planning, industrial development through protected local markets, and
vast resources directed to social development and large-scale public sector
employment.
Early positive results. Early on, this model had strong payoffs. The region as a
whole benefited from oil, either directly (for oil-producing economies) or
through aid and labor remittances. With oil and oil-related revenues, in addi-
tion to directing large resources toward public infrastructure, the Arab
economies directed vast resources toward education, with the resulting aver-
age level of education of the adult population increasing from less than a year
in 1960 to more than three years in 1985. Fueled by improvements in basic
education and heavy investments in health care, health indicators also
improved significantly, and poverty was substantially reduced.
Significant costs. But the model also has had strong implications for the eco-
nomic orientation of the Arab economies, on governance, and on employ-
Figure 1.7. Average Annual Growth in the Labor Force of Arab Countries, 1970–2010
0123
1970–90
1990–2010
456
Jordan
Yemen
Syria
Egypt
Morocco
Tunisia
percent
Source: ILO 1996; UN 2002.

Breaking the Barriers to Higher Economic Growth10
ment creation. Economically, heavy protection and regulation of industry,
along with overvalued and uncompetitive exchange rates, provided marked
disincentives for the growth of a tradable goods sector in the Arab states. Oil
resources relieved many governments of the need to tax their citizens, and
allowed governments to redistribute substantial resources through vast wel-
fare and social services systems. At the same time, this system of pervasive
redistribution of wealth reduced demands from Arab citizens for accountable
and inclusive public institutions. Arab governance mechanisms lacked trans-
parency, reflected in limited access to government information and carefully
monitored freedom of the press. They lacked contestability, as reflected in
some of the most centralized governments of all developing countries. And
they lacked inclusiveness, as reflected in rural-urban inequalities in access to
public services, gender inequalities in voice and participation in society, and
nepotism or patronage determining who received public services or access to
lucrative business opportunities and who did not.
The private sector that emerged under this model, developing under the
patronage of governments, flourished not so much by being dynamic in a
competitive environment, but often by supplying protected domestic markets
and generally “living off the state.” Despite high growth rates during the oil
boom years, investments in the Arab region became progressively unproduc-
tive. Though still positive, total factor productivity growth was cut in half in
the Arab region from the 1960s to the 1970s.
The model also had significant implications on employment creation.
Along with vast social welfare systems, Arab governments redistributed
resources through public sector jobs. Governments became the employer of
choice throughout the Gulf Cooperation Council (GCC), and while oil rev-
enues were high, nationals entering the labor force could be hired almost
exclusively by the public sector. Even in the non-oil-producing economies, aid
flows and labor remittances from citizens working in the Gulf permitted the
public sector to grow to unprecedented levels.
A major source of employment in the region also came in the form of labor
migration, with large numbers of Arab laborers working in other countries in
the Gulf. At the peak of the oil boom in the early 1980s, some 3.5 million Arab
migrant workers were employed in the Gulf states. During 1973-1984, official
remittances totaled almost US$22 billion in Egypt, US$8.2 billion in Morocco,
and US$6.5 billion in Jordan. This provided an enormous buffer to any lack
of job opportunities at home.
The oil price collapse and a decade of crisis. When oil prices collapsed in the
mid-1980s, government revenues throughout the region felt the blow, begin-
ning with the oil producers, but with the nonoil producers close behind.
Although cushioned by extensive external assistance, the flow of resources to
the public sector slowed considerably and government expenditures on social

Part I: Growth, Reform, and Governance 11
systems and physical capital shrank. For the region as a whole, public fixed
investment was cut heavily, and growth in physical capital stock per laborer
declined by more than 60 percent from the 1970s.
Facing declining public revenues, governments struggled to maintain their
redistributive commitments. The fiscal strains contributed to large macroeco-
nomic imbalances. Productivity growth, already declining by the 1970s, plum-
meted over the 1980s. Growth collapsed under the multiple blows of declining
public spending, an unattractive private investment climate, and continuing loss-
es in efficiency. Economic growth per capita averaged only 0.7 percent per year.
At the same time, and for the first time, unemployment began to emerge as
a serious problem in the Arab states. With reduced public sector revenues,
guaranteed employment in the public sector was no longer possible. Labor
migration opportunities were also diminished. Lower oil prices, rapidly rising
domestic supplies of national labor, and competition from lower-cost labor
elsewhere in the world all worked to dampen the GCC demand for labor from
the rest of the Arab region.
Initial efforts at reform. The lack of economic growth and deteriorating budg-
et deficits prompted several of the economies to undertake programs of
macroeconomic stabilization and structural reform aimed at encouraging pri-
vate sector development, so it would become an engine for growth and
employment creation. Macroeconomic stabilization was broadly achieved,
but the pace of structural reform differed markedly throughout the region.
• A group of resource-poor countries, including Tunisia, Morocco, and
Jordan, implemented earlier and more intensive reforms toward more open,
private sector-led economies than the rest of the Arab countries. Tunisia
Figure 1.8. GDP per Capita Growth in Developing Regions, 1970–2000
1970s
1980s
1990s
East Asia
and Pacic
High-Income
OECD
Latin America
and the
Caribbean
South Asia Sub-Saharan
Africa
Arab States
percent
7
6
5
4
3
2
1
0
⫺1
⫺2
Note: Arab States include all MENA countries except Iran, plus Sudan.

Breaking the Barriers to Higher Economic Growth12
introduced, as early as the 1970s, an “offshore” export-processing platform
to facilitate trade, and both Morocco and Tunisia joined the General
Agreement on Tariffs and Trade. Reforms also included exchange rate liber-
alization, tax reforms, trade and financial sector liberalization, and privati-
zation. All three signed Euro-Med agreements in the 1990s, and Jordan
signed a free trade agreement with the United States. In general, the reform
effort has been steady in these countries, and without policy reversals.
• Other resource-poor countries, including Egypt and Lebanon, have pursued
reform more slowly and sporadically. Despite early reforms in Egypt, and
aggressive macroeconomic stabilization in the 1990s, reforms were partially
reversed with escalating behind-the-border trade restrictions and signifi-
cant exchange rate overvaluation. More recently, reforms have resumed,
especially with the signing of a Euro-Med agreement in 2001. Lebanon, with
a legacy of destroyed physical and economic infrastructure and weaker
institutions, had to contend with large macroeconomic imbalances.
• Another set of countries with significant oil resources and large popula-
tions, including Algeria, Syria, and Yemen, also pursued reforms later, more
gradually, and more sporadically than the early reformers. Algeria, with
macroeconomic imbalances stemming from the collapse in oil prices,
aggressively pursued macroeconomic stabilization, but structural reforms
have been far more limited. Trade reforms initiated in the early 1990s were
reversed in 1998, and then taken up again in 2001 with the signing of a
Euro-Med agreement. Reforms in key areas such as the financial sector and
privatization remain limited. In Syria, the trade and investment liberaliza-
tion begun in 1991 was not sustained. While there has been modest
progress in exchange rate unification and private sector regulatory reforms,
broader structural reforms have been limited. And in Yemen, macroeco-
nomic stabilization reforms have not been accompanied by more aggres-
sive reforms to diversify the economy, despite relatively open trade policies.
The investment climate remains poor, reflecting weak rule of law and prop-
erty rights, ineffective regulatory frameworks, and security problems.
• The six GCC economies—Bahrain, Kuwait, Oman, Qatar, Saudi Arabia,
and the United Arab Emirates—have long maintained an open trade sys-
tem with free movement of capital and advanced financial systems. As oil
prices deteriorated, most countries cut expenditures, but aggregate budget
deficits have generally increased. Some of the smaller GCC countries have
encouraged growth in selected sectors such as entrepot trade (United Arab
Emirates), financial services, and tourism (Bahrain and United Arab
Emirates). Oman has made substantial efforts to broaden private sector
participation and improve the investment climate, with privatizations and
changes in its foreign capital investment law. In Saudi Arabia, reforms have
progressed more slowly. The public sector still dominates economic activi-
ties; and government revenues remain dependent on oil.

Part I: Growth, Reform, and Governance 13
Failure of growth to improve. In general, the structural reform agenda in the
Arab region, despite its intentions, has been cautious, selective, and often sub-
ject to pause and to reversals. The reform agenda avoided most governance
reform or the opening of the political space needed for any type of deeper
reforms that depend on the compliance and participation of the social groups
whose well being the reforms are intended to improve. As a result, the reform
effort stopped short of providing a substantially improved climate for growth
and investment, and the recovery in growth over the 1990s was weak, with per
capita GDP growing by an average of only 1.5 percent per year.
Declining employment opportunities from the old model. Now, as the Arab
region faces an unprecedented challenge in job creation over the next two
decades, the major sources of employment creation of the past are quickly
dwindling. The public sector can no longer be the employment outlet it has
been in the past. Most branches of public sectors are already overstaffed, by as
much as a third or more in some countries. More important, the public sec-
tor, with the marked change in fiscal circumstances throughout the region,
cannot act as a refuge to vast numbers of unemployed.
The Arab countries’ development has relied heavily on three financial
sources: oil, aid inflows, and workers’remittances. These three sources provid-
ed an essential supply of public revenues and private earnings, and supported
the large-scale public sector employment policies of the past. But all three of
these revenue sources are under greater pressure.
Figure 1.9. GDP per Capita Growth in the Arab World, 1970–2000
1970s
1980s
1990s
Bahrain
Jordan
Kuwait
Oman
Saudi Arabia
Syria
Sudan
Algeria
Egypt
Morocco
Tuni si a
⫺8
⫺6
⫺4
⫺2
0
2
4
6
8

Breaking the Barriers to Higher Economic Growth14
Oil revenues will decline for all countries in the Arab region, continuing
the pattern of the past two decades. Known oil resources will be depleted in
some countries of the region, such as Algeria, in about four decades, while in
some others, such as Egypt and the Republic of Yemen, much sooner. Exports
will fall as domestic energy consumption ratchets up and the population
grows. With declining oil production, the decline in per capita oil rents will be
steeper than in the past two decades.
Aid flows are expected to similarly decline, except in temporary periods of
strategic importance and conflict resolution. Finally, labor remittances are not
projected to increase significantly, a result of deteriorating prospects for labor
migration.
Prospects for another major source of employment creation—labor
migration—are also diminished. Throughout the 1970s and 1980s, labor
migration provided a critical employment outlet for workers in many of the
non-oil-producing Arab economies. In the mid-1980s, close to 10 percent of
the Egyptian and almost 15 percent of the Yemeni labor forces were employed
abroad, primarily in the Gulf. The last decade,however,has seen a rapid decel-
eration in the net outflow of Arab workers to other countries in the region.
Migration has slowed to the GCC countries of the region.
Combined, the region’s traditional sources of job creation are fast dwin-
dling, just as the need for job creation is accelerating.
Three Major Economic Realignments Are Required
If the region is to ensure that the rapidly expanding labor force has both suf-
Figure 1.10. Per Capita Oil Exports in Arab Countries, 1980–2000 (US$)
non-GCC
GCC and Arab countries
Arab countries (left axis)
GCC (left axis)
non-GCC (right axis)
12,000 90
80
70
60
50
40
30
20
10
0
10,000
8,000
6,000
4,000
2,000
0
1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000
year
Source: Oil exports, COMTRADE 2002; population, WDI 2002.

Part I: Growth, Reform, and Governance 15
ficient employment opportunities and the prospects for real wage growth,
GDP growth in the region will need to double from its average of 3 percent
per year over the late 1990s to 6-7 percent a year for a sustained period. This
is an enormous growth challenge, but tantamount to meeting the enormous
labor market challenge ahead.
To meet the growth and employment challenges, the Arab countries must
address a set of long-standing policy and institutional challenges to complete
three fundamental and interrelated realignments in their economies:
• From public sector-dominated to private sector-dominated economies
• From closed economies to more open economies
• From oil-dominated and volatile economies to more stable and diversified
economies.
Figure 1.11 Aid to GDP Ratio in Arab Countries, 1980–2000
percent
9
8
7
6
5
4
3
2
1
0
1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000
year
Figure 1.12. Worker Remittances as Percentage of GDP, Egypt and Morocco: 1970–2000
0
2
4
6
8
10
12
14
16
1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000
percent
Egypt
Morocco
year

Breaking the Barriers to Higher Economic Growth16
Expanding the Role of the Private Sector in the Arab Economies
The private sector’s contribution to value added in the Arab region is low
compared to that in other regions, and while most Arab countries attempted
to expand private sector activity since the late 1980s, it increased only margin-
ally over the 1990s. The scope for private sector expansion is very large in the
Arab region, but it requires a conducive economic and political environment.
Improving the investment climate. Firm startups and operations in the Arab
world are significantly hindered by the time and financial costs of regulatory
and administrative barriers. For new firms, the costs of complying with regu-
lations as a proportion of gross national income are twice what new firms face
in Eastern Europe, and five times what firms face in East Asia, Latin America,
and the Caribbean. Other costs are also high, from licenses to domestic taxa-
tion, import duties, and regulations that slow customs clearance.
World Bank Investment Climate Surveys reveal that potential investors in
the Arab world are significantly hindered by obstacles to entry, including
cumbersome licensing processes, complex regulations, opaque bidding proce-
dures, and official acceptance of uncompetitive practices. In Morocco, about
half the firms say they have had to hire intermediaries or maintain full-time
staff to deal with the bureaucracy. In Jordan, a potential investor interested in
registering a new firm has to wait three months, with half of that time spent
on a single procedure—inspection by the ministry concerned.
Figure 1.13. Private Sector Contribution to GDP in Arab Countries
100
90
80
70
60
50
40
30
20
10
0
1995
1993
2002
1985
1997
1990
1999
2000
2002
1985
1999
1983
1997 1995
2000 1990 2001
1990 2000
1980
2000
Lebanon Egypt Morocco Jordan Palestine Syria Tunisia Yemen United
Arab
Emirates
Algeria Saudi
Arabia
percent
Source: Assaf and Benhassine 2003.

Part I: Growth, Reform, and Governance 17
Businesses are further handicapped by judicial systems that facilitate nei-
ther restructuring of viable businesses nor closure of nonviable ones. The
Arab region compares poorly with other regions in the complexity and time
needed to initiate and complete a legal claim. Unpredictability of enforcement
is an even more serious problem. Nearly half the surveyed firms expressed
concern over the unpredictability of the legal system.
Improving key finance and infrastructure for business. Weaknesses in infrastruc-
ture and in the financial system are yet another burden on business opera-
tions. According to World Bank Investment Climate Assessments, almost half
the private businesses in the region complain that infrastructure is a moder-
ate to major obstacle to conducting business. Telecommunications and trans-
port, two backbone services, are significantly underdeveloped. With public
banks dominating the banking system in many countries, and favoring state
enterprises, larger industrial firms, and offshore enterprises, smaller firms
have a hard time getting the startup and operating capital they need.
Accelerating privatization. Finally, several strategic services for the private
sector—banking, telecommunications, and transportation—remain under
public ownership in most Arab countries. Privatizations in the 1990s focused
on manufacturing firms, only recently extending to selected services such as
telecommunications. And even where privatization of services is advancing,
regulatory arrangements often remain weak.
Integrating with the World
The limited integration of Arab economies. The Arab region has largely missed
out on the trends of global trade integration. The region’s trade integration,
measured by the ratio of trade to GDP, fell from about 92 percent of GDP in
1980 to about 70 percent in 1985, and it remains around 65 percent today. The
region’s oil resource boom helped it realize high initial export-to-GDP ratios,
but with the windfall revenues invested at home in infrastructure and servic-
es (and not in tradables), the export-to-GDP ratio steadily declined.
The scope for expanding trade for the Arab countries is significant. Exports
other than oil are about a third of what they could be, given the characteris-
tics of the region in terms of endowments, size, and geography.
Manufacturing imports are one half of what they could be. Foreign direct
investment could be five to six times higher.
Arab country trade is marked by both a high degree of product and geo-
graphic concentration. Although the share of fuels in total exports has fallen,
fuels still constitute the most significant exports for the region. In 1978, 76
percent of merchandise exports were fuels. By 2001, that share had only fallen
to 58 percent. The manufacturing sectors of most Arab countries are small by

Breaking the Barriers to Higher Economic Growth18
international standards—almost half the typical levels in other lower-middle-
income economies. Some of the resource-poor countries, including Tunisia
and Morocco, have seen a large fall in their trade product concentration, com-
parable with Chile or Malaysia, and countries like Lebanon and Jordan,
already well diversified, have seen further declines in concentration. But the
diversification by labor-abundant, resource-rich economies (Algeria, Egypt,
Syria) was limited, and for the major oil producers—Libya, the GCC
economies—there has been little improvement in diversification.
Barriers to integration. Several factors have hindered Arab integration into the
world. The trade regimes in the Arab states are among the most protective in
the world. Tariffs remain high and dispersed. Nontariff barriers, including
lengthy processes to comply with customs and quality control standards, are
still widespread. The trade-impeding effect of these barriers has been com-
pounded by often persistent overvaluation of exchange rates.
In addition, behind-the-border constraints are considerable. Transport,
logistics, and communication costs are high, effectively raising the costs of
trade. Add to that an investment climate that generally discourages the start-
up of small and medium firms, often critical to success in trading.
Removing the barriers to trade. The Arab region needs to significantly increase
the speed of integration into the global economy, reducing the level of tariff
protection, dismantling inefficient nontariff barriers, and removing the
Figure 1.14 Arab Nonoil Export Potential
weak-performing countries
3.5
3.0
2.5
2.0
1.5
1.0
.5
0
high-performing countries
Algeria
Egypt
Lebanon
Syria
Arab9
Tunisia
Saudi Arabia
Jordan
Morocco
Iran
Colombia
Argentina
Brazil
Russian Fed.
South Africa
Turkey
Guatemala
Poland
Bulgaria
El Salvador
Mauritius
Bolivia
Chile
Jamaica
Ecuador
Slovak Rep.
Czech Rep.
Costa Rica
Korea, Rep. of
Mexico
Hungary
Thailand
Philippines
Malaysia
Note: Regression is based on 42 countries, but values for 8 low-income countries, including Yemen, are not reported because of negative values.
Arab9 = Algeria, Egypt, Jordan, Lebanon, Morocco, the Republic of Yemen, Saudi Arabia, Syria, and Tunisia.

Part I: Growth, Reform, and Governance 19
behind-the-border constraints to trade that have raised the costs of trade and
have discouraged both domestic and foreign investment. Real exchange rates,
which have been significantly overvalued in the Arab region for the last two
decades, need to be corrected as well.
The imperative for trade reform is high. The costs of wait-and-see policies are
also becoming increasingly expensive for global trade integration. The region
faces intensified competition in world markets, at both the skill-intensive and
the labor-intensive ends, affecting important industries and activities such as
textiles and garments, light engineering, and manufacturing. On the skill-
intensive end, two major competitive threats are the upcoming accession of
some Eastern European countries to the European Union and Turkey’s cus-
toms union agreement with the European Union, the Arab region’s major
export market. Other regional groupings and free trade agreements also pres-
ent challenges. Both the Europe and Central Asia and the Latin America and
Caribbean regions, two middle-income regions with labor skill profiles simi-
lar to the Arab region, have enjoyed faster trade reforms and integration with
global markets in recent years, and are rapidly gaining market share as the
Arab world lags behind. A mismatch between the skills required by firms and
the skills produced by an education system still organized to turn out new
government employees intensifies the competitive problem.
Labor-intensive products and manufactures are also facing a new challenge
from low-wage, high-productivity countries such as Bangladesh, China,India,
Indonesia, and Vietnam. That competition will escalate with the abolition in
2005 of quota preferences in textiles and garments, important to the region’s
manufacturing employment and export earnings. These labor-intensive sec-
tors, employing more than a million workers, have been among the few in
which the Arab region has been gaining market share—partly because of the
protected quota markets in the European Union. Adjusted for labor produc-
tivity, labor costs in the garments sector in Arab counties are significantly
higher than in populous countries of Asia, but lower than in Eastern European
countries. The longer the region delays trade and related reforms, the more
disadvantaged exporters will be in facing these competitive pressures.
How Arab trade can flourish. The future for Arab trade will involve specializa-
tion in manufacturing, cutting up the production chain, and permitting finer
gradations of specialization within that chain among skills, labor costs, and
productivity. Small, resource-poor countries in the region stand to benefit
from such production chains, and given their size, the prospects are virtually
unlimited in world markets. Larger countries will also benefit from such spe-
cialization. Their larger domestic markets, and proximity to major interna-
tional markets, will drive a much larger range and scale of domestic manufac-
turing possibilities.

Breaking the Barriers to Higher Economic Growth20
The role of regional trade agreements. Regional trade agreements can be an
important anchor for reforms. Trade with Europe, the natural geographic
partner of the Arab region, falls far short of its potential. Agreements with the
EU could provide immediate and expanded access to its markets for agricul-
ture, as well as increased temporary migration, funds for managing transition
costs, and more efficient rules of origin.
At the same time, however, the Arab countries need to maintain open
access to world markets, anchoring their trade and investment reforms in a
multilateral framework, such as the WTO, to give them greater credibility.
Only half of the Arab population lives in a WTO member country—the low-
est share of all developing regions. This is partly due to the disproportionate
share of Arab countries under trade and foreign investment sanctions. But it
is also by choice. Arab countries have generally opted for preferential trading
arrangements with the EU, which have immediate appeal because of geo-
graphic proximity and EU financial and technical support. In addition, pri-
mary oil exporters with little other trade are wary of bringing their oil trade
under the WTO.
But speeding up WTO accession processes is necessary. The transition
process for the Arab region depends upon signals. The enhanced credibility of
government policies that are bound as commitments to the WTO gives the
world’s investors the greatest degree of confidence in the permanence of the
reform process.
Limited intraregional trade and scope for expansion. There have been many
attempts to foster intraregional integration among Arab countries over the last
half century. But they have mostly failed to produce enhanced trade within the
region. Intraregional exports among Arab countries remain of the order of 8-
9 percent since the 1980s, compared to 22 percent for the ASEAN (Association
of Southeast Asian Nations), and 25 percent for MERCOSUR (Argentina,
Brazil, Uruguay). Many factors prevent an enabling environment for trade
among the Arab economies. They include restrictions on imports—either
directly through customs or more indirectly through the financial sector, but
they also include a host of factors that prevent this facilitating environment for
regional trade. There are a variety of infrastructure issues, such as inadequate
telecommunications, as well as trade-chilling contingent protection and obsta-
cles created by frontier frictions, such as frontier red tape and differences in
national product standards. The Arab region needs to work to eliminate these
inhibiting factors to trade, and allow Arab exports to the region to at least be
on a level playing field with exports from the rest of the world.
The potential for increased intraregional Arab trade and investment is large
and could be a source of increased investment, production and employment.
The recent Pan-Arab Free Trade Area Agreement improves on past efforts and

Part I: Growth, Reform, and Governance 21
opens new opportunities. But its impact is likely to be limited for the reasons
discussed above, and because it remains restrictive on trade in agriculture and
services. In addition, the rules of origin remain loosely defined, leaving open
the possibility that they will continue to be used to protectionist effect.
It is important to notice, however, that even a successful process of intrare-
gional integration is unlikely to be a substitute for increased integration with
the world economy at large. Analysis and lessons from international experi-
ence show that intraregional integration is best and most successful if it is
pursued within a broader agenda of openness with the rest of the world. In
addition, even integrated, the Arab economy would remain a relatively small
player, and would not be able, with an intraregional focus only, to harvest the
opportunities of the global economy in terms of production, technology, and
investment.
Managing Oil Resources
In addition to greater private sector participation and openness, diversifica-
tion out of oil will involve fundamental changes in institutions managing oil
resources and their intermediation to economic agents.
Economic diversification requires proper fiscal policy and public expendi-
ture management in addition to measures to expand globally competitive pri-
vate businesses outside the oil sector. That means setting up institutions and
fiscal rules that insulate public expenditures from oil price volatility, and save
oil revenues so that they can continue to benefit citizens when oil resources
decline. That also means improving the efficiency of public expenditures
through better systems of budgeting that emphasize performance and
accountability.
Diversifying productive activities is a growing priority, not only for coun-
tries whose known oil reserves will soon be depleted, but for all oil producers.
Diversification is essential if the region is to ensure vital public expenditures.
Per capita exports of hydrocarbon products have been declining across the
region, with falling real prices, rising domestic demand for energy, and rapid
population growth. Governments will need to develop new sources of revenue
to ensure the efficiency of public expenditures.
Diversification is also critical for job creation. The labor force is rapidly
expanding, in far greater numbers than the public sector can afford to employ,
especially given the declining oil revenues. Significantly greater job opportu-
nities need to be created in the Arab economies. To do that, the region must
diversify into a broad range of productive activities, which can provide the
basis for substantially greater employment. This diversification will not only
create the needed new jobs, it will also provide substantially greater protection
to the labor force from commodity price shocks.

Breaking the Barriers to Higher Economic Growth22
A Transition Depending upon Improved Governance, Higher-Quality
Education, and Gender Equality
The three realignments—key for managing the region’s employment
challenge—cannot be accomplished by policy change alone. Fundamental to
each transition is improved governance, across the board. Each transition
implies deep changes in the role of government and strong improvements in
its effectiveness. The governance agenda is not a separate challenge, to be
worked on at its own pace. It is a complementary and reinforcing agenda to
reform efforts—in private investment, trade, and economic
diversification—that change governance mechanisms, thereby improving
capacity and incentives within government, while fostering a larger role for
civil society in governance. While better governance cannot guarantee opti-
mal economic policies, it is indispensable to guard against persistently poor
policies and to ensure that the good policies needed to meet the Arab
region’s growth potential enjoy legitimacy and are implemented faithfully
and with celerity.
The fundamental governance challenges are to strengthen the incentives,
mechanisms, and capacities for more accountable and inclusive public institu-
tions, and to expand allegiance to equality and participation throughout soci-
ety. Those good governance mechanisms are first steps in improving econom-
ic policies that are themselves instruments for improving the climate and
incentives for efficient growth.
To improve the region’s governance, both governments and the people
must commit publicly to formulating and implementing a broad program of
inclusiveness and accountability of government, and increased transparency
and contestability in public affairs. Such programs would vary across coun-
tries, but the process of formulating these programs should itself set high
standards for including all segments of society, and for making all delibera-
tions public to ensure maximum transparency.
Enhancing inclusiveness. The first step to enhanced inclusiveness is to adopt
laws and regulations that secure access to widely accepted basic rights and
freedoms, including participation and equality before the law, particularly for
women. Broader public consultation, greater freedom for the media, fewer
restrictions on civil society, more equitable channels of access to social servic-
es, and an end to discriminatory laws and regulations are examples of meas-
ures that secure inclusion. They need to be supplemented with strong moni-
toring to ensure that public officials treat all citizens equally.
Enhancing accountability. But a particular focus must rest on putting in place
better systems of accountability, especially transparency and contestability.
External and internal accountability are related; the former is critical for pro-

Part I: Growth, Reform, and Governance 23
viding incentives for governments to strengthen their structures of internal
accountability, so action is needed on both fronts. External accountability, to
help citizens participate in and monitor government, can come from actions
at both the national and local levels. Such measures include greater freedom
of information and public disclosure of government operations, including
data on government quality; wider public debate by an independent and
responsible media and by representative civil society groups; and regular and
competitive elections with external oversight to ensure they are fair and open.
At the local level, external accountability can come from providing citizens
with effective feedback mechanisms on public agencies, expanding choices in
public service delivery, devolving responsibilities to empowered local commu-
nities, and by actively soliciting the participation of community empower-
ment organizations.
Enhancing internal accountability is about creating mechanisms and
incentives within government to ensure effective functioning in the public
interest. At the national level, it involves strengthening the checks and bal-
ances within government, in particular increasing the independence and
capacity of the legislative and judicial branches, and installing independent
oversight agencies like ombudsmen. Within the executive, it involves reform-
ing public administration, strengthening the performance orientation of gov-
ernment budgets, enhancing the service orientation of the civil service, aug-
menting the capacity of local governments, and ensuring the independence of
regulatory bodies.
Enhancing quality education also critical. The transition to more market-
driven, globally oriented economies requires continuing progress in widening
and deepening the stock of human capital, and more critically, changes in the
qualitative outputs of the region’s educational systems. Firm surveys in both
labor-abundant and labor-importing countries in the region suggest that lack
of skills is an important constraint in hiring. Qualitative deficiencies in edu-
cation have resulted from a variety of factors, including overly centralized
management of education, little assessment of performance, and promotion
based on seniority rather than performance.
Improving the relevance of the region’s educational outputs must be partly
addressed from within the educational sector, but it will be partly accomplished
by the transitions themselves. The past policies of ensuring public sector
employment to those with higher or intermediate education, with little atten-
tion to content or quality, has resulted in Arab individuals seeking higher
degrees that bear little relation to the skills needed by either the private sector or
for the efficient functioning of the public sector. As the private sector’s role in
the economy expands, more appropriate signals will be sent to students of the
types of education in need, and the types of education that will be rewarded.

Breaking the Barriers to Higher Economic Growth24
Greater openness will also foster improvements in education. Greater
openness fosters competition, encourages modern technology, increases the
demand for highly skilled labor, and promotes learning by doing. Openness
obliges industries to confront their inefficiencies. To compete successfully,
industries must adapt, thereby creating demand for the new skills and trades
to do so.
Greater economic participation by women needed. The success of these transi-
tions hinges also on progress in enhancing gender equality.Progress on bridg-
ing the gender gap in education and health in the region has been substantial.
But this has not translated into a commensurate increase in women’s partici-
pation in the labor force. The participation of women in the labor force in the
Arab region, at less than 33 percent, is the lowest in the world (compared to
45 to 75 percent in other parts of the world). This is partly the result of a range
of civil, commercial, labor, and family laws and practices in the region that
discourage women from entering the labor force.
At this time of exceptional demands on Arab economies to create jobs,
there has been little incentive to improve economic opportunities for women,
with some arguing that greater participation by women in the labor force
would exacerbate unemployment among men in the current climate of slow
growth and rising unemployment. But there is no empirical evidence to sug-
gest this. International experience shows that in the long run, greater labor
force participation by women is not associated with higher total unemploy-
ment rates.
The region needs to be able to draw on all of its talents and human capital.
Women’s greater participation in the labor force improves the overall growth
of the region. With the highest economic dependency rates in the world (that
is, each working person supporting two nonworking dependents, a burden
that is double what is observed in East Asia), the Arab region is missing
opportunities for increased welfare of families and society. Higher female
labor force participation, in line with what would be expected given female
education, age structure, and fertility rates, could allow family incomes to
increase by 15-30 percent in most countries. In addition, this participation
enhances the flexibility of families to adapt to changing economic conditions.
The agenda for addressing gender disparities includes the need for realign-
ing legal provisions that fail to give effect to the recognition of equal rights
under constitutions in most countries. It would also mean ensuring a sup-
portive infrastructure that facilitates women’s participation in the public
sphere, as well as continued attention to education, particularly in areas that
provide better market skills. Finally, it would involve reforming the labor laws
and regulations that raise the cost of women’s labor relative to that of men,
and serve as disincentives to private employers’ demand for female labor.

Part I: Growth, Reform, and Governance 25
An Ambitious, but Feasible Agenda for the Arab Region
Large growth and employment gains. The impact of an integrated package of
policy realignments that improves the business and investment climate for the
private sector, and fosters integration with the world economy, is potentially
very large. Based on the experience of comparable countries, output per
laborer could increase by some 2–3 percent a year, with greater integration
into the global economy. Using similar international evidence of high-
performing countries, it is estimated that improving the institutions of
accountability and public administration could boost output growth per capi-
ta by 0.8 and 1.3 percent a year. Increasing the participation of women in the
labor force to levels comparable to the highest performers in the region may
add 0.7 percent or more to GDP growth per capita.
While these effects are not additive, and reflect changes in policies and
institutions that are not exclusive, a conservative estimate of the sum of these
projected effects, taking into account overlap in the channels through which
the policy changes operate, would be an increase in output growth per labor-
er of 2.5–3.5 percent a year.
The suggested economic transformation and deep reforms would generate
millions of new jobs, and more productive jobs in traded sectors across man-
ufacturing and services. For instance, the Arab region’s share of nonoil mer-
chandise exports in total exports averages only 6 percent, versus more than 20
percent in East Asia and the Pacific. Bridging only half that gap, with associ-
ated increases in domestic and foreign private investment, would create more
than 4 million new jobs over the next five years. That is equivalent to cutting
the unemployment rate by 4 percentage points of the labor force. The broad-
er reform agenda would bring even larger benefits.
Feasibility. The underlying concern for many Arab countries is whether this
broad transformation of their economies is even feasible. But the experience of
countries around the world provides evidence that such a transformation is
possible. Around the world, many countries have faced labor market crises and
responded to the political challenges with reforms. Governments have reduced
the scope of state intervention in markets, reorganized regulatory frameworks,
and undertaken wide-ranging economic restructuring, including the large-
scale privatization of public enterprises, while trying to preserve important
aspects of social welfare, as well as wage and employment protections.
In some postsocialist states of Eastern and Central Europe, these transi-
tions were facilitated by the urgency of establishing new systems of economic
and political governance. Elsewhere, policy makers have advanced economic
reforms within stable policies. In several Latin American countries, for exam-
ple, politicians and reformers effectively navigated complex political environ-
ments containing entrenched, often competing interests. Despite the presence

Breaking the Barriers to Higher Economic Growth26
of supporters of the status quo, governments built effective reform coalitions
and successfully re-regulated economies around market-based principles.
Arab countries have also undertaken selective reforms, with positive, if limit-
ed, effects on economic performance.
The challenges confronting Arab governments do not arise from a lack of
information about what needs to be done. Pathways to reform are much bet-
ter mapped today than they were only two decades ago. Past decades have pro-
duced extensive knowledge about what works in development strategies and
policy reform, and what does not.
A new path for greater employment and a new social contract. The limitations of
past approaches to reform are visible in the deteriorating conditions in Arab
labor markets. The old model that guided the region’s development for the last
40 years no longer provides the needed employment opportunities. To ensure
that the citizens of the Arab world have these opportunities as the need for jobs
accelerates, the region must shift to market-based strategies of growth, which
can provide the substantially higher growth rates to ensure better labor market
outcomes. It involves fundamental transitions in each of the economies.
A new social contract as the basis for more productive relations among the state,
labor, and the private sector. The new social contract would balance the needs
for market flexibility with the rights of workers, would link reform to the
principles of poverty reduction, income equality, and income security. It
would restructure the rigid, exclusionary, and inefficient aspects of the old
social contract. The exact agenda, the sequencing and emphasis will vary from
country to country.
Need to link economic and political reform. Moving the reform program
beyond its current limits requires reviving national conversations about labor
market reform, restructuring redistributive programs, and redefining the
terms of the new social contract. The top-down approach to economic
reform—which has prevailed—sidesteps the need for political reform to
secure the legitimacy of reform and the credibility of government commit-
ment. It is no longer adequate. The strong preference of Arab governments for
political and social stability has been often advanced as a justification for not
proceeding with the strong and deep reform agenda that is needed for Arab
economies to regain dynamism and meet their challenges. But today the costs
of maintaining a nonviable status quo are becoming severe, and may under-
mine that much valued stability, as stresses in the labor markets continue to
build. At the same time the need to establish consensus around a new social
contract is critical. Political reform, which would strengthen political inclu-
siveness and accountability, becomes critically linked to economic reform.

Part I: Growth, Reform, and Governance 27
The transition may be costly and needs to be managed. The impact of these
transitions would be an improved business and investment climate, higher
growth, and, most important, substantially greater opportunities for employ-
ment, improved standards of living, and lower poverty for the Arab region.
But costs of transitions are likely, and may be significant. They may entail
greater insecurity and job losses for some groups. Some economic sectors may
be negatively affected. Inequality may increase as wages and returns to educa-
tion become more market oriented. These risks highlight the need for careful
design and sequencing of reform, and the development of mitigating strate-
gies. Careful monitoring and early corrections are needed, but without back-
tracking and stop-and-go policy change.
External versus internal responsibilities. Embarking on such an ambitious agen-
da for political and economic reform is challenging by any measure. But the
biggest challenge is for societies to own such a vision and to develop the con-
fidence that they can achieve it. Pessimism in the region is pervasive. It is pes-
simism about the region’s trading potential. It is fear about competing in
world markets. It is pessimism about the capacity to innovate. It is pessimism
about markets and the private sector. It is pessimism about the ability to
develop an inclusive and accountable governance. But it is unfounded pes-
simism. In recent history, countries of the region have been able to achieve
much, and to change when they mustered the will for doing so. Their histo-
ries have shown time and again that it can achieve success, and it can harness
progress. Pessimism has to be replaced by ambition and confidence. The main
responsibility for meeting the challenges is internal. It has to be with the gov-
ernments and the people of the region.
There is no doubt that external factors and interference have contributed
to the poor state of the region. Violence and conflict have severely impeded
the pace of reform. Persistent regional conflict has large neighborhood effects
that have spilled throughout the region, making it less attractive and more
costly for business and investment, both domestic and foreign. Conflicts have
drained resources to less productive uses in the military and for security. They
have undermined the development of good governance. External partners
bear a responsibility for helping the region establish regional stability and
security. But the central responsibilities remain within the region itself.
Notes
1. The material draws on the diagnosis and conclusions from four major regional reports produced by
the World Bank on the occasion of the World Bank–International Monetary Fund Annual Meetings in
Dubai in September 2003. The reports address the issues of trade and investment, governance, gender,
and employment.
2. The countries included in the Arab region for the purposes of this paper are those that are part of
the MENA region according to the World Bank. They include Algeria, Bahrain, Djibouti, Egypt, Iraq,

Breaking the Barriers to Higher Economic Growth28
Jordan, Kuwait, Lebanon, Libya, Morocco, Oman, Palestine, Qatar, Saudi Arabia, Syria, Tunisia, United
Arab Emirates, and Yemen. Data for Iraq and Libya are often missing, and these two countries are gen-
erally not included in the analysis, save for population or labor force figures. In addition, four Arab
countries not included in the MENA region, Comoros, Mauritania, Somalia, and Sudan, are included
only in Arab totals for population and labor force figures. In 2000, the total population of the Arab
region was estimated at 287 million people. GDP per capita averages US $1,850, ranging from a low of
US$315 in Yemen and Sudan, to a high of US$13,900 in Kuwait.
3. Computed from Barro-Lee Educational attainment database, including Algeria, Bahrain, Egypt,
Iraq, Jordan, Kuwait, Syria, and Tunisia.

29
Reform Complementarities
and Economic Growth in the
Middle East and North Africa
Mustapha Kamel Nabli
Marie-Ange Véganzonès-Varoudakis*
2
This paper explores the relationship between economic reforms and econom-
ic growth in the Middle East and North Africa (MENA) region. Despite
apparent reforms starting in the mid-1980s, the growth performance of the
region has often been disappointing. Since the large fall in international oil
prices in the mid-1980s, most MENA economies have experienced a marked
slowdown and/or macroeconomic crisis. In spite of a small recovery of GDP
per capita in the 1990s (from 1 percent in the 1980s to 1 percent in the 1990s)
the MENA region is, for the second decade,the most slowly growing region in
the world (see Dasgupta et al., 2004).
As a possible explanation, we try to understand in this paper whether the
growth performance of the region has been disappointing because MENA
economies have lagged behind in terms of reforms, or because the growth divi-
dend of the reforms has been small. We illustrate that both explanations are rel-
evant. On the one hand, economic reforms have been insufficient to boost the
growth performance of the countries. This is the case for macroeconomic
reforms during the 1980s, and for structural reforms during the entire period.
On the other hand, our estimations reveal a low impact of some economic
reforms. This is the case when structural reforms are implemented in a volatile
*Centre d'Etudes et de Recherches sur le Développement International and Centre National de la
Recherche Scientifique, Université d’Auvergne, France.
Published in Journal of International Development, January, 2007, Vol. 19. Reprinted by permission
from John Wiley and Sons; see page 17–54.

Breaking the Barriers to Higher Economic Growth30
macroeconomic environment (which corresponds to the situation of the MENA
countries in the 1980s) and when macroeconomic reforms are implemented
without a sufficient level of structural reform (as observed during the 1990s).
This result has been obtained by the introduction of a multiplicative term in our
growth equation. It illustrates the complementarities between economic reforms
as elaborated by Mussa (1987) and by Williamson (1994), and brings new empir-
ical evidence to the subject of economic policy and economic growth.
A first critical step for the paper has been to measure the economic reform
effort of the countries. The originality of our approach has consisted in gen-
erating aggregated reform indicators using principal component analysis.
This methodology permits the aggregation of basic indicators in a more rig-
orous way than would a subjective scoring system (see, for example, the rat-
ing system elaborated by the International Country Risk Guide, ICRG). It also
avoids multicollinearity problems when estimating an equation that includes
several disaggregated indicators. Furthermore, the use of panel data estima-
tion techniques has allowed some comparative analysis among the different
regions, as well as among the MENA countries.1In addition to economic
reforms, the link of human capital and physical infrastructure to economic
growth has also been discussed.
The paper is organized as follows. The second section presents the aggre-
gated indicators of economic reforms, human capital, and physical infrastruc-
ture, and summarizes the progress of these indicators in the MENA region.
Our analysis is based on a panel of 44 developing countries over 1970–80 to
1999.2In the third section, we estimate a growth equation that includes the
different composite indicators. We verify that economic reforms, human cap-
ital, and physical infrastructure have had a positive influence on the growth
pattern of our sample of countries. The fourth section quantifies the contri-
bution of economic reforms, human capital, and physical infrastructure to the
economic growth of the MENA region. We illustrate that the lack of reforms
has had a negative influence on the growth accomplished by the MENA
economies. The last section offers conclusions.
Indicators of Economic Reforms, Human Capital, and Physical Infrastructure
MENA countries differ considerably among themselves, as well as with regard to
the rest of the world. This is, in particular, the case in terms of economic reforms,
physical infrastructure, and human capital. These differences have been assessed
using various indicators that have been aggregated by means of principal com-
ponent analysis. This method has been used to generate five aggregate indicators
(see details on methodology in Nagaraj et al., 2000; appendix 2).3
In this section, we present succinctly the five composite indicators. More
details on their composition and evolution are given in Appendices 2.3–6. The
objective of this section is to highlight the MENA economic strengths and

Part I: Growth, Reform, and Governance 31
weaknesses that support the empirical analysis developed in the rest of the
paper. A more in-depth analysis on economic reforms and their historical con-
text in MENA can be found in the abundant literature on the subject (see in par-
ticular Safadi, 1998; Hoekman and El-Din, 2000; Hakimian and Nugent, 2004).
Macroeconomic Reforms
The first composite indicator of macroeconomic stability (MS) is based on:
• Inflation (p) and public deficit as percentage of GDP (PubDef), which can
be disruptive to investment and growth if they lead to unsustainable
macroeconomic imbalances (see Fischer, 1993; De Gregorio, 1992)
• Foreign exchange parallel market’s premium (lBmp, in logarithm) as a
proxy of various distortions in the economy that can also lead to macro-
economic instability.4
In MENA, macroeconomic stability clearly improved in the 1990s com-
pared to the 1980s, when economies faced great instability (see figure 2.1).5, 6
Our results confirm the findings of several studies on the subject (see in par-
ticular Iqbal, 2001). In some countries, however, macroeconomic reforms
could have shown better progress compared to the rest of the world (with the
exception of Latin America, where stability was not achieved in the 1990s).
Efforts still need to be made, in particular, in Iran, Syria, and Algeria (see
appendix 3 for more details).
External Stability
The second composite indicator of external stability (ES) is represented by:
• External debt as percentage of GDP (DebExGdp), as well as of exports of
goods and services (DebExX), which represent the risk to an economy of
Figure 2.1. Macroeconomic Reform Indicators
⫺0.6
⫺0.5
⫺0.4
⫺0.3
⫺0.2
⫺0.1
0
0.1
0.2
0.3
0.4
MENA
a. Region b. Country
Latin
America
Asia Africa
(CFA)
Africa
(non-CFA)
⫺0.6
⫺0.5
⫺0.4
⫺0.3
⫺0.2
⫺0.1
0
0.1
0.2
0.3
0.4
Iran Syria Jordan Djibouti Tunisia Morocco Egypt
1970–79
1980–89
1990–99
1970–79
1980–89
1990–99
Source: Authors’ calculations.

Breaking the Barriers to Higher Economic Growth32
encountering difficulties in reimbursing its debt and facing a financial
crisis
• Current account in percentage of exports of goods and services (CurAcX),
which gives another signal of the fragility of the external position of the
country.
MENA countries performed rather poorly in term of external stability,
behind Latin America and Asia (see figure 2.2).7This is largely the result of the
unsustainable level of foreign debt, which indicates that a significant scope for
debt reduction exists in the region (see figure 2.11 in appendix 4). Debt reduc-
tion is a concern shared by the majority of MENA countries in our sample;
however, Syria, Jordan, and, to a lesser extent, Morocco need special attention
(see figure 2.12 and a more indepth analysis in appendix 4).
Structural Reforms
The third composite indicator of structural reforms (SR) includes an indicator of
trade policy (TradeP) calculated as the ratio of imports plus exports to GDP, from
which has been deducted the “natural trade openness” of the economies calculat-
ed by Frankel and Romer (1999),8as well as the exports of oil and mining prod-
ucts, which introduce a bias in the sample due to natural resource endowment.
This indicator is based on the fact that trade reforms can be at the origin of
economies of scale and of productivity gains as a result of increased competitive-
ness and increased access to larger markets (Balassa, 1978; Feder, 1983).
Private credit by deposit money banks and other institutions (in percent-
age of GDP, PCrBOG), as a proxy for the development of the banking system,
can have a positive effect on productivity owing to a better selection of invest-
ment projects and to higher technological specialization through a
diversification of risks (see Levine, 1997, for a synthesis).
Figure 2.2. External Stability Indicators
⫺
0.6
⫺
0.4
⫺
0.2
0
0.2
0.4
0.6
⫺0.8
⫺0.6
⫺0.4
⫺0.2
0
0.2
0.4
0.6
0.8
1.0
1.2
MENA Latin
America
Asia Africa
(CFA)
Africa
(non-CFA)
Iran Syria Jordan Djibouti Tunisia Morocco Egypt
1970–79
1980–89
1990–99
1970–79
1980–89
1990–99
a. Region b. Country
Source: Authors’ calculations.

Part I: Growth, Reform, and Governance 33
In terms of structural reforms, MENA seems to perform relatively well
compared to the rest of the world (see figure 2.3). This result is largely the
result of the apparent high financial depth of the MENA economies.
Financial development, however, is usually not considered satisfactory in
MENA (see appendix 5 for a discussion). This is also the case for trade open-
ness, which has on average been particularly low across the region. This
makes of structural reform a sector with a large potential for improvement in
MENA (see figures 2.14a and 2.15a, and appendix 5 for a more in-depth
analysis).
Human Capital
The fourth composite indicator of human capital (MH1) is represented by the
logarithm of:
• The infant mortality rate (lMort) as a proxy of the health conditions of the
population,
• The number of years of primary schooling of the population (lH1).
Both health and education increase the productivity of physical capital and
can be at the origin of positive externalities (Lucas, 1988; Psacharopoulos,
1988; Mankiw et al., 1992).9
In MENA, human capital improved significantly throughout the period.
Progress was among the best across regions (see figure 2.4a). Good perform-
ances are noticeable everywhere (see figure 2.4b). MENA human capital
development, however, still lags behind that of Asia and Latin America. This
means that a large potential gain can be expected from an increase of health
and education in the region.
Figure 2.3. Structural Reform Indicators
1.2
0.8
0.4
0
⫺0.4
⫺0.8
⫺1.2
1.2
0.8
0.4
0
⫺0.4
⫺0.8
⫺1.2
a. Region b. Country
MENA Latin
America
Asia Africa
(CFA)
Africa
(non-CFA)
Iran Syria Jordan Djibouti Tunisia Morocco Egypt
1970–79
1980–89
1990–99
1970–79
1980–89
1990–99
Source: Authors’ calculations.

Breaking the Barriers to Higher Economic Growth34
Core Infrastructure
The fifth composite indicator of physical infrastructure (Phys) is based on the
logarithm of:
• The density of the road network (lRoads, in km per km2)
• The number of telephone lines per 1000 people (lTel).
The complementarities among physical infrastructure and physical and
human capital lead to higher productivity and increase the incentive to invest
(see Barro, 1990; Aschauer, 1989; Murphy et al., 1989).
Despite real progress throughout the period, MENA’s endowment in infra-
structures has remained insufficient (see figure 2.5). The road network has
Figure 2.4 Human Capital Indicators (reduceda)
Figure 2.5. Physical Infrastructure Indicators
1.5
1.0
.5
0
⫺.5
⫺1.0
⫺1.5
⫺2.0
1.0
.5
0
⫺.5
⫺1.0
⫺1.5
⫺2.0
a. Region b. Country
MENA Latin
America
Asia
Africa
(CFA)
Africa
(non-CFA)
Iran Syria Jordan Djibouti Tunisia Morocco Egypt
1970–79
1980–89
1990–99
1970–79
1980–89
1990–99
⫺2.0
⫺1.5
⫺1.0
⫺0.5
0
0.5
⫺2.0
⫺1.5
⫺1.0
⫺0.5
0
0.5
1.0 1.0
a. Region b. Country
MENA Latin
America
Asia Africa
(CFA)
Africa
(non-CFA)
Iran Syria Jordan Djibouti Tunisia Morocco Egypt
1970–79
1980–89
1990–99
1970–79
1980–89
1990–99
Source: Authors’ calculations.
a. This indicator includes infant mortality and primary schooling only.
Source: Authors’ calculations.

Part I: Growth, Reform, and Governance 35
been very deficient (see figure 2.20 in appendix 7), and the telephone network
also inadequate compared to Latin America and Asia (see figure 2.21 in
appendix 7). Actually, closing the gap with more advanced developing coun-
tries constitutes a significant challenge for MENA (see appendix 7 for a more
detailed analysis).
Growth Diagnostic for Our Sample of MENA Countries
Despite the noticeable progress in various areas of reform, MENA growth per-
formance has shown disappointing results. Although growth improved
almost everywhere during the 1990s, MENA results have been largely sur-
Figure 2.6. GDP per Capita Growth Rates (%)
⫺1.5
⫺0.5
0.5
1.5
2.5
3.5
⫺8
⫺6
⫺4
⫺2
0
2
4
6
a. Region b. Country
MENA Latin
America
Asia Africa
(CFA)
Africa
(non-CFA)
Iran Syria Jordan Djibouti Tunisia Morocco Egypt
1970–79
1980–89
1990–99
1970–79
1980–89
1990–99
Source: Authors’ calculations.
Figure 2.7. GDP per Capita (international dollars)
0
500
1,000
1,500
2,000
2,500
3,000
3,500
4,000
4,500
500
1,500
2,500
3,500
4,500
5,500
a. Region b. Country
MENA Latin
America
Asia Africa
(CFA)
Africa
(non-CFA)
Iran Syria Jordan Djibouti Tunisia Morocco Egypt
1970–79
1980–89
1990–99
1970–79
1980–89
1990–99
Source: Authors’ calculations.

Breaking the Barriers to Higher Economic Growth36
passed by the Asian economies, where GDP growth rates reached 4 percent a
year on average (1.7 percent in MENA; see figures 2.6 and 2.7).
These low growth performances explain why MENA GDP per capita
increasingly lags behind that in Asia and Latin America (see figure 2.7a).
Economic Reforms, Human Capital, Physical Infrastructure, and Growth
In this section, we address the link between economic reforms and economic
growth by estimating a conditional convergence equation [Equation (1)]
inspired by Barro and Sala-I-Martin (1995). In this equation, the long-run
GDP per capita growth rate [Δln(yi,t)] depends on the logarithm of the initial
level of the GDP per capita [ln(yi, t–1)] and on a set of other variables. These
variables aim at explaining the differences across countries in the long-term
level of GDP per capita. In our case, we have added—in addition to the ratio
of investment to GDP, which represents the increase in the productive capac-
ities of the economy—the composite economic reform indicators, as well as
the physical infrastructure and human capital indicators10, 11:
Δln(yi,t) = c + a*ln(yi,t–1)+b*ln(Invi,t)+d1*(MSi,t)+d2*
(ESi,t)+d3*(SRi,t)+el*(Physi,t)+e2*(MH1i,t)+ εi,t
(1)
with: yi,t – 1 = GDP per capita of the previous period
Invi,t = investment ratio to GDP
MSi,t = macroeconomic stability indicator
ESi,t = external stability indicator
SRi,t = structural reform indicator
Physi,t = physical infrastructures indicator
MH1i,t = human capital indicator
c= intercept, a,b,d1to d3,e1to e2, = parameters, t= time index and
εt=error term.
A variant of this equation was introduced: two multiplicative terms
(MS*SR and ES*SR) to capture for the possible complementarities amongst
reforms [Equation (2)]. This case has been developed by Mussa (1987) and
Williamson (1994), who have elaborated, in particular, the idea that structur-
al reforms are not efficient if the economy is not stable. This situation was
observed on several occasions during the 1980s and the 1990s in various
Eastern European and Latin American countries.In these countries, the intro-
duction of structural reforms, such as trade and financial liberalization, in an
unstable macroeconomic environment resulted in economic crisis. The mul-
tiplicative term also allows capture of the possible reverse effect: of the
efficiency of external and macroeconomic stability increases with the level of
structural reforms. This could be relevant to the MENA economies in which
structural reforms have generally been insufficient.

Part I: Growth, Reform, and Governance 37
Δln(yi;t)= c + a*ln(yi;t–1) + b*ln(Invi,t) + d1*(MSi,t) + d2*(ESi,t) + d3*(SRi,t)
+ d4*(MSi,t*SRi,tt) + d5*(ESi,tSRi,t) + e1*(Physi,t) + e2*(MH1i,t) + εi,t
(2)
The first step in the econometric analysis was to estimate the degree of inte-
gration of the variables entering Equations (1) and (2). Co-integration of
series is actually required if Equations (1) and (2) have to be estimated with
the standard methods.
Table A5-1 in appendix 9 provides the results of the Augmented-Dickey-
Fuller (ADF) test of the series entering the two equations. We used the Im et
al. (1997) methodology, which provides critical values of ADF tests in the case
of heterogeneous panel data. The results indicate that series are generally sta-
tionary at the 1 percent level (5 percent in the case of human capital). This
allows running Equations (1) and (2) using the standard methods.12
Equations (1) and (2) have been estimated for our unbalanced panel of 44
developing countries. The regressions have used the White estimator to cor-
rect for the heteroscedasticity bias. To control for sample heterogeneity, coun-
try dummy variables have also been introduced. These variables reflect differ-
ences in the quality of institutions or in different endowments of natural
resources,13 which can be at the origin of large discrepancies in the “natural
propensity” to grow. This hypothesis is supported by the data, as shown by the
value of the Fischer tests of equality of the intercepts across countries, as well
as by the value of the Hausman tests as far as the random hypothesis is con-
cerned.14 Results are presented in table 2.1.
The results of the estimations show that the relationship between GDP per
capita growth rates and its determinants is consistent with the theory.
Increases in the investment ratio, macroeconomic and external stability, struc-
Table 2.1. Estimate Results of the Growth Equations (1) and (2) Dependent Variable Δln(yt)
Independent variables Eq1 Eq2
ln(yt– 1 ) –0.17 (8.62)*** –0.17 (8.96)***
In(invt) 0.065 (5.27)*** 0.066 (5.38)***
MSt(1) 0.004 (1.76)* 0.028 (2.76)**
ESt(1) 0.03 (5.3)*** 0.027 (3.95)***
SRt(2) 0.015 (2.05)** 0.011 (1.56)
MSt(1)* SRt(2) 0.027 (2.65)**
ESt(1)* SRt(2) 0.006 (0.69)
Physt(2) 0.03 (2.07)** 0.031 (2.07)**
MH1t0.018 (1.70)* 0.016 (1.66)*
Adjusted R20.29 0.30
Fischer test 4.1 4.3
Hausman test 56.2 67.0
Source: Authors’ estimations.
Note: Student t statistics are within brackets. The number of observations used in the regressions is 589. Data have
been compiled from World Development Indicators, Global Development Finance, Global Development Network,,
and Live Data Base (World Bank).
(1): one lag; (2): two lags; (*) indicates significance at 10 percent; (**) indicates significance at 5 percent; (***) indi-
cates significance at 1 percent.

Breaking the Barriers to Higher Economic Growth38
tural reforms, physical infrastructure, and human capital lead to higher
growth rates.15
The impact of most variables on growth is, however, not instantaneous.
This is the case with macroeconomic and external stability, as well as with
structural reforms (respectively one and two lags). This means that structural
reforms need some time (two years, according to our estimations) to materi-
alize into growth. As expected, the effects of macroeconomic and external
reforms are faster (one year). The same conclusions can be drawn for human
capital and physical infrastructure (no lag and two lags, respectively).16
Another, and even more interesting, feature can be observed in the estima-
tion of Equation (2), which corroborates the sequencing of reforms developed
by Mussa (1987) and Williamson (1994): the introduction of the multiplica-
tive term MS*SR (macroeconomic multiplied by structural reforms) turns out
to be significant, while the structural reforms indicator alone (SR) is no longer
significant (see table 2.1). This result highlights the complementarities
between macroeconomic stability and structural reforms, as explained below.
On one hand, the efficiency of structural reforms depends on success in
stabilizing the economy. In other words, reforming the economy materializes
into growth if applied in a stable macroeconomic environment. In an increas-
ingly volatile environment, a high level of structural reforms increases the dis-
ruptive effect of macroeconomic instability. This means that structural
reforms should take place at least at the same time as macroeconomic
reforms, if not after.
On the other hand, the opposite is partly true. If stabilization programs
make more progress when the economy is reformed, and if the growth out-
come of macroeconomic stabilization is linked to the level of the structural
reforms (through the multiplicative term MS*SR), then macroeconomic
reform materializes into growth even in the absence of structural reforms
(through the MS index alone). This is not the case of structural reforms that
do not materialize into growth in an unstable macroeconomic environment
(the coefficient of structural reforms alone not being significant). This stress-
es the importance of macroeconomic reforms for the growth prospects and
the reform processes of the countries. This result clearly shows that macroeco-
nomic stability remains a priority for the government to address, and that
reforming the economy should not be undertaken prior to stabilizing.
Estimation of Equation (2) also reveals, in another way, the critical role of
macroeconomic stability in the growth performances of the economies. The
coefficient of the macroeconomic reform index is higher and more significant
than when estimating Equation (1) (see table 2.1). The coefficient of the mul-
tiplicative term is of the same magnitude as that of the macroeconomic
reforms index alone. This means that both effects are similarly significant.
This will be illustrated in the next section, when calculating the contribution
of each reform indicator to the economic growth of our sample of countries.

Part I: Growth, Reform, and Governance 39
This result, however, does not hold for external stability for which the mul-
tiplicative term (ES*SR) is not significant. The absence of correlation between
external stability and structural reforms could be interpreted as an indication
that structural reforms are not undertaken unless the external environment
has been adjusted. This is because our indicator of structural reforms incor-
porates improvement of trade reform, which is unlikely to be sustained in the
presence of external unbalance. These results show that external stability
remains an important initial condition for reforming the economy, as well as
for the growth prospects of the countries (the external stability index alone
being still significant; see table 2.1).
Finally, the robustness of our estimations can be evaluated by the fact that
the coefficients of the other variables are similar in both estimations. In the
rest of this document, we will focus on the results of Equation (2), which give
a more complete picture of the relation between reform and growth.
Assessing the Impact of Economic Reform, Human Capital, and Physical
Infrastructure on the Growth Performance of the MENA Region
As noted above, the use of aggregate indicators of economic reform, human
capital, and physical infrastructure overcomes the difficulties of estimating
the impact of a large number of indicators that may have collinear relation-
ships. This method allows for subsequent calculations of the contribution of
the initial indicators to the growth performance of the countries. The calcula-
tion is based on the estimated coefficients of the aggregate indicators in the
regression, as well as on the weights of each principal component in the aggre-
gate indicator combined with the loadings of the initial variables in each prin-
cipal component (see appendix 10 for more details on the methodology).
Table 2.2 presents the long-term coefficients/elasticities of the initial
reform indicators (see tables A11-1 and A11-2 in appendix 11 for the short-
term coefficients/elasticites). A number of conclusions can be drawn from
these calculations.
Physical Infrastructure and Human Capital
A first set of conclusion shows that improvements in human capital and phys-
ical infrastructure present a strong and approximately equal potential impact
on growth. In both equations, primary education, the road network, and the
health conditions of the population show strong elasticities. These variables
appear to be key factors in the growth prospects of our sample of countries.
Although studies on the impact of core infrastructure rarely use physical indi-
cators (such as the road or the telephone network),17 our results are quite in
line with the findings of some authors who have followed the same approach.18
These coefficients/elasticities have been used to calculate the contribution of
the different factors to the growth performances of the region. As shown,

Breaking the Barriers to Higher Economic Growth40
MENA’s development effort in health and education has been consequential. It
can be calculated that improvement in primary schooling contributed an annu-
al average of 0.3 to 0.4 points of GDP per capita growth rate during the period.
For the amelioration of health conditions, the contribution is even higher (0.8
and 0.5 points in the 1980s and the 1990s, respectively; see table 2.3).
Globally, improvements in human capital have explained between 0.9 and
1.1 points of growth per capita per year since the beginning of the 1980s. This
means that, without progress in human capital, GDP per capita annual
growth rate would have been -1.1 percent in the 1980s (instead of 0 percent)
Table 2.2. Long-Term Coefficients/Elasticities
Index Variables Equation (1) Equation (2)
MS alone** P ⫺0.004 ⫺0.025
ln(Bmp)⫺0.004 ⫺0.029
PubDef ⫺0.056 ⫺0.443
ES DebExX ⫺0.101 ⫺0.089
DebExGni ⫺0.003 ⫺0.003
CurAccX 0.453 0.396
SR alone*** TradeP 0.332 0.230*
PCrBOG 0.363 0.252*
Phys ln(Roads) 0.149 0.145
ln(Tel) 0.106 0.104
MH1 ln(Mort)⫺0.147 ⫺0.123
ln(H1) 0.169 0.142
Source: Authors’calculations.
*Not significant.
**In Equation (2), macroeconomic reforms impact growth through two channels: one is direct; the other depends
on the level of structural reforms. Elasticities/coefficients have not been calculated in this last case, due to the
difficulties linked to the multiplicative term (MS*SR). Nevertheless, the global impact of macroeconomic reforms will
be assessed for our different countries/groups of countries through their contribution to growth (see below).
***As for macroeconomic reforms, because the impact of structural reforms depends on the level of macroeco-
nomic stability, the coefficient/elasticity of the multiplicative term (MS*SR) has not been calculated. The contribu-
tions, however, will be presented in the section on contributions.
Table 2.3. Contribution of Human Capital— MENA Region
MENA Primary Total Growth of GDP
Years % education Health contribution per capita
1980–89 Annual growth rate 2.8 ⫺6.3 0
1990–99 2.3 ⫺4.5 1.7
Elasticity 0.14 0.12 Without H
1980–89 Contribution to 0.4 0.8 1.1 ⫺1.1
1990–99 growth 0.3 0.5 0.9 0.8
Regional summary
Years MENA Asia LA ACFA ANCFA
1980–89 1.1 0.7 0.8 0.7 0.6
1990–99 0.9 0.7 0.7 0.6 0.4
Source: Authors’calculations.

Part I: Growth, Reform, and Governance 41
and 0.8 percent in the 1990s (instead of 1.7 percent; see table 2.3). This con-
tribution is even higher in the case of Iran, Syria, and Algeria, because of the
initial gap in primary schooling compared to more advanced MENA coun-
tries (1.2 to 1.3 in the 1980s and 0.9 to 1.2 in the 1990s; see table A12-4 in
appendix 12). These results clearly highlight the strong contribution of
human capital to the growth performances of the economies.
The contribution of infrastructure to growth also appears to be substantial.
This has been the case despite MENA’s relatively low level of infrastructure.
This result is mainly the result of the rapid increase in the number of tele-
phone lines, which annually has accounted for 0.7 points of GDP per capita
growth rate.19 Globally, from the beginning of the 1980s, physical infrastruc-
ture has annually contributed from 1.0 to 1.4 points of the GDP per capita
growth rate (see table 2.4). This contribution has even been higher in Iran and
Syria (around 1.8 and 1.5, respectively; see table A12-5 in appendix 12). This
result also implies that infrastructure and primary education constitute a
potentially important source of growth for MENA, because the region lags in
comparison to more advanced developing economies (such as Asia and Latin
America) in these two fields of activity.
The importance of human capital and physical infrastructure for the
growth performance of the other regions is also illustrated in tables 2.3 and
2.4. The rapid progress of MENA health conditions accounts for a higher con-
tribution of human capital in that region. Closing the gap in education and
infrastructure with more advanced regions remains, however, a concern for
MENA.
Macroeconomic Stability and Structural Reforms
As far as macroeconomic stability is concerned, our calculations also stress the
significance of this factor for the growth prospects of our sample of
economies. As noted above, one part of this effect is direct. The other part
Table 2.4. Contribution of Physical Infrastructure—MENA Region
MENA Roads Total Growth of GDP
Years % Tel lines networks contribution per capita
1980–89 Annual growth rate 7.2 3.0 0
1990–99 7.5 1.8 1.7
Elasticity 0.10 0.15 Without Phy
1980–89 Contribution to 0.8 0.6 1.4 ⫺1.4
1990–99 growth 0.8 0.3 1.0 0.7
Regional summary
Years MENA Asia LA ACFA ANCFA
1980–89 1.4 1.5 0.9 0.4 0.6
1990–99 1.0 1.1 0.8 0.9 0.8
Source: Authors’calculations.

Breaking the Barriers to Higher Economic Growth42
depends on the level of structural reforms. Table 2.5 summarizes these two
effects, from which several conclusions can be drawn.20
During the 1980s, macroeconomic volatility in MENA was disruptive to
growth. This was mainly the result of the sharp increase in the foreign
exchange parallel market premium in some MENA countries (Iran, Algeria,
and Syria).21 Inflation and public deficit increased moderately, but stayed at a
relatively high level (especially in Iran and Syria, as well as for inflation in
Algeria). Both the direct and indirect effects contributed negatively, and with
the same magnitude, to growth (0.6 to 0.7 points per year of GDP per capita
growth, for a total of 1.3 points). This means that the negative effect of macro-
economic instability more than compensated for the positive impact of
human capital or physical infrastructure (see tables 2.3 and 2.4). This negative
contribution was even higher in Iran and Syria (2.6 and 2.4, respectively).
This finding stresses again the importance of macroeconomic stability for
the growth prospects of our sample of economies. In Latin America, inflation
and public deficit, as well as public deficit and foreign exchange parallel mar-
ket premiums in non-Communauté Francophone d’Afrique (CFA) countries,
contributed to the economic turmoil of the period (1.4 to 0.3 points, respec-
tively, of the GDP per capita growth rate; see table 2.5 and figures 2.8 to 2.10
in appendix 3).
In the 1990s, this situation reversed, and economic stabilization con-
tributed positively to MENA growth performance. Progress, however—which
essentially concerned the public deficit—could have been more significant.
The contribution of improved macroeconomic conditions reached only 0.4
points of GDP for the two effects (0.2 for the direct, 0.2 for the indirect). This
can be explained by several factors.
First, the macroeconomic stability of the MENA economies, whatever the
level, could have been further improved. This is the case for the foreign
exchange parallel market premium in Iran, Algeria, and Syria; as well as, to a
lesser extent, for inflation in Iran, Algeria, Egypt, Syria; and public deficit in
Tunisia and Morocco.
Second, our analysis has shown that structural reforms in MENA have
always lagged behind the experiences of more dynamic region such as Asia,
particularly in the 1990s. This lack of economic reform has already been
commented on in the literature. It has often been attributed to political eco-
nomic considerations (see, for example, Hoekman and Kheir El Din, 2000,
and Hakimian and Nugent, 2004).22 In fact, the slow pace of structural
reform has been an important factor limiting the benefits of macroeconom-
ic stabilization.23 The case of the MENA countries thus illustrates the com-
plementarities between macroeconomic stability and structural reforms.
These results also confirm that both macroeconomic stability and structural
reforms have still to progress if MENA wants to catch up with more advanced
developing countries.

Part I: Growth, Reform, and Governance 43
Conversely, the contribution of structural reforms—0.2 points in GDP per
capita growth rate during the two subperiods, despite the low level of reforms
and the macroeconomic instability of the 1980s (see table 2.6)—stresses that
reforming the economy is potentially a key factor for the growth prospects of
the MENA region. This is confirmed by the strong contribution of structural
reforms to the growth performances of the Asian countries (0.7 to 0.9 points;
see table 2.6), as well as by the experience of Tunisia and Jordan (0.5 to 0.7
points; table A12-3 in appendix 12). The important gap between MENA and
Asia shows that this factor constitutes another important potential source of
growth for the region.
External Stability
As for macroeconomic stability, the external instability of the 1980s strongly
contributed to the economic turmoil of the period. The increase in the debt
ratio led to an average deterioration of 1 percentage point in the annual GDP
per capita growth rate (see table 2.7). This deterioration was even greater in
Syria (2.4 points: table A12-2 in appendix 12). Globally, imbalances in exter-
nal stability cost the MENA countries an average of 1 percentage point per
year in GDP per capita growth rate. This means that the growth performance
of the region could have reached one percent on average per year (instead of
0 percent).
Table 2.5. Contribution of Macroeconomic Stability—MENA Region
MENA Total Growth of GDP
Years % p bmp Pubdef contribution per capita
1980–89 Annual growth rate 0.4 22 0.03 0
1990–99 ⫺0.2 2.6 ⫺0.6 1.7
Coef/elasticity ⫺0.025 ⫺0.029 ⫺0.443 MS alone With(2)/without(3) MS
1980–89 Contribution to ⫺0.01 ⫺0.6 ⫺0.01 ⫺0.7 0.7
1990–99 growth 0.01 ⫺0.1 0.26 0.2 1.5
Coef/elasticity ⫺0.024 ⫺0.028 0.427 MS*SR(1) With(2)/without(3)
MS*SR
1980–89 Contribution to ⫺0.01 ⫺0.6 ⫺0.01 ⫺0.6 0.6
1990–99 growth 0.01 ⫺0.1 0.25 0.2 1.5
Total With(2)/without(3)MS
1980–89 Contribution to ⫺0.02 ⫺1.2 ⫺0.03 ⫺1.3 1.3
1990–99 growth 0.01 ⫺0.15 0.5 0.4 1.3
Regional summary
Years MENA Asia LA ACFA ANCFA
1980–89 ⫺1.3 0.1 ⫺1.4 0.05 ⫺0.3
1990–99 0.4 0.5 0.8 0.2 1.0
Source: Authors’calculations.
(1) Without change in SR.
(2) With MS in the case of a negative impact on growth.
(3) Without125 MS in the case of a positive impact.

Breaking the Barriers to Higher Economic Growth44
During the 1990s, the process of debt reduction contributed positively to
the growth performance of the region (0.2 point per year in GDP per capita).
This low contribution, however, and the gap with other regions (Asia in par-
ticular) indicate that debt reduction still represents an important potential
source of growth.
Special attention needs to be drawn to the current account balance. The
strong long-term coefficient potentially makes of this variable a key factor that
can lead to large variations in the rate of growth of the economies. This was
the case in the 1990s, when improvement in the current account position con-
tributed substantially to the growth performance of the region (0.8 point of
GDP per capita growth rate). This improvement was even higher in Egypt (1.3
points) and Iran (1 point). During this period, MENA showed the best per-
formance among the regions. Table 2.7 also confirms that external stability
Table 2.6. Contribution of Structural Reforms (Structural Reforms*Macroeconomic Stability: SR*MS(1))—MENA Region
MENA Total Growth of GDP
Years % TradeP*MS(1) PCrBOG*MS(1) contribution per capita
1980–89 Annual growth rate ⫺0.1 0.8 0
1990–99 0.7 0.01 1.7
Coefficients 0.24 0.27 Without SR
1980–89 Contribution to ⫺0.03 0.2 0.2 ⫺0.3
1990–99 growth 0.2 0.0 0.2 1.5
Regional summary
Years MENA Asia LA ACFA ANCFA
1980–89 0.2 0.9 0.2 0.1 0.2
1990–99 0.2 0.7 0.3 ⫺0.3 0.3
Source: Authors’calculations.
a. Without change in macroeconomic reforms.
Table 2.7. Contribution of External Stability—MENA Region
MENA Total Growth of GDP
Years % DebX DebGNP CurAcc contribution per capita
1980–89 Annual growth rate 9.9 3.8 ⫺0.3 0
1990–99 ⫺1.8 1.4 2.1 1.7
Coefficients ⫺0.089 ⫺0.003 0.396 With/without ES
1980–89 Contribution to ⫺0.9 ⫺0.01 ⫺0.1 ⫺1.0 1
1990–99 growth 0.2 ⫺0.004 0.8 1.0 0.7
Regional summary
Years MENA Asia LA ACFA ANCFA
1980–89 ⫺1.0 0.2 ⫺1.4 ⫺1.0 ⫺1.9
1990–99 1.0 0.5 0.9 ⫺0.9 ⫺0.2
Source: Authors’calculations.

Part I: Growth, Reform, and Governance 45
has constituted a key factor in explaining the growth achievements of all the
regions in our sample.
Summary
Table 2.8 summarizes the global impact of economic reforms, human capital,
and physical infrastructure on the growth performances of our sample of
countries. In the 1980s, macroeconomic and external instability contributed
significantly to MENA’s low growth performances. These factors contributed
to lowering the annual GDP growth rate by 2.2 percentage points. Growth
could have reached 2.2 percent per year (instead of 0 percent) if no degrada-
tion of macroeconomic and external conditions had occurred. This negative
contribution was even stronger in the case of Syria (5.4 points) and Iran (2.6
points for macroeconomic stability alone; see table A12.6 in appendix 12).
As far as positive contributions are concerned, human capital, physical
infrastructure, and, to a lesser extent, structural reforms contributed during
the same period to a 2.7 percent annual average GDP per capita growth rate.
This positive influence more than compensated for the negative impact of
macroeconomic and external instability. This positive contribution was even
higher in the case of Tunisia, Iran (3.1 points), and Syria (2.8 points).
In the 1990s, economic reforms, and growth in human capital and physical
infrastructure, together explain the improvement in the economic situation of
MENA. Human capital and physical infrastructure still contributed the most
(1.9 percent in GDP per capita growth rate), followed by macroeconomic and
external stability (1.4 points in GDP per capita growth rate).24 This finding
reflects the efforts of the MENA countries to reform their economies.
Reforms, however, have still lagged behind Asia and Latin America. Indeed,
the low level of structural reforms, the still high debt ratio, the foreign
Table 2.8. Total Contributions—Summary MENA Region
Contribution to growth
% Years MS + ES H + Phy + SR Investment Total
1980–89 ⫺2.2 2.7 ⫺0.3 0.1
1990–99 1.4 2.1 ⫺0.1 3.4
Regional
summary MENA Asia LA ACFA ANCFA
Total contribution
to growth
1980–89 0.1 3.6 ⫺1.0 0.2 ⫺1.0
1990–99 3.4 3.6 3.5 0.4 2.4
GDP per capita
growth rate
1980–89 0 3.4 ⫺0.8 ⫺1.2 ⫺0.9
1990–99 1.7 3.9 1.7 ⫺0.8 0.9
Source: Authors’calculations.

Breaking the Barriers to Higher Economic Growth46
exchange parallel market premium in some countries, the insufficient devel-
opment of infrastructure and primary schooling compared to Asia and Latin
America, as well as the high potential of reforms, highlight the urgent need for
MENA to accelerate the reform agenda in the region.
Conclusion
The empirical analysis carried out in this paper has clearly underlined the
importance of economic reforms, human capital, and physical infrastructure
to improve the growth prospects of the economies. These factors have been
shown to have a powerful impact on growth. They have greatly contributed to
the growth process in our sample of MENA countries. An original contribu-
tion of this paper has been to also bring to light new empirical evidence on
the subject of economic reforms and economic growth. Our quantitative
analysis has revealed the complementarities across macroeconomic reforms
and structural reforms, as elaborated by Mussa (1987) and Williamson
(1994). These complementarities have greatly contributed to the low growth
performance of our sample of MENA countries.
During the 1980s, macroeconomic volatility in MENA was disruptive to
growth. In the 1990s, this situation was reversed and economic stabilization
contributed positively to MENA’s development. Our empirical analysis shows,
however, that although most MENA countries implemented better macroeco-
nomic policies in the mid-1980s or in the 1990s, consolidating macroeconom-
ic stability is still a priority for the success of structural reforms, as well as for
successful competition with more successful developing countries.
The region is also concerned with achieving progress in structural reforms,
which have always lagged behind faster-growing countries in terms of trade
openness and financial development. In the 1990s, the slow pace of these
reforms limited the benefits of macroeconomic stabilization. In fact, and as
illustrated by our econometric results, improving trade openness and finan-
cial development would strongly contribute to the economic growth of
MENA countries, in addition to facilitating the integration of the region into
the world economy.
As for macroeconomic and structural reforms, external stability has been
identified as another factor affecting growth performances in MENA. The
external instability of the 1980s strongly contributed to the economic turmoil
of the period. In the 1990s, the renegotiation of the external debt helped to
improve the growth process in the region. The gap in terms of external debt
and the significant scope for debt reduction indicate, however, that this still
represents a potentially important source of growth for the future.
The last set of conclusions concerns the substantial role of human capital
and physical infrastructure, improvements that have greatly contributed to
MENA growth performances. Progress achieved in the health, education, and

Part I: Growth, Reform, and Governance 47
infrastructure sectors has been substantial in MENA. The level of education,
however, remains lower than in other regions—such as Asia and Latin
America—and the amount of infrastructure equipment available is still
insufficient. The development gaps in education and infrastructure represent
areas of potentially strong improvement in the region’s growth prospects. This
is particularly true in the current context of economic reform and integration
into the world’s global economy, particularly with the European Union. In
fact, a more qualified labor force and an adequate level of infrastructure are
essential for the success of economic reforms in MENA, as well as to stimulate
and develop international competitiveness.
Notes
1. The comparative advantage of panel data regressions compared to time series estimation techniques
can also be seen in:
a. The double dimension (time series-cross-section), which improves estimates by adding informa-
tion
b. The country dummy variables, which generally ask for an important number of degree of freedom,
and which add precision to the results of the estimations.
2. Among these countries, 16 are African countries (7 CFA and 9 non-CFA), 12 are Latin American
countries, 9 are Asian countries, and 7 are MENA countries (see appendix 1 for the list of countries).
These economies have been selected based on their level of income per capita. To preserve the coher-
ence of the sample, we have chosen mostly intermediate-income countries that are comparable with
MENA economies.
3. As part of our empirical work, we have tried, without success, to introduce in the principal compo-
nent analysis the ratio of current account balance to GDP as a component of the macroeconomic sta-
bility index, exchange rate stability as part of the external stability, cumulative privatization receipts as
a factor in the structural reforms, and real exchange rate misalignment processed by Nabli and
Véganzonès-Varoudakis (2002). Other interesting indicators had to be ignored because of lack of
information. This has been the case with the ratio M2 to GDP, the import cover of international
reserves, the ratio of short-term to total debt, and net international liquidity as months of import
cover, which could have reinforced the external stability index. Similarly, the structural reform index
could not benefit from information on mean tariff rates, or highest marginal individual and corporate
taxes.
4. The Bmp indicator is traditionally a measure of the distortions on the capital markets, which can
hinder the mobilization of resources for investing, especially in tradable goods. This indicator is also
used as a proxy for real exchange rate (RER) misalignment,and particularly for RER overvaluation. In
fact, many governments, to ration scarce foreign exchange to the private sector,use exchange controls.
The excess of foreign exchange demand arising from the official exchange rate being kept below its
market clearing level leads to an overvaluation of the currency (see Pinto, 1990, for an analysis of the
various foreign exchange markets’ distortions captured by Bmp).
5. We have listed the MENA economies in decreasing order of welfare. The first countries presented in
the graphs are the ones that exhibited higher GDP per capita in the 1970s.
6. As a result of the methodology of calculation of principal component analysis, the level and sign of
the aggregate indicators have no particular interpretation. These indicators have to be read as follows:
a. A rise in the indicator indicates progress in reform. Conversely,a deterioration denotes a regression.
b. The intensity of reforms is measured as the difference of the indicator across countries and
between periods. In the last section, we will use the annual growth rate of the composite, as well as
of the initial indicators, to measure their impact on growth.

Breaking the Barriers to Higher Economic Growth48
7. These results are partly due to the exclusion of the Gulf economies, in which the average debt ratio
has been lower.
8. The “natural openness” of the economy is calculated by Frankel and Romer (1999) by taking into
account the size and the distance of the markets of the countries.
9. We have also processed a more complete human capital indicator (H) that includes, as well, second-
ary and superior schooling (see appendix 2, table A2.4.a). This indicator does not give better results
when estimating the growth equation. We will, however, give some details on the outcomes in second-
ary and superior education in Appendix 6.
10. It is worth noting that the reduced form used here (which includes the ratio of investment to GDP)
is derived from a log-linear approximation around the steady state of a Solow (1967) type growth
model. This model has been extended by several authors, by incorporating other explanatory variables
in order to better explain the residual of the equation (the technical progress—see Mankiw, Romer,
and Weil, 1992, as far as human capital is concerned). In this paper, we have chosen the same approach.
We should, however, point out that reforms also materialize into growth by increasing the investment
ratio. This should be kept in mind when analyzing our estimation results, which, in this context, may
have underestimated the impact of reforms on growth.
11. This method, consisting of directly introducing the initial reform indicators into the regressions,
has been tried with less success. Because of the correlation of the disaggregated variables, the
significance of some of the variables turned out to depend on the specification of the model. These
initial tests have not been reported here because of the large number of combinations.
12. This is also the case because our sample is sufficiently ‘’big,’’ with T(number of time periods) and
N(number of countries) being large enough (see Im, Pesaran, and Shin, 1997).In this case, despite the
presence of the lag variable y– 1, which is correlated with the constant terms, the distributions of the
tests tend to converge.
13. This applies in particular to the oil-producing and mining economies in our sample.
14. Other estimations have consisted in testing the heterogeneity of the estimated relationship across
regions. We have in particular tested the difference of slope of the economic reforms, human capital,
and physical infrastructure indicators.
15. The coefficient of macroeconomic stability in Equation (1), and of human capital in Equations (1)
and (2) are significant at the 10 percent level.
16. The lag of the reform variables also corrects for the potential bias due to the possible endogeneity
of the right-hand variables. In this case, the reverse causality that might exist is not supposed to bias
the value of the estimated coefficients.
17. Most of them employ aggregated indicators such as public investment.
18. Comparisons among studies are, however, also difficult because of the differences in indicators
and/or of specifications used.
As far as the road network is concerned, Pouliquen (2000)—in his study of Indian infrastructure
at the village level—found a short-term elasticity of 0.07 for the 1971-1981 period (here 0.03).
Nagaraj, Varoudakis and Véganzonès (2000) obtained an elasticity of 0.05 at the Indian States level
studied from 1970 to 1994, and Véganzonès (2001) found an elasticity of 0.06 for a panel of 87 coun-
tries during 1970-1995. Conversely, Nagaraj, Varoudakis and Véganzonès (2000) have come out with
a long-term elasticity of 0.34 and Véganzonès (2001) of 0.16 (here 0.15). The impact of the develop-
ment of the telephone network has been evaluated by, among others, Canning (1999) for a large sam-
ple of countries during 1965-1990, using cross-countries regressions. Canning has estimated a short-
term elasticity of 0.013 (here 0.019) and a long-term elasticity of 0.15 (here 0.10). Véganzonès (2001)
has obtained a short-term elasticity of 0.007 and a long-term one of 0.18. As far as education is con-
cerned, Benhabib and Spiegel (1994) have estimated cross-countries regressions for the 1965-85 peri-
od. They have obtained a short-term elasticity of 0.01 for total education (here 0.025 for primary edu-
cation) and a long term one of 0.04 (here 0.14). Véganzonès (2001) has found a short-term elasticity
of 0.01 for primary education and a long-term one of 0.19. Infant mortality is an indicator rarely used
in the growth literature, which focuses more on life expectancy. Nagaraj,Varoudakis, and Véganzonès

Part I: Growth, Reform, and Governance 49
(2000) have estimated a short-term elasticity of 0.06 (here 0.02) and a long-term elasticity of 0.38
(here 0.12). Véganzonès] (2001) has obtained a short-term elasticity of 0.02 and a long-term one of
0.37. Our comparisons are constrained by the indicators and/or specifications chosen. Other close
interesting results can be found, for example,in Grace and others (2001), Bougheas, Demetriades, and
Mamuneas (2000), and Easterly and Levine (1997 and 2000).
19. The strong impact of the amelioration of the telephone network-which is, however, in line with the
finding of other studies-may also be explained by the fact that this indicator, along with the road net-
work, possibly captures the impact of other infrastructures.
20. As far as coefficients/elasticities are concerned, our results have also been compared to those of
other studies. In the case of inflation, Barro (1996)—who studied a panel of 110 countries for three
sub-periods between 1965 and 1990—found a short-term coefficient of 0.006 (here 0.005).
Guillaumont, Guillaumont-Jeanneney, and Varoudakis (1999) have estimated a short-term coefficient
of 0.006 and a long-term one of 0.14 (here 0.025) for a panel of 44 African countries studied on a five-
year average basis during 1960–95. In the case of public deficit, the same authors found a short-term
coefficient of 0.06 (here 0.08) and a long-term one of 0.14 (here 0.43). For the black market exchange
rate premium, Easterly and Levine (1993), as well as Easterly, Kremer, Prichet and Summer (1993),
obtained a short-term coefficient of 0.004 (here 0.005) for a panel of 115 countries studied during the
1960s, 1970s, and 1980s.
21. This indicator is considered here as a proxy of various distortions in the economy.
22. In the 1960s, most MENA governments implemented an import substitution strategy which iso-
lated the countries from outside competition. Although this was also the case in a majority of devel-
oping countries, in MENA this policy was reinforced in the 1970s and in 1980s.The abundance of for-
eign reserve from oil revenues, as well as from workers’ remittances and Arab assistance, encouraged
the governments to assume a still more assertive role in the economy. In the 1990s, this policy was pro-
gressively reverted in the non-oil-producing countries, as a result of the growing imbalances in the
government budget and in the balance of payments. The pace of reform was, however, slow in the oil-
producing countries. The political instability and the rigidities of the economies explain also why this
process has always remained less intensive in the region.
23. The same conclusions can be drawn for CFA Africa.
24. The relatively high potential growth of the 1990s (3.4 percent compared to 1.7 observed),contrary
to the 1980s, shows that some adverse factors to growth have not been taken into consideration in our
regressions. These factors could be political stability or quality of governance institutions, as developed
recently in the literature (see, for example, Knack and Keefer,1995, Acemoglu, Johnson and Robinson,
2001, and Rodrik, Subramanian and Trebbi, 2002. Some authors have in particular pointed out that
the MENA deficit in “good’’ institutions—(essentially political rights, civil liberties, quality of admin-
istration and corruption)—have significantly contributed to the slow economic activity of the region
(see El Badawi, 2002 and the World Bank, 2004).
References
Acemoglu, D., S. Johnson, and J. A. Robinson. 2001. “The Colonial Origins of Comparative
Development: An Empirical Investigation.” American Economic Review 91(5): 1369–401.
Aschauer,D. A. 1989. “Is Public Expenditure Productive?”Journal of Monetary Economics 23: 177–200.
Balassa, B. 1978. “Exports and Economic Growth—Further Evidence.” Journal of Development
Economics 5: 181–189.
Barro, R. J. 1990. “Government Spending in a Simple Model of Endogenous Growth. Journal of
Political Economy 98(5, part II): S103–S125.
———. 1996. “Determinants of Economic Growth: A Cross-Country Empirical Study.” Working
Paper Series, No 5698. National Bureau of Economic Research, Washington, DC (August).
Barro, R. J., and X. Sala-I-Martin. 1995. Economic Growth. New York: McGraw-Hill.

Breaking the Barriers to Higher Economic Growth50
Benhabib, J., and M. M. Spiegel. 1994. “The Role of Human Capital in Economic Development:
Evidence from Aggregate Cross-Country Data.”Journal of Monetary Economics 34(2): 143–173.
Bougheas, H., P. O. Demetriades, and T. P. Mamuneas. 2000. “Infrastructures, Specialization and
Economic Growth.” Canadian Journal of Economics 33(2): 506–522.
Canning, D. 1999. “Telecommunications, Infrastructures, Human Capital and Economic Growth.”
Discussion Paper No 55. Consulting Assistance on Economic Reform II, Cambridge, the Harvard
CAER II project consortium, September.
Dasguptta, D.,J. Keller, and T. G.Srinivasan. 2004.“Reforms and Elusive Growth in the Middle East—
What Happened in the 1990s?”In Trade Policy and Economic Integration in the Middle East & North
Africa: Economic Boundaries in Flux, ed. by H. Hakimian and J.B. Nugent. London: Curzon-
Routledge.
De Gregorio, J. 1992. “The Effects of Inflation on Economic Growth: Lessons from Latin America.”
European Economic Review 36(2/3): 417–425.
Easterly, W., and R. Levine. 1993. “Is Africa Different? Evidence from Growth Regressions.” Internal
document. World Bank, Washington, DC (April).
———. 1997.“Africa’s Growth Tragedy: Policies and Ethnic Divisions. Quarterly Journal of Economics
50(3): 112–120.
———. 2000. “It’s Not Factor Accumulation: Stylized Facts and Growth Models.” IMF Seminar Series
No. 2000–12: pp. 1–52. International Monetary Fund, Washington, DC (March).
Easterly, W., M. Kremer, L. Prichet, and L. Summer. 1993. “Good Policy or Good Luck: Country
Growth Performance and Temporary Shocks. Journal of Monetary Economics 32: 459–483.
El Badawi, I. A.2002. “Reviving Growth in the Arab World.” Working Paper Series, 0206. Arab Planing
Institute, Safat, Kuwait.
Feder, G. 1983. “On Exports and Economic Growth.” Journal of Development Economics 12(1/2):
59–79.
Fischer, S. 1993. “The Role of Macro-economic Factors in Growth.” Working Paper N. 4565. National
Bureau of Economic Research, Washington, DC.
Frankel, J.A., and D.Romer. 1999.“Does Trade Cause Growth.” The American Economic Review 89(3):
379–399.
Grace, J., C. Kenny, J. Liu, C. Qiang, and T. Reynolds. 2001. “Information and Communication
Technologies and Broad-Base Development: A Partial Review of Evidence.” World Bank,
Washington, DC (February).
Hakimian, H., and J.B. Nugent.2004. Trade Policy and Economic Integration in the Middle East & North
Africa: Economic Boundaries in Flux. London: Curzon-Routledge.
Hoekman, B., and H. Kheir-El-Din. 2000. Trade Policy Development in the Middle East and North
Africa. Washington D.C. and Cairo: World Bank and Economic Research Forum.
Iqbal, Z. 2001. Macroeconomic Issues and Policies in the Middle East and North Africa. Washington, DC:
International Monetary Fund.
Im, K. S., M. H. Pesaran, and Y. Shin. 1997. “Testing for Unit Roots in Heterogeneous Panels.” DAE
Working Paper Amalgamated Series, No 9526, University of Cambridge.
International Country Risk Guide (ICRG). 1999. “Brief Guide to the Rating System.” Political Link
Service Group, New York.
Knack, S., and P. Keefer. 1995. “Institutions and Economic Performance: Cross Country Tests Using
Alternative Institutional Measures.” Economics and Politics 7(3): 207–27.
Levine, R. 1997. “Financial Development and Economic Growth: Views and Agenda.” Journal of
Economic Literature 35: 688–726.
Levine, R., and S. Zervos. 1993. “Looking at the Facts. What We Know about Policy and Growth from
Cross-Country Analysis.” Policy Research Working Paper No. 1115. World Bank, Washington, DC.

Part I: Growth, Reform, and Governance 51
Lucas, R. E.1988. “On the Mechanism of Economic Development.” Journal of Monetary Economics 22:
3–42.
Mankiw,N.G., D.Romer,and D. N.Weil. 1992.“A Contribution to the Empirics of Economic Growth.”
Quarterly Journal of Economics 107(2): 407–427.
Murphy, K. M., A. Schleifer, and R. W. Vishny. 1989. “Industrialization and the Big Push.” Quarterly
Journal of Economics 106(2): 503–530.
Mussa, M. 1987. “Macroeconomic Policy and Trade Liberalization: Some Guidelines.” World Bank
Research Observer 57(6): 61–77.
Nabli, M. K. 2000. “From a Growth Crisis to Prospects for a Strong Growth Recovery in the MENA
Region.” Internal document, MNSED. World Bank, Washington DC.
Nabli M. K., and M-A Véganzonès-Varoudakis. 2002.“Exchange Rate Regime and Competitiveness of
Manufacture Exports: The Case of MENA Countries.” In Trade Policy and Economic Integration in
the Middle East & North Africa: Economic Boundaries in Flux, ed. by H.Hakimian and J. B.Nugent.
Curzon-Routledge: London.
Nagaraj, R., A. Varoudakis, and M. A.Véganzonès. 2000. “Long-Run Growth Trends and Convergence
across Indian States: The Role of Infrastructures.” Journal of International Development 12(1):
45–70.
Pinto, B. 1990. “Black Market Premium, Exchange Rate Unification and Inflation in Sub-Saharan
Africa.” The World Bank Economic Review 3(3): 321–338.
Pouliquen, L. 2000. “Infrastructure and Poverty.” Internal document. World Bank, Washington, DC.
Psacharopoulos, G. 1988. Economics of Education: Research and Studies. Elmsford,NY: Pergamon Press.
Rodrik, D., A. Subramanian, and F. Trebbi. 2002.“Institutions Rule: The Primacy of Institutions over
Geography and Integration in Economic Development. NBER Working Paper 9305, National
Bureau of Economic Research, Cambridge, MA.
Safadi, R. 1998. Opening Doors to the World: A New Trade Agenda for the Middle East. Cairo: Economic
Research Forum and American University Press.
Véganzonès-Varoudakis, M-A. 2001. “Infrastructures, Investissement et Croissance: Nouvelles
E´Vidences Empiriques.” Revue d’Economie du De´veloppement 4: 31–46.
Williamson, J. 1994. The Political Economy of Policy Reforms. Washington, DC: Institute of
International Economics.
World Bank. 2004. “Better Governance for Development in the Middle East and North Africa:
Enhancing Inclusiveness and Accountability.” MENA development report. The World Bank,
Washington, DC.

Breaking the Barriers to Higher Economic Growth52
Appendix 1. Countries in the Sample
Africa
MENA CFA Non-CFA Asia Latin America
Algeria (DZA) Burkina Faso (BFA) Botswana (BWA) Bangladesh (BGD) Argentina (ARG)
Egypt, Arab Rep. (EGY) Côte d’Ivoire (CIV) Ghana (GHA) Indonesia (IDN) Bolivia (BOL)
Iran, Islamic Rep.(IRN) Gabon (GAB) Kenya (KEN) India (IND) Brazil (BRA)
Jordan (JOR) Cameroon (CMR) Madagascar (MDG) Korea, Rep.(KOR) Chile (CHL)
Morocco (MAR) Gambia, The (GMB) Mauritius (MUS) Sri Lanka (LKA) Colombia (COL)
Syrian Arab Republic (SYR) Niger (NER) Malawi (MWI) Malaysia (MYS) Costa Rica (CRI)
Tunisia (TUN) Togo (TGO) Nigeria (NGA) Pakistan (PAK) Ecuador (ECU)
South Africa (ZAF) Philippines (PHL) Guatemala (GTM)
Zambia (ZMB) Thailand ( THA) Mexico (MEX)
Peru (PER)
Paraguay (PRY)
Uruguay (URY)

Part I: Growth, Reform, and Governance 53
Appendix 2. Principal Component Analysis*
The principal components of the basic indicators are extracted for each group
of indicators from an annual panel of 44 countries over 1970–80 to 1999. The
five composite indicators are constructed as the weighted sum of one or two
principal components, depending on the explanatory power of each compo-
nent. We chose the most significant principal components whose eigenvalues
are higher than one. In this case, we explain around 70 percent of the variance
of the underlying individual indicators (see tables A2.1 to A2.5). The weight
attributed to each principal component corresponds to its relative contribu-
tion to the variance of the initial indicators (calculated from the cumulative
R2).1The contribution of each individual indicator to the composite indicator
can then be computed as a linear combination of the weights associated with
the one or two principal components and of the loadings of the individual
indicators on each principal component (see below). The calculations show
that all initial indicators contribute as expected to the composite indicators.2
*Source: Authors’ calculations.
Table A2.1. Macroeconomic Stability Variables
Component Eigenvalue Cumulative R2
P1 1.19 0.40
P2 0.96 0.72
P3 0.85 1.00
Loadings P1 P2 P3
P0.48 0.86 0.19
lBmp 0.72 ⫺0.16 ⫺0.68
PubDef ⫺0.67 0.45 ⫺0.60
MS=0.40/0.72*P1+(0.72–0.40)/0.72*P2
Table A2.2. External Stability Variables
Component Eigenvalue Cumulative R2
P1 1.92 0.65
P2 0.81 0.92
P3 0.25 1.00
Loadings P1 P2 P3
DebExX 0.92 0.03 0.37
DebExGni 0.73 0.60 ⫺0.26
CurAccX ⫺0.71 0.67 0.21
ES=P1

Breaking the Barriers to Higher Economic Growth54
Table A2.3. Structural Reform Variables
Component Eigenvalue Cumulative R2
P1 1.49 0.75
P2 0.51 1.00
Loadings P1 P2
TradeP1 0.86 0.50
PCRBOG 0.86 ⫺0.50
SR=P1
Table A2.4.A. Human Capital Variables (complete)
Component Eigenvalue Cumulative R2
P1 2.95 0.74
P2 0.50 0.87
P3 0.34 0.95
P4 0.19 1.00
Loadings P1 P2 P3 P4
ln(Mort) 0.79 0.590 0.15 0.08
ln(H1) ⫺0.84 0.340 ⫺0.43 0.04
ln(H2) ⫺0.91 ⫺0.005 0.26 0.32
ln(H3) ⫺0.89 0.210 0.27 0.29
H=P1
Table A2.4.B. Human Capital Variables (reduced)
Component Eigenvalue Cumulative R2
P1 1.52 0.76
P2 0.48 1.00
Loadings P1 P2
ln(Mort) 0.87 0.49
ln(H1) ⫺0.87 0.49
MH1=P1
Table A2.5. Physical Infrastructure Variables
Component Eigenvalue Cumulative R2
P1 1.42 0.71
P2 0.58 1.00
Loadings P1 P2
ln(Roads) 0.84 0.54
ln(Tel) 0.84 ⫺0.54
Ph7ys=P1

Part I: Growth, Reform, and Governance 55
Appendix 3. Macroeconomic Stability indicators
Most MENA countries adopted better macroeconomic policies—some in the
late 1980s (Morocco, Tunisia, and Jordan) and others in the 1990s— after a
decade of regression (Iran, Syria, Algeria, Egypt). In the 1990s, some MENA
countries (Morocco, Tunisia, Jordan, and Egypt) undertook a level of macro-
economic reforms similar to the average in the South and East Asian
economies.
In the 1990s, inflation was successfully contained (particularly in Morocco,
Jordan, and Tunisia; figure 2.8b). Public deficit was reduced everywhere in the
region (figure 2.9b) and the foreign exchange parallel market premium ended
in the majority of countries in our sample (figure 2.10b).
Although MENA macroeconomic policy has globally improved, further
progress in Tunisia and Morocco would have reduced the public deficit, which
has been higher than the MENA average (3.4 percent and 2.8 percent of GDP,
on average, in the 1990s; see figure 2.9b). In Iran and Algeria, the black market
exchange rate (owing to capital controls and political instability) and inflation
(which reached 20 percent in the 1990s) should have been better controlled
(see figures 2.8b and 2.10b). This is also the case, to a lesser extent, for the for-
eign exchange parallel market premium in Syria and for inflation in Egypt.
Figure 2.8. Inflation (%)
⫺10
40
90
140
190
240
0
5
10
15
20
25
a. Region b. Country
MENA Latin
America
Asia Africa
(CFA)
Africa
(non-CFA)
Iran Syria Jordan Djibouti Tunisia Morocco Egypt
1970–79
1980–89
1990–99
1970–79
1980–89
1990–99
Source: Authors’calculations.

Breaking the Barriers to Higher Economic Growth56
Figure 2.9. Public Deficit (% of GDP)
⫺17
⫺15
⫺13
⫺11
⫺9
⫺7
⫺5
⫺3
⫺1
⫺17
⫺15
⫺13
⫺11
⫺9
⫺7
⫺5
⫺3
⫺1
a. Region b. Country
MENA Latin
America
Asia Africa
(CFA)
Africa
(non-CFA)
Iran Syria Jordan Djibouti Tunisia Morocco Egypt
1970–79
1980–89
1990–99
1970–79
1980–89
1990–99
Source: Authors’calculations.
Source: Authors’calculations.
Figure 2.10. Black Market Premium (%)
0
50
100
150
200
250
300
350
400
0
50
100
150
200
250
300
350
400
a. Region b. Country
MENA Latin
America
Asia Africa
(CFA)
Africa
(non-CFA)
Iran Syria Jordan Djibouti Tunisia Morocco Egypt
1970–79
1980–89
1990–99
1970–79
1980–89
1990–99

Part I: Growth, Reform, and Governance 57
Appendix 4. External Stability indicators
In MENA, the level of foreign debt increased dramatically in the 1980s (see
figure 2.11a and 2.12a, because of the high public investment ratio of the
1970s and 1980s (see appendix 8). The exceptions were Iran, and to a lesser
extent, Tunisia and Algeria.3
In the 1990s, the MENA countries’ external debt accounted for 60 to 70
percent of gross national investment (GNI) in Morocco, Egypt, Algeria,
Tunisia (around 150 to 200 percent of exports), and more than 130 percent of
GNI in Syria and Jordan (see figures 2.11b and 2.12b).4In the 1990s, Morocco
and Egypt were the only two countries to reduce their external debt, follow-
ing major debt forgiveness after the Gulf War.
These difficulties in containing the external debt have been partly compen-
sated for by improvements in the current account balance. In fact, MENA
countries are among the best performers from all regions (see figure 2.13a).
Even though only a few MENA economies exhibited a positive balance in the
1990s—this was the case for Iran, Syria, and Egypt—efforts to reduce deficits
are noticeable in almost every MENA country (except Jordan, see figure
2.13b).
Source: Authors’calculations.
Figure 2.11. External Debt (% of exports)
0
50
100
150
200
250
300
350
400
0
50
100
150
200
250
300
350
400
a. Region b. Country
MENA Latin
America
Asia Africa
(CFA)
Africa
(non-CFA)
Iran Syria Jordan Djibouti Tunisia Morocco Egypt
1970–79
1980–89
1990–99
1970–79
1980–89
1990–99

Breaking the Barriers to Higher Economic Growth58
Figure 2.12. External Debt (% of GNI)
⫺10
10
30
50
70
90
110
130
150
⫺10
10
30
50
70
90
110
130
150
a. Region b. Country
MENA Latin
America
Asia Africa
(CFA)
Africa
(non-CFA)
Iran Syria Jordan Djibouti Tunisia Morocco Egypt
1970–79
1980–89
1990–99
1970–79
1980–89
1990–99
Source: Authors’calculations.
Source: Authors’calculations.
Figure 2.13. Current Account Balance (% of exports)
⫺55
⫺45
⫺35
⫺25
⫺15
⫺5
5
15
25
⫺55
⫺45
⫺35
⫺25
⫺15
⫺5
5
15
25
a. Region b. Country
MENA Latin
America
Asia
Africa
(CFA)
Africa
(non-CFA)
Iran Syria Jordan Djibouti Tunisia Morocco Egypt
1970–79
1980–89
1990–99
1970–79
1980–89
1990–99

Part I: Growth, Reform, and Governance 59
Appendix 5. Structural Reform indicators
The ratio of private credit as a share of total credit from banking systems and
other institutions has averaged 35 percent during the 1980s and the 1990s.
Financial depth has been particularly significant in Jordan and Tunisia
(respectively 70 percent and 60 percent of GDP), but has improved almost
everywhere—except in Egypt, Iran and Algeria in the 1990s (figure 2.14b).
This achievement has been better only in Asia, where the financial ratio
reached, on average, 60 percent in the 1990s (see figure 2.14a). Financial
development has, however, been weaker in Syria, Algeria, and, to a lesser
extent, Iran and Egypt in the 1990s (see figure 2.14b).
Our findings do not mean, however, that the MENA economies have ben-
efited from a strong banking system or a developed Financial sector. In fact,
other studies highlight the deficiencies of the financial sector as an effective
means of boosting the development of the private sector and the growth
prospects of the region (Nabli, 2000). In this context, our results might be due
to the fact that the proxy used (PCrBOG) is unable to capture either the qual-
ity of the banking system (which could be better analyzed through other spe-
cific indicators),5or the development of the financial markets (which have
also been deficient in the MENA economies).
Trade openness has often been particularly low in MENA—with a ratio, on
average, of about 30 percent of GDP, compared to 45 to 70 percent in Asia6—
and rather similar to that of Latin America (figure 2.15a). This is the case
despite some exports diversification of the non-oil-producing countries in the
1990s, which explains why trade openness has been rather high in Jordan and
Tunisia (around 60 percent of GDP; figure 2.15b). These countries (with
Morocco) have been the most diversified in the region (table A5-1). Trade
Source: Authors’calculations.
Figure 2.14. Credit from the Banking System Indicator
70
60
50
40
30
20
10
0
a. Region b. Country
MENA Latin
America
Asia
Africa
(CFA)
Africa
(non-CFA)
Iran Syria Jordan Djibouti Tunisia Morocco Egypt
1970–79
1980–89
1990–99
1970–79
1980–89
1990–99
70
60
50
40
30
20
10
0

Breaking the Barriers to Higher Economic Growth60
openness has remained weak in Iran, Syria, and Algeria (between 10 percent
and 20 percent of GDP in the 1990s) due to the difficulties in moving from oil
production and state-dominated management of the economies.7In these
countries, the scope for improvement of trade policy is still very significant.
Figure 2.15. Trade Policy (excluding exports of oil and mining, % of GDP)
0
10
20
30
40
50
60
70
0
10
20
30
40
50
60
70
a. Region b. Country
MENA Latin
America
Asia
Africa
(CFA)
Africa
(non-CFA)
Iran Syria Jordan Djibouti Tunisia Morocco Egypt
1970–79
1980–89
1990–99
1970–79
1980–89
1990–99
Source: Authors’calculations.
Table A5-1. Average Manufactured Exports of Selected MENA Countries
Algeria Egypt Iran Jordan Morocco Tunisia
%X % GDP % X % GDP %X % GDP %X % GDP %X % GDP %X % GDP
1970–79 3 0.6 27 3.1 2.9 0.6 26 1.9 16 2.1 24 4.6
1980–89 1.5 0.3 19 1.5 4.0 0.3 43 5.4 39 6.0 49 12
1990–99 3.3 0.8 37 2.4 6.6 1.5 49 9.5 53 7.5 75 21
Source: Authors’ calculations.
*For the first subperiod, four values were missing for Iran (1970, 1971, 1972, and 1973).
**As far as the thirrd subperiod is concerned, two values were missing for Iran (1991 and 1992) and one for Jordan (1996).

Part I: Growth, Reform, and Governance 61
Appendix 6. Human Capital indicators
As far as human capital is concerned, one important area of success is the
reduction of infant mortality. Mortality rates (which reached 30 per 1,000 in
the 1990s) were cut by two-thirds compared to the 1970s. Levels are now in
line with Asia and Latin America. In Egypt, the ratio of infant mortality fell
sixfold during the same period. In this country, however, because of a high
initial level, the ratio still surpasses the MENA average. Noticeable efforts were
also made by Iran, Algeria, Tunisia, and Morocco (see figure 2.16).
In the field of education, results are more mitigated. Despite visible
progress, the level of primary schooling has remained lower than in Asia,Latin
America, and non-CFA Africa. Tunisia is the exception, with almost five years
of primary schooling in the 1990s. Successful achievements were also realized
in Jordan, Algeria, Syria, and Iran (about four years of schooling; see figure
2.17). On average, achievements were greater in secondary and superior edu-
cation, where schooling was more in line with Asia and Latin America. The
performance in Syria, Iran, Jordan, and Egypt has been above the MENA aver-
age, with 1.5 to 2 years, and 0.6 years of schooling, respectively (but only 0.15
for superior education in the case of Iran, see figures 2.18b and 2.19b).
Source: Authors’calculations.
Figure 2.16. Mortality Rate (per 1,000 population)
⫺10
10
30
50
70
90
110
130
0
50
100
150
200
250
a. Region b. Country
MENA Latin
America
Asia
Africa
(CFA)
Africa
(non-CFA)
Iran Syria Jordan Djibouti Tunisia Morocco Egypt
1970–79
1980–89
1990–99
1970–79
1980–89
1990–99

Breaking the Barriers to Higher Economic Growth62
Figure 2.17. Primary Education (number of years of schooling of the population)
0
1
2
3
4
5
6
0
1
2
3
4
5
6
a. Region b. Country
MENA Latin
America
Asia
Africa
(CFA)
Africa
(non-CFA)
Iran Syria Jordan Djibouti Tunisia Morocco Egypt
1970–79
1980–89
1990–99
1970–79
1980–89
1990–99
Source: Authors’calculations.
Source: Authors’calculations.
Source: Authors’calculations.
Figure 2.18. Secondary Education (number of years of schooling of the population)
Figure 2.19. Tertiary Education (number of years of schooling of the population)
⫺1.5
⫺1.0
⫺0.5
0
0.5
1.0
1.5
2.0
⫺1.5
⫺1.0
⫺0.5
0
0.5
1.0
1.5
2.0
a. Region b. Country
MENA Latin
America
Asia
Africa
(CFA)
Africa
(non-CFA)
Iran Syria Jordan Djibouti Tunisia Morocco Egypt
1970–79
1980–89
1990–99
1970–79
1980–89
1990–99
⫺0.05
0.05
0.15
0.25
0.35
0.45
0.55
0.65
⫺0.05
0.05
0.15
0.25
0.35
0.45
0.55
0.65
a. Region b. Country
MENA Latin
America
Asia
Africa
(CFA)
Africa
(non-CFA)
Iran Syria Jordan Djibouti Tunisia Morocco Egypt
1970–79
1980–89
1990–99
1970–79
1980–89
1990–99

Part I: Growth, Reform, and Governance 63
Appendix 7. Physical Infrastructure Indicators
Despite real progress throughout the period, MENA physical infrastructures
remain on average insufficient. Iran, Syria, Tunisia, and to a lesser extent
Morocco, however, show better achievements than other MENA economies
(almost in line with Latin America and Asia; see figure 2.5).8
The road network, in particular, has been markedly insufficient, with a
density as low as in CFA Africa. Although a majority of countries have pro-
gressively improved their road infrastructure, the level was still low in Jordan,
Algeria, and Egypt in the 1990s. Better progress was made in Syria, Tunisia,
Morocco, and Iran, where construction equipment was in some ways equiva-
lent to that in Latin America (but worse than in Asia; see figure 2.20).
The same observation can be made for the telephone network. Despite real
improvements in almost all MENA countries, the level of equipment has
remained deficient compared to that in Latin America and Asia (but far bet-
ter than in Africa). Only Iran, Syria, and Jordan have revealed a pattern simi-
lar to that in Latin America (see figure 2.21).
Figure 2.20. Road Network (km per km2)
Source: Authors’calculations.
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
a. Region b. Country
MENA Latin
America
Asia
Africa
(CFA)
Africa
(non-CFA)
Iran Syria Jordan Djibouti Tunisia Morocco Egypt
1970–79
1980–89
1990–99
1970–79
1980–89
1990–99

Breaking the Barriers to Higher Economic Growth64
Source: Authors’calculations.
Figure 2.21. Telephone Network (number of lines per 1,000 people)
0
20
40
60
80
100
120
0
20
40
60
80
100
120
a. Region b. Country
MENA Latin
America
Asia
Africa
(CFA)
Africa
(non-CFA)
Iran Syria Jordan Djibouti Tunisia Morocco Egypt
1970–79
1980–89
1990–99
1970–79
1980–89
1990–99

Part I: Growth, Reform, and Governance 65
Appendix 8. investment
Total investment in MENA has always remained rather dynamic compared to
the rest of the world. At around 25 percent of GDP, total investment has been
stronger than in Latin America and Africa, but inferior from the 1980s to Asia
(figure 2.22a). This result is mainly the result of the high public investment
ratio, private investment having conversely always been rather weak
(Dasgupta, Keller, and Srinivasan, 2002).
In the 1970s, investment was particularly high in Egypt (more than 60 per-
cent of GDP) and to a lesser extent, in Algeria and Jordan (35–40 percent of
GDP). This situation can be explained by an abundance of liquidities result-
ing from the increase in oil revenues, workers’ remittances, and the profusion
of foreign capital. During this time, many countries were able to improve their
physical infrastructure and human capital.
In the 1980s, because of the marked economic slowdown, many MENA
economies faced overinvestment, in a context of macroeconomic imbalances
and a growing debt burden.9Total investment was still high, around 30 per-
cent of GDP in Algeria, Jordan, and Egypt (figure 2.22b), and public invest-
ment still more dynamic than private investment.
During the 1990s, many MENA countries were able to adjust their public
investment ratio with some success, and total investment ranges from 20 to 30
percent of GDP. This has been done in a context of macroeconomic stabiliza-
tion and structural reforms. Private investment remains, however, low com-
pared to the need for productivity gains in the MENA economies.
Source: Authors’calculations.
Figure 2.22. Investment (% of GDP)
0
5
10
15
20
25
30
0
10
20
30
40
50
60
70
a.
R
eg
i
on
b
.
C
ountry
MENA Latin
America
Asia
Africa
(CFA)
Africa
(non-CFA)
Iran Syria Jordan Djibouti Tunisia Morocco Egypt
1970–79
1980–89
1990–99
1970–79
1980–89
1990–99

Breaking the Barriers to Higher Economic Growth66
Appendix 9. Augmented Dickey-Fuller Test (ADF)
Table A9. Equations (1) and (2)
Variable ADF statistic k(1) Trend(2) Critical value(3) ADF test
Ln(y) ⫺3.22 1 Yes ⫺1.82* I(0)
Ln(Inv) ⫺3.12 1 No ⫺1.82* I(0)
MS ⫺3.19 1 No ⫺1.82* I(0)
ES ⫺3.62 1 No ⫺1.82* I(0)
SR ⫺1.86 1 No ⫺1.82* I(0)
MH1⫺1.76 1 No ⫺1.73** I(0)
Phys ⫺6.58 1 No ⫺1.82* I(0)
Source: Authors’calculations.
(1) kis the number of lags in the ADF test.
(2) The introduction of a trend when estimating Equations (1) and (2) shows, however, that this trend is not
significant in the growth equation.
(3) Im, Pesaran, and Shin (1997) critical values (respectively *10% and ** 5% level).
Data have been compiled from World Development Indicators, Global Development Finance, Global Development
Network, and Live Data Base (World Bank).

Part I: Growth, Reform, and Governance 67
Appendix 10. Methodology of Calculation of the Coefficient/Elasticities of the
Disaggregated Indicators
The impact of each disaggregated indicator can be computed as follows: Let P
be the vector (n*1) of the nprincipal components selected and δthe vector
(1*n) of their weights in the aggregate indicator. Furthermore, the nprincipal
components are expressed as a linear combination of initial variables such
that P= AX,Xbeing the vector (k*1) of kvariables, and Arepresents the
matrix (n*k) of loadings assigned to them. The composite indicator is
expressed as: δP= δ AX. Denoting by γthe estimated coefficient for this indi-
cator, the convergence equation can be written:
ln(yii,t) – ln(yi,t–1) = αi –βln(yii,t–1) + γδAXi,t + ηi,t + ui,t (3)
The vector (1*k) (E), expressing the impact on growth of the original vari-
ables, can be calculated such that E= γδA. These coefficients are estimated
from Equations (1) and (2), as well as from the loadings summarized in
Appendix 2. However, given the standardization procedure for the variables
associated with the principal components method, the contribution of varia-
tions in level of each indicator to growth is expressed by the previously calcu-
lated coefficient (ei), divided by the standard deviation for each variable
(ei/σi). The coefficients/elasticities of the long-run GDP per capita level with
respect to different types of indicator is then obtained by dividing the impact
coefficients by the convergence coefficient (β). Table 2.2 gives the long-term
coefficients/elasticities for each indicator (see also table A11-1 and A11-2 in
appendix 11 for intermediate calculations).

Breaking the Barriers to Higher Economic Growth68
Appendix 11. Short- and Long-Term Coefficients/Elasticities of the
Disaggregated Reform Indicators
Table A11-1. Equation (1)
Short-term coefficients/elasticities Long-term
Standardized Level coefficients/
Index Variables variables variables elasticities
MS P ⫺0.0026 ⫺0.0007 ⫺0.004
ln(Bmp) ⫺0.0013 ⫺0.0007 ⫺0.004
PubDef 0.0006 0.0094 0.056
ES DebExX ⫺0.027 ⫺0.0172 ⫺0.101
DebExGni ⫺0.022 ⫺0.0005 ⫺0.003
CurAccX 0.021 0.0770 0.453
SR TradeP 0.013 0.0565 0.332
PCRBOG 0.013 0.0618 0.363
Phys ln(Roads) 0.025 0.0253 0.149
ln(Tel) 0.025 0.0181 0.106
MH1 ln(Mort) ⫺0.0157 ⫺0.0249 ⫺0.147
ln(H1) ⫺0.0157 0.0287 0.169
Source: Authors’ calculations.
Table A11-2. Equation (2)
Short-term coefficients/elasticities Long-term
Standardized Level coefficients/
Index Variables variables variables elasticities
MS P 0.0176 0.0044 0.026
ES ln(Bmp) 0.0087 0.0050 0.029
SR PubDef 0.0039 0.0637 0.375
Phys DebExX 0.024 0.0154 0.091
MH1 DebExGni 0.020 0.0005 0.003
CurAccX 0.018 0.0687 0.404
TradeP 0.009 0.0414 0.244
PCRBOG 0.009 0.0453 0.266
ln(Roads) 0.026 0.0262 0.154
ln(Tel) 0.026 0.0187 0.110
ln(Mort) 0.0139 0.0222 0.130
ln(H1) 0.0139 0.0255 0.150
Source: Authors’ calculations.

Part I: Growth, Reform, and Governance 69
Appendix 12. Contribution to Growth: MENA Countries’ Experience
A12-1. Impact of Macroeconomic Stability
Years (%) MENA Tunisia Morocco Algeria Egypt Jordan Iran Syria
1980–89 Contribution ⫺1.3 0.2 ⫺0.01 ⫺1.0 0.4 0.1 ⫺2.6 ⫺2.4
1990–99 to growth 0.4 0.4 0.70 0.8 1.3 0.6 0.3 0.6
1980–89 GDP per capita 0.0 1.6 1.40 0.1 2.2 ⫺0.7 ⫺3.5 ⫺1.1
1990–99 growth rate 1.7 3.2 0.50 ⫺0.3 1.9 0.4 2.0 4.4
Source: Author’ calculations.
A12-1.a. Inflation
Years MENA Tunisia Morocco Algeria Egypt Jordan Iran Syria
1970s Inflation 9 6 7 8 16 11 10 8
1980s ( %) 13 9 8 9 17 7 20 23
1990s Average 11 5 4 20 13 5 24 8
1980–89 Annual 0.4 0.3 0.02 0.1 0.2 ⫺0.4 0.9 1.4
1990–99 change ⫺0.2 ⫺0.4 ⫺0.3 1.1 ⫺0.4 ⫺0.2 0.4 ⫺1.4
Total: p + p*SR
1980–89 Contribution to ⫺0.02 ⫺0.01 ⫺0.001 ⫺0.004 ⫺0.01 0.02 ⫺0.05 ⫺0.07
1990–99 growth 0.01 0.02 0.020 ⫺0.060 0.02 0.01 ⫺0.02 0.07
Source: Authors’ calculations.
A12-1.b. Black Market Exchange Rate Premium
Years MENA Tunisia Morocco Algeria Egypt Jordan Iran Syria
1970s bmp 23.0 19.0 6.0 79.0 30.0 3.5 20.0 7.0
1980s ( %) 207.0 9.0 6.0 330.0 28.0 4.3 793.0 282.0
1990s Average 268.0 5.0 4.0 191.0 13.0 4.5 1,349.0 307.0
1980–89 Annual 22.0 ⫺7.3 0.5 14.0 ⫺0.7 2.0 37.0 37.0
1990–99 growth rate 2.6 ⫺4.9 ⫺4.0 ⫺5.5 ⫺7.4 0.4 5.3 0.8
Total: bmp + bmp*SR
1980–89 Contribution to ⫺1.2 0.4 ⫺0.03 ⫺0.8 0.04 ⫺0.10 ⫺2.1 ⫺2.10
1990–99 growth ⫺0.1 0.3 0.20 0.3 0.40 ⫺0.02 ⫺0.3 ⫺0.05
Source: Authors’ calculations.

Breaking the Barriers to Higher Economic Growth70
A12-1.c. Public Deficit
Years MENA Tunisia Morocco Algeria Egypt Jordan Iran Syria
1970s Public deficit ⫺7.1 ⫺2.9 ⫺7.8 ⫺5.1 ⫺16.5 ⫺9.9 ⫺2.2 ⫺5.4
1980s ( % GDP) ⫺7.8 ⫺4.9 ⫺7.6 ⫺7.3 ⫺12.5 ⫺7.8 ⫺7.3 ⫺7.4
1990s Average ⫺1.6 ⫺3.4 ⫺2.8 ⫺0.9 ⫺2.5 ⫺0.9 0.6 ⫺0.3
1980–89 Annual 0.03 0.2 ⫺0.02 0.2 ⫺0.4 ⫺0.2 0.5 0.2
1990–99 change ⫺0.6 ⫺0.2 ⫺0.5 ⫺0.6 ⫺1.0 ⫺0.7 ⫺0.7 ⫺0.7
Total PubDef + PubDef*SR
1980–89 Contribution to ⫺0.03 ⫺0.2 0.02 ⫺0.2 0.4 0.2 ⫺0.4 ⫺0.2
1990–99 growth 0.5 0.1 0.4 0.6 0.9 0.6 0.6 0.6
Source: Authors’calculations.
A12-2. Impact of External Stability
Years % MENA Tunisia Morocco Algeria Egypt Jordan Iran Syria
1980–89 Contribution to ⫺1.0 0.1 ⫺0.8 0.1 ⫺0.9 ⫺0.4 ⫺2.4
1990–99 growth 1.0 0.3 1.2 0.1 2.5 ⫺1.1 0.6 0.8
1980–89 GDP per capita 0.0 1.6 1.4 0.1 2.2 ⫺0.7 ⫺3.5 ⫺1.1
1990–99 growth rate 1.7 3.2 0.5 ⫺0.3 1.9 0.4 2.0 4.4
Source: Authors’ calculations.
A12-2.a. External Debt
Years MENA Tunisia Morocco Algeria Egypt Jordan Iran Syria
1970s External debt 141 125 190 185 165 82 100
1980s ( % of exports) 219 139 300 178 350 136 42 389
1990s Average 201 129 230 231 220 220 84 365
1980–89 Annual 9.9 1.4 11 ⫺0.7 19 5.4 29.0
1990–99 change ⫺1.8 ⫺1.0 ⫺7 5.3 ⫺13 8.4 4.2 ⫺2.4
LT Elasticity ⫺0.089
1980–89 Contribution to ⫺0.9 ⫺0.1 ⫺1.0 0.1 ⫺1.6 ⫺0.5 ⫺2.6
1990–99 growth 0.2 0.1 0.6 ⫺0.5 1.2 ⫺0.7 ⫺0.4 0.2
Source: Authors’ calculations.
A12-2.b. External Debt
Years MENA Tunisia Morocco Algeria Egypt Jordan Iran Syria
1970s External debt 28 33 33 37 48 30 17
1980s ( % of GNI) 66 56 92 39 107 81 4 83
1990s Average 80 59 72 63 61 147 21 135
1980–89 Annual 3.8 2.3 5.9 0.1 5.9 5.0 6.6
1990–99 change 1.4 0.4 ⫺2.0 2.4 ⫺4.6 6.6 1.6 5.2
LT Elasticity ⫺0.003
1980–89 Contribution to ⫺0.010 ⫺0.01 ⫺0.02 ⫺0.0004 ⫺0.02 ⫺0.02 ⫺0.02
1990–99 growth ⫺0.004 0.00 0.01 ⫺0.0100 0.01 ⫺0.02 0.00 ⫺0.02
Source: Authors’ calculations.

Part I: Growth, Reform, and Governance 71
A12-2.c. Current Account Deficit
Years MENA Tunisia Morocco Algeria Egypt Jordan Iran Syria
1970s Current account ⫺17 ⫺26 ⫺35 ⫺24 ⫺50 ⫺11 26 ⫺19
1980s ( % Exports) ⫺20 ⫺20 ⫺31 ⫺14 ⫺30 ⫺8⫺20 ⫺14
1990s Average 1 ⫺15 ⫺16 0 4 ⫺16 4 4
1980–89 Annual ⫺0.3 0.6 0.4 1 2.0 0.3 ⫺4.6 0.5
1990–99 change 2.1 0.5 1.5 1.4 3.4 ⫺0.8 2.4 1.8
LT Elasticity 0.396
1980–89 Contribution to ⫺0.1 0.2 0.2 0.4 0.8 0.1 ⫺1.8 0.2
1990–99 growth 0.8 0.2 0.6 0.6 1.3 ⫺0.3 1.0 0.7
Source: Authors’calculations.
A12-3. Impact of Structural Reforms
Years (%) MENA Tunisia Morocco Algeria Egypt Jordan Iran Syria
1980–89 Contribution to 0.2 0.6 0.1 ⫺0.1 0.40 0.5 ⫺0.05 ⫺0.05
1990–99 growth 0.2 0.7 0.5 ⫺0.8 ⫺0.03 0.6 ⫺0.10 0.40
1980–89 GDP per capita 0.0 1.6 1.4 0.1 2.20 ⫺0.7 ⫺3.50 ⫺1.10
1990–99 growth rate 1.7 3.2 0.5 ⫺0.3 1.90 0.4 2.00 4.40
Source: Authors’ calculations.
A12-3.a. Trade Policy
Years MENA Tunisia Morocco Algeria Egypt Jordan Iran Syria
1970s Trade openness 27 29 29 26 36 47 19 6
1980s ( % GDP) 26 47 31 11 41 43 7 3
1990s Average 33 62 37 15 37 57 10 15
1980–89 Annual ⫺0.1 1.8 0.2 ⫺1.4 0.4 ⫺0.4 ⫺1.2 ⫺0.3
1990–99 change 0.7 1.5 0.6 0.3 ⫺0.4 1.4 0.3 1.3
LT Elasticity 0.24
1980–89 Contribution to ⫺0.03 0.4 0.1 ⫺0.3 0.1 ⫺0.1 ⫺0.3 ⫺0.1
1990–99 growth 0.20 0.4 0.1 0.1 ⫺0.1 0.3 0.1 0.3
Source: Authors’calculations.
A12-3.b. Financial Development
Years MENA Tunisia Morocco Algeria Egypt Jordan Iran Syria
1970s Private credit 27 41 20 43 18 39 24 6
1980s ( % GDP) 35 47 21 52 28 59 33 7
1990s Average 35 59 34 19 31 69 26 9
1980–89 Annual 0.80 0.6 0.1 1.0 1.1 2.0 0.9 0.1
1990–99 change 0.01 1.2 1.3 ⫺3.3 0.3 1.0 ⫺0.6 0.3
LT Elasticity 0.27
1980–89 Contribution to 0.2 0.2 0.03 0.3 0.3 0.5 0.2 0.02
1990–99 growth 0.0 0.3 0.40 ⫺0.9 0.1 0.3 ⫺0.2 0.10
Source: Authors’calculations.

Breaking the Barriers to Higher Economic Growth72
A12-4. Impact of Human Capital
Years % MENA Tunisia Morocco Algeria Egypt Jordan Iran Syria
1980–89 Contribution to 1.1 1.2 1.0 1.2 0.7 0.6 1.3 1.3
1990–99 growth 0.9 0.9 0.8 1.2 0.8 0.4 1.2 0.9
1980–89 GDP per capita 0.0 1.6 1.4 0.1 2.2 ⫺0.7 ⫺3.5 ⫺1.1
1990–99 growth rate 1.7 3.2 0.5 ⫺0.3 1.9 0.4 2.0 4.4
Source: Authors’calculations.
A12-4.a. Primary Education
Years MENA Tunisia Morocco Algeria Egypt Jordan Iran Syria
1970s Primary Education 2.4 2.7 1.2 2.1 4.9 2.6 1.6 1.7
1980s (number of years) 3.0 4.0 1.8 3.1 3.1 3.5 2.6 2.7
1990s Average 3.7 4.8 2.3 3.9 3.6 4.0 3.9 3.7
1980–89 Annual 2.8 3.9 3.9 3.7 ⫺4.4 3.2 4.9 4.6
1990–99 growth rate 2.3 1.8 2.4 2.5 1.3 1.4 4.1 3.4
LT Elasticity 0.14
1980–89 Contribution to 0.4 0.5 0.5 0.5 ⫺0.6 0.4 0.7 0.6
1990–99 growth 0.3 0.2 0.3 0.3 0.2 0.2 0.6 0.5
Source: Authors’calculations.
A12-4.b. Infant Mortality
Years MENA Tunisia Morocco Algeria Egypt Jordan Iran Syria
1970s Infant mortality 122 99 117 123 277 43 118 79
1980s (/1000 population) 65 56 81 71 95 37 70 47
1990s Average 41 33 56 36 57 32 43 34
1980–89 Annual ⫺6.3 ⫺5.8 ⫺3.7 ⫺5.6 ⫺10.7 ⫺1.5 ⫺5.2 ⫺5.3
1990–99 growth rate ⫺4.5 ⫺5.1 ⫺3.8 ⫺6.8 ⫺5.1 ⫺1.6 ⫺4.9 ⫺3.2
LT Elasticity ⫺0.12
1980–89 Contribution to 0.8 0.7 0.4 0.7 1.3 0.2 0.6 0.6
1990–99 growth 0.5 0.6 0.5 0.8 0.6 0.2 0.6 0.4
Source: Authors’ calculations.
A12-5. Impact of Physical Infrastructures
Years % MENA Tunisia Morocco Algeria Egypt Jordan Iran Syria
1980–89 Contribution to 1.4 1.3 1.0 1.0 1.0 1.6 1.8 1.5
1990–99 growth 1.0 1.0 1.3 0.5 ⫺0.9 0.5 1.7 1.4
1980–89 GDP per capita 0.0 1.6 1.4 0.1 2.2 ⫺0.7 ⫺3.5 ⫺1.1
1990–99 growth rate 1.7 3.2 0.5 ⫺0.3 1.9 0.4 2.0 4.4
Source: Authors’calculations.

Part I: Growth, Reform, and Governance 73
A12-5.a. Telephone Lines
Years MENA Tunisia Morocco Algeria Egypt Jordan Iran Syria
1970s Telephone lines 14 11 7 9 9 17 17 16
1980s (/1000 population) 28 26 11 24 17 52 28 39
1990s Average 59 58 40 41 37 74 75 78
1980–89 Annual 7.2 8.1 4.7 10.0 6.8 11.2 4.9 8.5
1990–99 growth rate 7.5 8.1 13.1 5.5 8.2 3.6 9.8 7.0
LT Elasticity 0.1
1980–89 Contribution to 0.8 0.8 0.5 1.0 0.7 1.1 0.5 0.9
1990–99 growth 0.8 0.8 1.3 0.5 0.8 0.4 1.0 0.7
Source: Authors’calculations.
A12-5.b. Road Network
Years MENA Tunisia Morocco Algeria Egypt Jordan Iran Syria
1970s Road network 0.068 0.121 0.094 0.032 0.025 0.043 0.035 0.092
1980s (km/km2) 0.093 0.168 0.130 0.032 0.032 0.060 0.085 0.141
1990s Average 0.111 0.188 0.133 0.030 0.009 0.065 0.135 0.218
1980–89 Annual 3.0 3.3 3.2 ⫺0.1 1.8 3.2 8.9 4.2
1990–99 growth rate 1.8 1.1 0.2 ⫺0.5 ⫺11.3 0.8 4.6 4.4
LT Elasticity 0.15
1980–89 Contribution to 0.6 0.5 0.50 ⫺0.01 0.3 0.5 1.3 0.6
1990–99 growth 0.3 0.2 0.03 ⫺0.10 ⫺1.7 0.1 0.7 0.7
Source: Authors’ calculations.

Breaking the Barriers to Higher Economic Growth74
Appendix Notes
1. In the case of macroeconomic stability, for example, the first component is weighted by 40/72 and
the second by (72-40)/72, where 40 is the explanatory power of the first principal component and (72-
40) the explanatory power of the second one. These coefficients are normalized by dividing by 72,
which corresponds to the percentage of the variance of the initial indicators, explained by the two
principal components selected.
2. For a better reading of the graphs and of the composite indicators, we have inverted the sign of the
macroeconomic and external stability indexes. They can be interpreted as the efforts to reform the
economy. The same thing has been done for the human capital indicator (see signs of the initial and
composite indicators in tables A2.1 to A2.5)
3. Iran could not find long-term loans in the international market.Algeria and Tunisia followed a more
cautious policy.
4. The case of Syria has to be treated with caution because of the Russian debt.
5. These indicators could, for example, be the nature and the quality of the loans, the ratio of reserves
of the banks, or the percentage of loans through private banks. If such indicators were taken into con-
sideration, the picture of the banking system would be different, with, in particular, a not always
healthy sector, a very present State, and a slow pace of privatization (specifically in Algeria, Egypt,
Tunisia, Iran, Syria; see Nabli, 2000).
6. Our trade openness indicator excludes the “natural” openness of the economy (see Frankel and
Romer, 1999), as well as the exports of oil and mining products.
A12-6. Contributions to Growth: MENA Countries Summary
Years (%) MENA Tunisia Morocco Algeria Egypt Jordan Iran Syria
Macroeconomic stability
1980–89 Contribution to ⫺1.3 0.2 ⫺0.01 ⫺1.0 0.4 0.1 ⫺2.6 ⫺2.4
1990–99 growth 0.4 0.4 0.70 0.8 13.0 0.6 0.3 0.6
External stability
1980–89 Contribution to ⫺1.0 0.1 ⫺0.8 0.1 ⫺0.9 ⫺0.4 ⫺2.4
1990–99 growth 1.0 0.3 1.2 0.1 2.5 ⫺1.1 0.6 0.8
Structural reforms
1980–89 Contribution to 0.2 0.6 0.1 ⫺0.1 0.4 0.5 ⫺0.05 ⫺0.05
1990–99 growth 0.2 0.7 0.5 ⫺0.8 ⫺0.03 0.6 ⫺0.10 0.4
Human capital
1980–89 Contribution to 1.1 1.2 1.0 1.2 0.7 0.6 1.3 1.3
1990–99 growth 0.9 0.9 0.8 1.2 0.8 0.4 1.2 0.9
Physical infrastructures
1980–89 Contribution to 1.4 1.3 1.0 1.0 1.0 1.6 1.8 1.5
1990–99 growth 1.0 1.0 1.3 0.5 ⫺0.9 0.5 1.7 1.4
Investment
1980–89 Contribution to ⫺0.3 0.1 0.1 ⫺0.3 ⫺1.5 ⫺0.24 ⫺0.1 ⫺0.03
1990–99 growth ⫺0.1 ⫺0.1 ⫺0.1 ⫺0.2 ⫺0.4 0.01 0.15 0.15
Total
1980–89 Contribution to 0.1 3.5 1.4 0.9 0.1 2.2 0.4 ⫺2.1
1990–99 growth 3.4 3.2 4.4 1.6 3.3 1.0 3.9 4.3
GDP per capita
1980–89 Annual growth 0.0 1.6 1.4 0.1 2.2 ⫺0.7 ⫺3.5 ⫺1.1
1990–99 rate 1.7 3.2 0.5 ⫺0.3 1.9 0.4 2.0 4.4
Source: Authors’calculations.

Part I: Growth, Reform, and Governance 75
7. Trade policy should, however, be analyzed through other indicators,such as average tariffs and non-
tariff barriers, which are not available on an yearly basis for a large sample of countries. By using these
kinds of indicators from the mid-1980s and for a smaller sample of countries, Dasgupta, Keller and
Srinivasan (2002) have shown that trade policy in MENA countries has historically been among the
most restrictive in the world.
8. The analysis is, however, restricted by our limited number of indicators; that is to say the road net-
work and the number of telephone lines per ,1000 population. Results would also have been different
if we had included in our sample the Gulf economies, as well as other oil-producing countries.
9. Tunisia and Morocco offer a clear example of this situation. Despite a sharp economic slowdown
during the 1980s, investment increased during this period.

77
After Argentina:
Was MENA Right to Be Cautious?
Mustapha K. Nabli
3
Much has happened over the past few months to put the Middle East region
into the spotlight of international attention. Perhaps most notable were the
events following September 11. It is an increasing phenomenon that we see
articles in the newspapers devoted to the region, sometimes looking to under-
stand the causes of social exclusion, but almost always trying to understand
better the region’s economic and social structures. Our regional department
in the World Bank is regularly called upon by some journalist seeking infor-
mation about where the region stands in terms of economic growth, employ-
ment prospects, and political stability.
In this investigation, which has heightened of late, the progress that the
region has made over the last decade, in terms of both achieving macroeco-
nomic stability and initiating programs of structural reform, has become bet-
ter known. Starting in the late 1980s, a handful of countries in the region—
Morocco and Tunisia and, soon after, Jordan—embarked on programs of
macroeconomic stabilization and policy reform. By the 1990s, nearly all of the
non-Gulf Cooperation Council (GCC) countries in the region followed suit,
as did several of the Gulf economies. The reasoning, of course, was to create
an environment in which the private sector could emerge and become an
engine for higher and sustainable economic growth—crucial for employment
creation as well as for social cohesion. But at the same time, it must be con-
ceded that the Middle East and North Africa region’s (MENA) progress with
structural reform has been incomplete. Financial sectors remain weak; trade
Speech at the Third Mediterranean Social and Research Meeting; Florence, Italy; March 2002.

Breaking the Barriers to Higher Economic Growth78
liberalization remains incomplete, with continuing moderate to high protec-
tion levels; public ownership remains high; and the regulatory framework and
supportive institutions for growth have not materialized.
As a result, the region has often been chastised by institutions such as the
World Bank for not proceeding more quickly and deeply with its commit-
ments to liberalization. But juxtaposed with this sense of frustration about the
region’s lack of continued progress with economic liberalization comes
another event that has served in some circles to cast some doubt on what can
be referred to as the “Washington Consensus” on economic reform programs:
that being the recent crisis in Argentina. Until quite recently, Argentina was
the darling of the international financial community. Within a single decade,
it had embarked on an unprecedented effort, and had undone four decades of
state interventionism. It deregulated most markets, but deregulation of labor
markets was not deep enough. It privatized most state-owned enterprises,
including difficult giants like the oil industry, railroads, and airlines. It radi-
cally reformed its social security system, substantially strengthened its finan-
cial sector, and swiftly aligned its economy to market forces. These policies
revived the economy and brought stability to the country. But in a sudden
shockwave, the country became engulfed in an unprecedented institutional,
political, economic, and social crisis. Within a few days in December 2001, it
changed presidents five times, its payment systems collapsed, it defaulted on
its foreign and domestic debt, and it violated property rights. Argentina’s real
economy is imploding and the country’s social fabric is being torn apart.
Looking at the current crisis in Argentina, the countries in the MENA
region may be asking themselves: Were we right to be cautious? Did Argentina
commit itself to an economic model that, in the end, suffers from some fatal
flaws?
Before taking the Argentine crisis as confirmation of the need for guarded
steps into economic liberalization and deregulation, one needs to look at its
roots. I believe there are indeed lessons for the MENA economies from
Argentina’s example. Broadly, three factors, in combination, contributed to
the collapse of the Argentine economy: first, a strict exchange rate rule that
reduced competitiveness and the capacity to grow; second, a loose fiscal poli-
cy mix that generated an increase in public indebtedness; and third, a contin-
uous weakening of the institutional environment. However, the collapse was
not caused by liberalization of the economy. Liberalization strengthened the
economy. It initially reduced economic imbalances, and the economy per-
formed remarkably well until 1997. But as I intend here to speak to the MENA
economies, if there is an additional lesson from Argentina regarding the role
that liberalization played in the crisis there, it is that liberalization cannot be
piecemeal. The liberalization process should present domestic enterprises
with heightened competitive pressures, both international and domestic, but
it should also allow them to respond —both positively to new opportunities

Part I: Growth, Reform, and Governance 79
and effectively to economic downturns. In Argentina, the lack of liberalization
of labor markets reduced the economy’s ability to adapt when it needed to
adjust under reduced competitiveness abroad.
Let me take a few minutes to highlight the events that drove Argentina into
crisis, and come back to these elements as they apply to lessons that can be
taken by the economies of MENA. While all three of the factors I mentioned
contributed in combination to the Argentine crisis, let me begin with the first
factor:
The Exchange Rate Regime
To understand the story, we have to go back to the 1980s and early 1990s. In
1989, Argentina experienced hyperinflation at the same time as its first dem-
ocratically elected president since the military dictatorship had to take an
early leave. The incoming authorities at first failed with several stabilization
attempts. But under very difficult circumstances, in 1991, the government
made a radical change. It established a currency board that fixed the U.S. dol-
lar price of the Argentine peso on a one-to-one basis, replacing its floating rate
regime. This currency board, set up by law, provided a strong and reliable
anchor. It succeeded in arresting inflation and providing price stability over
the next 10 years.
Fixed exchange rate regimes require different preconditions from floating
rates in order to be sustainable. Moreover, they also require a flexible econo-
my that can swiftly adjust to exogenous shocks. During the late 1970s and
early 1980s, Chile implemented a stabilization plan based on a fixed exchange
rate. This plan failed, mainly because during its implementation the dollar
appreciated substantially, and Chile could not withstand the appreciation of
its own currency vis-à-vis the rest of the world.
On the other hand, a few years later, in 1985, Israel initiated a stabilization
plan very similar to that of Chile and anchored its economy to the dollar
through a fixed exchange rate.Israel’s plan was successful, but this success was
due in no small measure to the fact that in this period the dollar actually
depreciated substantially.
The lesson of many experiments with fixed exchange rates is that most
economies adjust better to real depreciation than to real appreciation of their
currency.
During the 10-year period that Argentina kept its currency board, the U.S.
dollar appreciated substantially, and Argentina’s main trading partner, Brazil,
devalued its currency by about 40 percent in real terms. Consequently,
Argentina’s multilateral real exchange rate appreciated substantially, and this
disequilibrium required adjustment through deflation and unemployment.
During the early 1990s, the economy grew at very high rates, driven by a
surge in capital inflows and the fact that the country was bouncing back from

Breaking the Barriers to Higher Economic Growth80
a deep crisis. In 1995, the economy suffered the “tequila effect”of the Mexican
crisis, with a run on the banks and the sudden reduction in capital inflows.
Later, in 1997, the Asian crisis led to a sharp fall in commodity prices. This
meant that the Argentine equilibrium real exchange rate had depreciated. But
instead of depreciating, the actual real exchange rate continued to appreciate,
first because of the dollar’s appreciation, and later because of the 1998 sharp
depreciation of the Brazilian Real.
The exchange rate could not adjust to its new equilibrium value by nomi-
nal depreciation. It had to adjust through prices and quantities. Given the
inflexibility of prices to downward pressures, the adjustment to lower nomi-
nal income had to be done through quantities, and real income took a dive.
It is at this point that the effects of the other factor propelling the crisis
became evident:
Lack of Fiscal Discipline and Increasing Public Debt
A central issue in the crisis was the lack of fiscal discipline and its contribu-
tion to increasing public debt. During the years of massive privatization in
Argentina, government expenditures were adjusted upward with the huge
influx of privatization revenues (see figure 3.1). Between 1989 and 1994, for
example, expenditures kept almost identical pace with overall revenues—
including privatization proceeds. Once privatization revenues began to
decline, however, an adjustment of spending never took place. Spending as a
percentage of GDP remained almost unchanged between 1994 and 1997,
despite the decline in overall revenues of about 1.5 percent of GDP.
The last three years of the Menem administration (1997-99) were charac-
terized by excessive government spending, at both the federal and provincial
levels. Government spending as a percent of GDP increased from 14 percent
to 22 percent. At the time, the political system was split between the national
Figure 3.1. Government Spending as a Percentage of GDP and Revenues (including debt and privatization)
0
5
10
15
20
25
1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999
expenditures
revenues
percent of GDP

Part I: Growth, Reform, and Governance 81
government (following a proprivatization, promarket policy) and the provin-
cial governments (oriented to high-cost social programs). Because the provin-
cial governors had direct control over national legislators, the Menem regime
used revenues gathered from privatization and foreign investment to shower
the provinces with money, to win their support for the national program.1
Lack of economic growth, combined with an expansion in government
expenditures, generated a fiscal deficit that grew from 0.15 percent of GDP in
1994 to 2.4 percent in 2000. To achieve fiscal balance without distorting the
economy, the government could have lowered its expenditures. Instead, it
chose to finance the fiscal deficit by raising taxes and incurring additional
debt in financial markets and with the IMF. Raising taxes proved to be a bad
tactic; it generated more tax evasion and suffocated an already hobbled pri-
vate sector.2
The extra tax burden and the reduction in credit to the private sector
diminished the investment rate and reduced productivity. This, in combina-
tion with the loss of competitiveness, drove the economy into recession.
The recession led to double-digit unemployment. Early in 2000, the new
government made the crucial mistake of increasing tax rates in the middle of
a deep recession. Naturally, this exacerbated the recession, and instead of
increasing fiscal revenues, the tax hikes reduced them. The higher fiscal deficit
had to be adjusted with lower expenditures and financed with higher debt
issues. Lower expenditures also worsened the recession. In this setting, the
country debt became very risky, and interest rates on public debt soared.
With the soaring interest rates, the burden of the debt became unsustain-
able. The IMF and other international financial institutions (IFIs) continued
supporting the currency board with huge multibillion-dollar loans. This did
not solve the underlying disequilibria. By late 2001, the government had
defaulted on its external debt held by domestic creditors, mainly banks and
pension funds. It forced a swap of its high-yielding debt for new debt “guar-
anteed with future fiscal revenues” and yielding only 7 percent. The policy to
restructure, on a “voluntary” basis, a good part of the domestic debt improved
finances in the short term but complicated debt management for years to
come, and signaled to the markets clear problems with Argentina’s debt sus-
tainability capacity, despite the seemingly comfortable levels in traditional
debt indicators. The market interpreted this as a clear sign of unsustainabili-
ty, which resulted in large withdrawals from the banking system.
The Central Bank issued a regulation putting a ceiling on interest rates.
This was another element that helped to trigger the banking crisis. Banks had
been paying a rate much higher than the new ceiling to keep deposits inside
the system. The fixing of interest rates was the fatal blow to stability; a full-
fledged currency crisis ensued. The flawed and time-inconsistent macroeco-
nomic policy, the price fixing, a very weak president, and this not-so-
“voluntary” debt swap with institutional investors and banks led to a run on

Breaking the Barriers to Higher Economic Growth82
the banks and the collapse of the payment system. The financial fortress built
over a decade was destroyed overnight.
Weakening of Institutional Environment
This component played a crucial role in destroying confidence in Argentina’s
economy and political institutions. It started gradually, during President
Menem’s unsuccessful attempt at reelection, which included a significant
loosening of the fiscal stance. It was followed by the Alianza strategy, in 1999,
to defeat Menem rather than govern. There followed a crisis in the senate in
2000, associated with the mild labor reform and the resignation of the Vice
President in October 2000. This crisis tarnished the possibility of serious
reform in the labor market in the future.
The weakest link was the fiscal regime that tied the federal government and
the rest of the country together, which remained discretional and full of loop-
holes. Provincial governments depended upon on taxes raised at the federal
level, while the municipalities depended upon taxes raised at the provincial
level, with no balance between the political benefits of public services and the
political costs of raising revenues. As the budget situation deteriorated, trans-
fers from Buenos Aires to the provincial governments became a matter for
day-to-day political bargaining.3
As the situation became increasingly complex, the rescue package from the
IMF, in December 2000, gave some breathing room, but generated further
(and excessive) loosening of the fiscal stance and culminated with the IMF
losing confidence in Argentina’s capacity to adjust.
The fiscal crisis, political mess, and debt overhang led the government to
seek an orderly solution—restructuring public debt—but this proved impos-
sible, as ordinary Argentines and foreign investors opted out of the system.
Figure 3.2. GDP Growth, 1988–2001
⫺8
⫺6
⫺4
⫺2
0
2
4
6
8
10
12
14
percent
1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

Part I: Growth, Reform, and Governance 83
The market attempt at working out the money overhang led to the controver-
sial decision to freeze the deposits. The public banks became a great liability
as the crisis developed, pushing the country deeper into crisis, as the govern-
ment lacked the political support to deal with them.
What can the MENA economies learn from the crisis in Argentina? Let’s
begin by talking about financial system, and the relationship between the
financial system and macroeconomic fundamentals.
In 1999, Argentina had one of the strongest banking systems in Latin
America. This was the result of several strategic decisions, including: requir-
ing banks to hold capital ratios much higher than those demanded by the
Basel Accord principles; imposing high loan-loss provisions and strict inter-
national accounting standards; actively favoring the entrance of large foreign
banks; requiring that banks issue subordinated debt; demanding that banks be
rated by international rating agencies; imposing reserve requirements higher
than 20percent on liquid deposits, not withstanding their initial time struc-
ture; addressing the lack of a lender of last resort by buying insurance from
international banks, in the form of contingent credit lines that would trigger
in the case of a currency run.
During its convertibility experiment, Argentina increased its debt to the
IFIs in an unprecedented way. In exchange for credit, the IFIs required and
agreed with the government on the enactment of multiple and difficult meas-
ures to strengthen its capital markets. On every aspect related to capital mar-
kets, Argentina “went by the book” as written by the IFIs and, in this way, con-
structed a real financial fortress. However, one year later its banking system
was on the brink of complete insolvency. The meltdown was instantaneous.
The question is “Why?” The answer is: economic shock resulting from the
accumulation of disequilibria related to time-inconsistent macroeconomic
policies and the loss of credibility in economic and political institutions. The
“books” had left out a critically important chapter.
The lesson from this recent Argentine episode, as well as from others, is
that microeconomic reform in the financial sector is very important, but not
sufficient. No matter how strong the financial sector is; no matter how much
expertise and confidence foreign banks may bring to the financial system, and
how large their share of total lending; no matter how well the Basel regula-
tions are put into place and enforced; if macroeconomic policies generate
large disequilibria, they may lead to a real systemic implosion. This is a lesson
that should be learned by other countries that are currently running signifi-
cant macroeconomic imbalances.
Another lesson from the Argentine episode relates to the deleterious influ-
ence of government’s debt finance on capital markets. Even during the exu-
berant period of the currency board experience and the substantial strength-
ening of the banking system, the Argentine stock exchange languished. The
number of listed instruments was steadily reduced from 204 in 1989 to 125 in

Breaking the Barriers to Higher Economic Growth84
1999. The number of listing companies also declined sharply. The pension
reform provided funds that were mainly used for government debt and bank
CDs. The stronger companies listed in foreign stock exchanges. Liquidity was
substantially reduced.
The decline of the Buenos Aires Stock Exchange (BASE) during the 1990s
can be directly related to the government’s debt finance crowding out the pri-
vate sector. During 1996-2000, as an average, 86 percent of net primary issues
dealt through the stock exchange were fixed income instruments, and of those
76 percent were government issues. The overwhelming presence of the gov-
ernment was even more dramatic in relation to liquidity: More than 90 per-
cent of trade was done on the Treasury’s fixed income products.
Now, let me return to the MENA region, and look at what we have learned
from the Argentine crisis. To begin, let’s address the issue of a fixed exchange
rate. In the MENA region, fixed exchange rates are still pervasive. In two-
thirds of the economies of the MENA region, fixed exchange rate arrange-
ments, whether explicit or implicit, are in place, despite the fact that world-
wide, only half of economies operate under a de facto fixed exchange rate sys-
tem (and only one-third have official fixed arrangements). The preoccupation
with macroeconomic stability during the 1990s played a large role in main-
taining nominally fixed and stable exchange rates, but it has meant that one
important tool to make exports more profitable was surrendered.
There are many other reasons for the region continuing to rely on more
rigid exchange rate arrangements, but both debt and oil play a role. By main-
taining a fixed regime, and in particular an overvalued exchange rate, the gov-
ernments in the region have an easy means to decrease their payments on
foreign-denominated debt at the expense of domestic borrowers, who face
tighter monetary policy, and of national industries whose competitiveness is
weakened. In addition,the public sector’s role as exporter of oil provides addi-
tional incentive to adopt a fixed regime, as it allows for directly subsidizing
imports by the public sector.
But the lesson of Argentina on the fixed exchange rate should send a sig-
nal about what can happen when an overvalued rate is maintained, and what
that implies for domestic industries and laborers. In Argentina’s case, the
fixed exchange rate contributed to reduce the competitiveness of the econo-
my in a period of strong appreciation of the dollar. The devaluation of the
Brazilian Real in 1999 deepened competitiveness problems from which the
economy never recovered. The loss in competitiveness, and the sustained
decline in real output since 1999, helped to destroy Argentina’s credibility
and increase its cost of capital. As the cost of capital increased, the needed
downward adjustment on wages became too much to bear, particularly when
labor market flexibility was weak. For each 1 percent increase in the interest
rate, wages needed to decline by about 3 percent to maintain competitive-

Part I: Growth, Reform, and Governance 85
ness. During 1990-98, the adjustment in unit labor costs came from increas-
es in productivity that preserved the level of wages. Since 1999, the adjust-
ments have come through reduction in labor costs and wages, which has
depressed consumption. Since the tax structure of Argentina is based on
indirect taxes, the reduction in consumption reduced tax receipts and deep-
ened the fiscal imbalances, generating a fiscal crisis when government expen-
ditures were impossible to contain.
What are the lessons that emerge on debt, in terms of its role in the
Argentine crisis? The most important is that debt, even small debt, can
become a major issue once credibility is lost. In Argentina’s case, debt sustain-
ability became an issue as the economy imploded, despite what we would
term moderate levels of public debt. But when the spreads increased, the
capacity to rollover or refinance public debt abroad vanished,forcing the gov-
ernment to impose a “voluntary” debt swap on domestic banks and other
institutional investors. Although restructuring part of the debt improved
finances short term, it also sent a powerful signal to markets of Argentina’s
debt sustainability, despite the appearance of comfortable debt levels by tradi-
tional indicators.
Within the MENA economies, outside of the Gulf economies, the size of
public sector debt is substantially higher than that Argentina exhibited at the
time of the crisis. At the time the crisis unfolded, public debt as a percentage
of GDP was about 56 percent in Argentina. That compares to debt ratios of 85
percent in Egypt, over 60 percent in Algeria, Morocco, and Tunisia, over 100
percent in Jordan, and close to 170 percent in Lebanon. So, at the start, the
MENA economies are saddled by extensively higher amounts of debt.
What is of greater importance than the debt level, however, is the maturity
of these debts. It is not enough to comfort oneself into security by observing
that the structure of debt is predominantly long term. Even long-term debt at
some point becomes due. And when debt becomes due, the issue of credibili-
ty becomes important. If an event makes the payment of that debt impossible,
then the economy must have credibility that the debt will be repaid, so that
debt restructuring or rollovers can occur. In the case of Argentina, the lack of
credibility put a complete halt to the possibility of rolling over the debt after
the first “voluntary” debt swap. The problem was not how fast the debt was
projected to grow, but that no one wanted to lend more resources to
Argentina. The loss in the credibility of Argentina’s capacity to repay its debt
became an anchor in the crisis, and a run on the banking system by deposi-
tors occurred.
How is credibility maintained? In the case of Argentina, a decade of credi-
bility was ultimately lost through reluctance to address serious reform of the
labor market combined with the significantly loosened fiscal stance leading up
to the 1999 elections.

Breaking the Barriers to Higher Economic Growth86
What should the MENA countries learn from the Argentine crisis regard-
ing the loss of credibility? There are many lessons, but they cannot be encap-
sulated into a few key economic indicators. Credibility is the fusion of many
factors. Certainly, though, macroeconomic stability is vital, and upholding a
tight fiscal stance sends a strong signal of the government’s commitment to
maintaining that stability. Most of the MENA economies, after deteriorating
budget deficits in the early 1990s, embarked on programs of macroeconomic
stabilization and structural reform designed to bring budgets back into bal-
ance. But several economies have also let their fiscal stance loosen since then.
In Egypt, the near budget balance realized in 1997 turned to deficit again by
1998, which through the late 1990s averaged 4.6 percent of GDP. In Jordan,
the budget surplus of 1 percent of GDP in 1995 also turned to deficit the fol-
lowing year, and by 1998 was close to 6 percent of GDP. In Algeria, the budg-
et surplus of 3 percent of GDP in 1996 was followed by a budget deficit of 4
percent of GDP by 1998. While fiscal imbalances are not steadily growing in
any countries in the region—which would send a signal of an impending
crisis—we are seeing a laxness in maintaining fiscal determination that is sim-
ply unsafe given the levels of debt in the region.
Second, credibility cannot be bought through rhetoric. Markets are keenly
observant of actions. When Egypt’s progress in structural reform lagged,
Standard and Poor’s revised its outlook on the economy from stable to nega-
tive (though it did not adjust the country’s risk rating), and Thomson
Financial revised Egypt’s sovereign risk rating outlook from positive to stable.
While the comprehensive macroeconomic and structural reform programs
espoused by many of the MENA economies in the early 1990s created an exu-
berant boost in their economic outlooks, markets will not wait forever. The
MENA region must move beyond “stroke of the pen” reforms to the more
serious, and challenging, issues that obstruct the development of a strong pri-
vate sector. Unless the private sector begins to see itself as an independent
source of growth and productivity in the economy, and society begins to
underpin this change economically and politically,it is unlikely that any of the
past economic reforms in themselves will be adequate.
It is not possible to go into detail about all of the issues that hinder the
development of strong private sector-led growth in the region, but I will point
out a few.
To begin, there is the size of the public sector. Governments may account
for as much as 40-60 percent of gross domestic output and of employment in
the region. This includes continuing high expenditures on military and social
services, which account for the large size of the public sector. The big role of
the state—a sector that essentially has low productivity and limited inherent
potential for productivity gain—is a drag on growth in most economies in the
region. Efforts to reduce the public sector through rationalizing public

Part I: Growth, Reform, and Governance 87
employment, improving performance through better incentives and institu-
tions, and privatization of goods and services that could be produced more
efficiently in the private sector have begun in many countries. But by and
large, these efforts remain slow and halfhearted to date.
Second, there is the issue of trade reform. Trade policies in the region
remain among of the most restrictive in the world. The degree and speed of
integration into the world economy are low. Tariff rates remain high and the
extent of nontariff barriers significant. A number of policy moves across the
region are expected to lead to greater trade openness, stimulating integration
and, it is hoped, growth. Most notable are the EU association agreements
signed by Tunisia, Morocco, Jordan, and Algeria. This type of effort needs to
be systematic.
Among the most important reforms is that of the banking sector. However,
it is lagging, particularly as evidenced by the slow progress of privatizing state
banks in countries including Algeria, Egypt, Tunisia, Iran, and Syria. While in
a number of countries the banking sectors are relatively healthy, this is not the
case in most, and financial sector development remains a principal constraint
for the development of the private sector and for growth.
Finally, the region needs a virtual overhaul of its system of property rights,
better legal systems, and improved contract enforcement mechanisms.
I will end this talk now, going back to where I started, and ask whether the
MENA countries should be cautious, given the events in Argentina. The
answer, of course, is yes. Yes, they should be cautious about the factors that
precipitated the crisis in Argentina, including debt sustainability, fiscal deter-
mination, and exchange rate management. And in terms of reform, yes, there
is a lesson of caution here for the MENA countries as well. But the caution
does not suggest slowing the reform agenda in order to avert the type of cri-
sis that Argentina is experiencing. If anything, the caution is that liberaliza-
tion, whatever its speed, must be complete. The liberalization of one sector
can be undone by the lack of liberalization of another sector. The elimination
of trade barriers, for example, can be undermined if the financial sector is not
liberalized, and if public sector banks control the financing of imports.
Changes in the investment code are moot if labor laws impede the ability of
firms to maintain competitiveness.
And so, I hope that the crisis in Argentina does serve some purpose for the
countries of MENA in evaluating their steps forward, and that as a region, it
renews its commitments to furthering and deepening the economic and
structural reforms it began—so that higher and sustainable economic growth
can be ensured, so that employment creation will meet the demands of a
growing workforce, so that needs are met and opportunities expanded for the
poor, and so that a crisis such as the Argentines are currently facing never
occurs in the MENA region.

Breaking the Barriers to Higher Economic Growth88
Notes
1. Mark Reutter; News Bureau; University of Illinois at Urbana-Champaign; December 21, 2001.
2. Ana I. Eiras and Brett D. Schaefer; Backgrounder; The Heritage Foundation; April 9, 2001.
3. Center for International Development.“The Grand Illusions.”Andres Velasco. 2002.

89
Restarting Arab
Economic Reform
Mustapha K. Nabli
4
Achieving greater international competitiveness often depends upon policy
reform. Increasingly, the development community has come to appreciate not
only the “what” of policy reform, but the “how.” Economic policy reform is far
from a straightforward undertaking. It is a deeply political issue. It affects the
balance of power between actors in society; at its core, it involves finding the
economic rents that have built up over the years and cutting them back; it
attacks the economic privilege that some have enjoyed for generations. It is lit-
tle surprise that carrying forward comprehensive economic reform is a pro-
foundly difficult task.
This has been acutely apparent from the reform experience of the Arab
region. Beginning in the mid-1980s, when several early reformers in the
region—Egypt, Jordan, Morocco, and Tunisia—all implemented a broad
range of macroeconomic policy and structural reform efforts, there was great
optimism for change in the region. These hopes were heightened when other
countries in the region successively followed suit. Now however, two decades
after the nascent reform movement, and observing the subsequent slowdown
in the pace of reform, and the inability to tackle deeper and more fundamen-
tal arrangements inhibiting growth, those in the development community
must contend with the importance of the myriad forces behind policy reform.
In the following paper, I review the reform experience in the Arab region
and discuss some of the factors which have inhibited the implementation of a
deeper reform agenda. Among other factors, soft budget constraints and
Published in the Arab World Competitiveness Report 2005. Reprinted by permission from Palgrave
Macmillan.

Breaking the Barriers to Higher Economic Growth90
weaknesses in governance have both contributed to a lack of momentum in
carrying forward the reforms needed for higher and sustained growth. The
region’s ability to continue to rely on oil, aid, and other strategic rents has
allowed it to delay implementing deeper and more difficult reforms.
Moreover, the lack of both accountable and inclusive governance mechanisms
has prevented the emergence of coalitions for reform in the region, a central
element of successful reform programs.
This paper first discusses the Arab reform agenda and the progress made to
date. It then examines the reasons behind the stalled reform agenda in the
region and ways in which to move the reform agenda forward. Concluding
remarks are then presented.
The Arab Reform Agenda and Progress to Date
Four recent reports published by the World Bank1spell out fundamental tran-
sitions needed in the Arab economies to move to higher and more sustainable
sources of growth, identifying three realignments in particular: a transition
from public sector- dominated to private sector-led economies; from closed to
more open economies; and from oil-dominated and volatile economies to
more stable and diversified economies. Critical for completing these transi-
tions will be better governance, improved quality of education, and greater
gender equality throughout the region. As one of the reports points out, few of
these recommendations are new to the citizens or government of the region.2The
Arab reform agenda is well known. And the urgency for reform is growing.
One of the fundamental challenges facing the Arab world over the next two
decades is job creation. In the next 10 years,3some 31 million new workers
will have been added to the stock of Arab laborers, expanding the labor force
by more than a third.4In the next 20 years, that figure grows to some 65 mil-
lion new jobs needed, just to absorb the growing labor force, or an expansion
in the labor force of almost 70 percent.
In addition, there are the region’s unemployed, who constitute about 13
percent of the labor force. If the region is also to address job creation for the
stock of unemployed, some 77 million new jobs will need to be created
between 2004-24, almost doubling the number of jobs in the region.
To give perspective on these numbers, figure 4.1 presents the employment
growth rates in several East Asian economies during periods of expansive job
creation. Some of these countries managed the highest rates of sustained
employment growth in modern history. In China, employment growth over
the two decades between 1980 and 2000 averaged slightly over 40 percent.
Korea’s employment growth over the 1980s and 1990s was slightly less than 60
percent. Malaysia’s employment growth was highest in East Asia over the
1980s and 1990s, with almost a 90 percent increase in jobs. These countries
listed can be considered to have had some of the greatest spurts in employ-

Part I: Growth, Reform, and Governance 91
ment growth over a sustained period of time. And the Arab region needs to do
even better.
Add to these demographics that the traditional modes of employment cre-
ation in the region are fast coming to an end. Public sectors are unable to be
the employment outlets they have been in the past. Labor migration prospects
have diminished.
This is an extraordinary challenge that is facing this region. It is estimated
that economic growth will need to average over 6 percent a year for a sus-
tained period of time to create these kinds of employment levels.
It is against this backdrop that the state of economic reform in the region
needs to be examined. We generally trace the Arab reform movement to the
efforts by a handful of economies in the region in the mid to late 1980s toward
macroeconomic stabilization and structural reform. Following the collapse in
oil prices, and facing high debt, deteriorating budget deficits, and a lack of
growth, a few countries—Morocco, Tunisia, and Jordan, and subsequently
Egypt—adopted programs aimed at restoring macroeconomic balances and
promoting the private sector as an engine for growth. By the late 1980s and
early 1990s, most other countries in the region had followed suit, adopting
some form of economic stabilization.
The macroeconomic stabilization achievements were non-negligible, as
most Arab countries undertook better macroeconomic policies. Debt renego-
tiations and write-offs helped reduce the very large and unsustainable debt
burdens and helped achieve improvements in the fiscal deficits. Continuing
macroeconomic stabilization efforts have helped to contain inflation to less
than 2 percent over the past three years, down from an average of 11 percent
in 1991.5Fiscal deficits have been reduced by about two-thirds, and total
external debt has declined from an average of 40 percent of gross national
income in 1990 down to 28 percent in 2002,6with the largest declines
Figure 4.1. Required Job Growth in Arab Region, versus International Experiences
0
10
20
30
40
50
60
70
80
90
100
percent
China
1980–2000
Korea
1980–2000
Malaysia
1980–2000
Singapore
1980–2000
Arab World
2004–24
Required
Growth
Japan
1960–80
Source: World Development Indicators; U.S. Department of Labor; World Bank data.

Breaking the Barriers to Higher Economic Growth92
achieved in Egypt, Morocco, Yemen, and Jordan. These achievements have
largely been sustained when one looks at the standard indicators of macroeco-
nomic stability. However, macroeconomic stability is not a foregone conclu-
sion. Contingent liabilities have been building up in many of the Arab coun-
tries, related to the accumulating implicit debts from many sources, such as
the pension systems, the banking sectors, the public enterprises, and a variety
of explicit and implicit government guarantees. Thus, while macroeconomic
stability has been broadly achieved, the fundamentals behind stability have,in
some cases, weakened.
In terms of accompanying structural reforms, the results have been more
mixed. A few early reformers have implemented more intensive reforms
toward market-oriented, private sector-led economies, signing EU-Med
agreements, implementing tax reform, undertaking trade and financial sector
liberalization. Others have pursued reform more sporadically and slowly. Still
others have made more modest progress.But by and large, the pace and inten-
sity of the reform effort have been weak.
As evidence of this, some of the standard indicators of market-orientation
or private sector development point to the Arab region remaining well below
potential:
• The ratio of private investment to public investment hovers around 1.8,
and has remained substantially unchanged since the early 1990s. In com-
parison, the private to public ratio averages 5 in East Asia.
• Arab manufacturing sectors are small—almost half the typical levels of
other lower-middle-income economies.
• Trade integration—the ratio of trade to GDP—has actually declined over
the last 25 years, averaging about 90 percent in 1980, compared with only
65 percent today.
• Exports other than oil are about a third of what they could be,7given the
region’s characteristics in terms of natural endowments, size and geography.
••• Foreign direct investment could be five or six times higher.8
As a result, economic growth has lagged well below what is required to
meet the region’s employment challenge. Over the last decade, economic
growth in the region has averaged about 3.7 percent per year9—less than two-
thirds the rate needed over the next 20 years to meet the coming employment
challenges.
The Failure to Sustain Deep Reforms
What has prevented a more intensive reform effort? There are certainly the
regional security issues and conflicts which have been used to justify the
maintenance of the status quo and the need to avoid risks of instability due to
change and reform. And while there is a heated debate about the merits of

Part I: Growth, Reform, and Governance 93
such justifications, and it can be argued that such factors have played some
role—which varies from country to country—in delaying reforms, there are
two other major fundamental factors which stand out, and which are certain-
ly more important.
The first has been the ability to continue to rely on oil and strategic aid to
delay implementing a deeper economic reform agenda. The substantial rev-
enues from oil, which declined throughout the 1980s and 1990s,remained sig-
nificant, and have even spiraled upwards over the last several years, giving
Arab leaders and the public a temporary sense of economic health (figure
4.2). This, along with foreign aid and strategic rents, has permitted Arab gov-
ernments to adopt limited reforms and postpone the fundamental reforms
needed for higher growth and employment creation.
There is a line of thinking that deep economic reform movements result
only from fundamental change, either from leadership change (regime
change, or a shift in governing coalition); or from dramatic economic
change—in other words economic crisis. In many Latin American countries,
for example, economic reforms were undertaken only when the “economic
conditions had deteriorated sufficiently so that there emerged a political
imperative for better economic performance.”10 Or, put another way, reform
often is only adopted “once the possibilities of throwing money at the prob-
lem are foreclosed.”11 Indeed, the reform movement initiated in the Arab
region was a direct consequence of the oil price collapse of the early 1980s.
There is certainly room for debate about whether crisis is necessary for
reform, and there have been at least a few examples of gradualism in reform
which were not driven by open crises—in China and Vietnam, to name two.
But what is true is that the availability of a soft budget constraint in the region
Figure 4.2. Net Oil Export Revenues in Arab OPEC, Selected Years
0
10
20
30
40
50
60
1972 1980 1998 2003 2004
0
50
100
150
200
250
Algeria
Iraq
Kuwait
Libya
Qatar
Saudi Arabia
constant 2004 $ billions
Saudi Arabia
United Arab
Emirates
Source: U.S. Energy Information Administration.
Note: 2004 revenues based on estimated price per barrel of $39.

Breaking the Barriers to Higher Economic Growth94
has delayed a change in the economic and social perspective needed to imple-
ment deeper reforms.
But the second and assuredly more important factor behind why reforms
have stagnated has been the lack of coalitions emerging to press the govern-
ment for deeper reforms and better policy choices. And underlying the lack of
coalitions for change in the region are i) a resistance to reform by those
included and protected by the status quo arrangements, and ii) an inability to
mobilize and demand changes by those who might benefit from change.
The theory of collective action rests on the hypothesis that organized
groups apply more pressure on politicians than unorganized groups.12
Inherently, a central problem with successfully implementing reform is that
those whose rents will be lost know who they are, but those who might bene-
fit from reform are generally more dispersed and have a much vaguer notion
of how economic reform will benefit them. Indeed, they may feel that they will
lose with reform. In the Arab world, this situation is complicated by gover-
nance issues that further hinder the ability for potential beneficiaries of
reform to band together for change.
Resistance by Those Protected by the Status Quo
Following independence, and in part to redress social inequity, a social con-
tract between the government and the public developed throughout the Arab
region, characterized by stateplanning, protection of local markets, and an
encompassing view of the state in providing welfare, social services, and even
employment. Economies became heavily protected and labor markets became
highly regulated, with a strong emphasis on employment protection.
Businesses developed under the patronage of governments, basically living off
the state.
After the initial reforms in the 1980s and early 1990s, often undertaken
under crisis conditions, and with economic conditions deteriorating or not
improving, resistance to reform has deepened among those groups whom the
social contract protects.
Internationally, one of the most effective lobbies for policy change is pri-
vate business. Although the yardstick of firm preferences comes down to the
ability to make profits, their interests are often closely aligned with those of
society: greater profits can allow for greater job creation, growth, and poverty
reduction.
But the private sector in most Arab economies is underdeveloped, with a
predominance of small and family businesses. This is particularly evident in
the manufacturing sector. In a recent paper,13 we constructed an index of the
lobby power of the manufacturing sector worldwide, estimated as an integra-
tive of the size of the manufacturing sector in total exports and the concen-
tration of manufacturing exports within the sector.14 Thus, countries in
which the manufacturing sector accounted for a large proportion of exports,

Part I: Growth, Reform, and Governance 95
other things equal, would be expected to be better able to influence govern-
ment policy. Likewise, the more concentrated the manufacturing sector, in
terms of a few export categories, the more manufacturers could be expected
to band together with a common interest to influence the government’s poli-
cies. From these estimates, we found that in the Arab region, the lobby power
of the manufacturing sector was lower than in any other region of the world
but Sub-Saharan Africa, with a measured index of approximately 0.22 in the
late 1990s,15 compared with an average worldwide of 0.43, and lobby indices
of over 0.4 in the OECD, South Asia, and East Asia.
The large-scale private sector, when it exists, is dominated by industries
that benefit from the status quo, manipulating government policy for their
own gain—the so- called state capture of government regulation. Actions
range from persistent awarding of large public sector contracts to a few well-
connected groups, to actual change in laws and regulations that lower the
costs or increase the profitability of such groups.16
Internationally, trade unions have often proven instrumental in organizing
support for comprehensive reform. In several Eastern European countries, for
example, trade unions played an important role in designing and implement-
ing large scale privatization programs. In the Arab world, independent trade
unions are rare17—trade unions were banned in several GCC economies until
recently, and in several other Arab countries, trade union membership is lim-
ited to a single national trade union. Throughout the Arab region, union
activity is strictly controlled. As a result, without real independence from the
political system, trade unions have not been effective in organizing the labor
force to press for reform. At the same time, for those within the system, tran-
sitioning to a new development model implies abandoning the job security
from which they benefited under the old social contract.
Figure 4.3. Manufacturing Export Lobby Power in Arab Region, Late 1990s
0
10
20
30
40
50
60
percent
East Asia
and
Pacic
Latin
America
Arab
Region
OECD South
Asia
Sub-
Saharan
Africa
World
Average
Source: Nabli, Keller, and Véganzonès, 2004.
Note: Regional averages weighted by national populations.

Breaking the Barriers to Higher Economic Growth96
In reaction to these deep-seated fears, the intellectual climate and underly-
ing ideologies in the Arab world have continued to be supportive of ideas
which are not friendly to reforms that promote markets, integration in the
world economy, a stronger role for the private sector, and less interference of
government in the economic sphere.
Inability to Contest Policies by Those Who Would Benefit from Change
But additionally, the ability of those “outside the net” to contest for changes
has been muted. Not all groups have been well represented in the region’s
social contract—indeed the region’s development paradigm has suffered from
great weaknesses in inclusiveness. Under-represented groups include the
young and educated who are unemployed; small businesses and those young
entrepreneurs who seek to enter protected markets but have difficulty access-
ing finance; consumers who pay high prices and get low-quality public serv-
ices; small farmers, and so on. But these groups have not been able to unite for
change.
The inability to form coalitions for change in the Arab region stems from
two central problems. The first is the intrinsic logistical problem of potential
beneficiaries of reform mobilizing for change when these groups are often dis-
persed. This is not a problem unique to the Arab world. Consumers, for exam-
ple, are generally the primary beneficiaries of trade reforms that would allow
for lower import prices, but it is intrinsically difficult for them to come
together for change, and this is the case throughout the world. However, what
exacerbates the problem in the Arab world is that the region’s governance sys-
tems directly hinder groups from effectively uniting for change.
To unite for change, groups need certain central rights. They need access to
information to formulate choices, they need the ability to mobilize, and they
need the ability to contest policies that are poor. But these rights are not pres-
ent in the region. Government information is not accessible by the public.
Freedom of the press is carefully monitored and circumscribed in most coun-
tries. There are restrictions on civil society. There are restrictions on freedom
of association. And the ability to contest government policies is weak.
More generally, the Arab region suffers from fundamental weaknesses in
governance, in terms of both inclusiveness and public accountability.
Inclusiveness reflects the notion that everyone who has a stake in development
and wants to participate in governance processes can, on an equal basis with
all others. Accountability reflects the notion that governance processes are
known and can be contested.
In the Arab region, weaknesses in inclusiveness are reflected not only in dif-
ferential business opportunities, with large or privileged firms being awarded
contracts or favorable legislation, but also in rural dwellers having fewer pub-
lic services; in gender inequalities in voice and participation; and in nepotism,
tribal affinity, patronage, or money determining who gets public services and

Part I: Growth, Reform, and Governance 97
who does not. Weaknesses in accountability are reflected in limited access to
government information, limited freedom of the press, and restrictions on
civil society.
It is this combination—the entrenched interests and uncertainties of those
“inside the net,” combined with lack of information and an inability to mobi-
lize and contest for better policies by those “outside the net”— that prevents
deeper reform from taking place.
Moving the Reform Agenda Forward
Perhaps the central challenge to moving the economic reform forward is
addressing the governance agenda. It cannot be viewed as a separate agenda. It
is a complementary and reinforcing agenda—to reform efforts in private
investment, trade, and economic diversification by changing governance
mechanisms, thereby improving capacity and incentives within government—
while fostering a larger role for civil society in governance. While better gover-
nance cannot guarantee optimal economic policies, it is indispensable to guard
against persistently poor policies, and to ensure that the good policies needed
to meet the Arab region’s growth potential enjoy legitimacy and are imple-
mented faithfully and with celerity.
The fundamental governance challenges are to strengthen the incentives,
mechanisms, and capacities for more accountable and inclusive public insti-
tutions and to expand allegiance to equality and participation throughout
society. Those good governance mechanisms are first steps in improving eco-
nomic policies that are themselves instruments for improving the climate and
incentives for efficient growth. Economic reform cannot proceed without
reform of the incentive structure in which reforms are embedded.
Enhancing inclusiveness requires adopting laws and regulations that secure
access to widely accepted basic rights and freedoms, including participation
and equality before the law. Broader public consultation, greater freedom for
the media, fewer restrictions on civil society, more equitable channels of
access to social services, and an end to discriminatory laws and regulations are
examples of measures that secure inclusion.
Enhancing accountability requires putting in place better systems of trans-
parency and contestability, including greater freedom of information, public
disclosure of government operations, wider public debate by an independent
media and civil society groups, and regular and competitive elections. It
involves strengthening the checks and balances within the government. It
involves reforming public administration.
Early in the reform process and beginning in the 1980s, many of the Arab
countries accepted an instrumental connection between economic and polit-
ical reform. Governments recognized that they needed to secure popular sup-
port for market-oriented economic reforms, and so several countries initiat-

Breaking the Barriers to Higher Economic Growth98
ed experiments in political reform, including increased opportunities for par-
ticipation by opposition political parties, expanded civil liberties and freedom
of the press, and increased participation in political life by civil society groups.
But these political openings have largely been reversed as Arab govern-
ments have sought to contain and manage the scope of political change.
Constraints on civil society and non-governmental organization (NGOs),
restrictions on the press, and other measures have restricted mobilization and
autonomous collective action.As a result, Arab government have also serious-
ly restricted the range of reforms that could be implemented. Reforms have
been limited to those that could be implemented through top-down manage-
ment, rather than those that would require the compliance and participation
of social groups whose well-being might be adversely affected by reforms.
Now, opening up the political space for greater public participation in pol-
icy is needed to move further with the economic reform agenda. Greater voice
in development is needed to take into account the needs and values of those
who have been excluded from the Arab development model. It is equally
important to ensure that in the transition to a new development model, the
economic outcomes are socially acceptable among those who have benefited
from the old social contract. The agenda is large. But without reforming the
governance structure, economic reform cannot proceed forward.
Managing Comprehensive Reform Programs
In addition to the governance agenda, the international experience with eco-
nomic reform can also provide some useful insights into managing compre-
hensive reform programs.
First, the region should look for ways to lock in reforms. This is where
international agreements can play a role. Following the 1994 peso crisis,
Mexico’s reform program would undoubtedly have gone far off course had it
not been for the North American Free Trade Agreement (NAFTA), which tied
the continuation of Mexico’s reform program to the international agreement.
Similarly, the reform programs in the accession countries of Eastern Europe
have undoubtedly been spurred forward by the desire for European Union
membership.
The EU Association agreements could be a positive step in that direction,
but they have not been able to act as the catalyst for change in the region orig-
inally envisaged. For a variety of reasons, the EU association agreements have
not provided enough of a “carrot” for the Med partner countries to launch
substantially deeper economic reforms. The main incentives of the EU agree-
ments have been financial, and the trade incentives have been limited.
Agricultural protection has been off limits, and many of the short-term
adjustments in terms of barrier lower, meeting standardization requirements,
and so forth have fallen to the Med countries. Combined with the problems
with rights of migration, trade in services, restrictive rules of origin, and

Part I: Growth, Reform, and Governance 99
rights of establishment, the EU association agreements have fallen short of
expectations.18
But with new members expanding the size and scale of the EU market, the
potential gains for Arab countries is expanding as well. With this is mind, the
region should look to strengthen these commitments, not only to realize the
increasing trade benefits, but also to lock in reforms and enhance credibility
for reform in general. The Euro-Med agreements will need to be improved by
both sides to strengthen their potential. The new EU neighborhood approach
is aiming at overcoming some of these limitations by linking the nature and
scale of benefits to the scope and extent of reforms in the EU neighboring
countries. The Arab countries should look to accelerate trade barrier reduc-
tions, liberalize services, and phase in domestic agricultural reforms. In turn,
the EU could offer immediate expanded access to its markets for agriculture,
as well as increased temporary migration, funds for managing transition costs,
and more efficient rules of origin.19 The region can also look to membership
in the World Trade Organization (WTO) as a means to enhance credibility
and lock in reform.
Second, the recent escalation of oil prices presents a valuable opportunity
for the region in terms of adjustment. Oil has always been a double-edged
sword for the region: the vast resources from oil, especially since the 1970s,
contributed to tremendous gains in development. At the same time, the con-
tinued existence of these oil revenues has certainly diminished the urgency for
reform and has contributed to procyclical fiscal policy. As a forthcoming
World Bank study puts it,20 a “tragedy of the commons” sets in during good
times when government revenues are high—no claimant on the government’s
budgetary resources internalizes the need for fiscal solvency, and political
imperatives thus cause the government to spend all of its resources in the
boom, leaving little margin of solvency to draw upon in order to finance fis-
cal deficits when times are bad.
But now, with the region on the path of reform, it has an invaluable oppor-
tunity to buffer the adjustment costs of economic transition with these rising oil
revenues. Several economies have already created oil stabilization funds to save
windfalls and buffer against future oil price declines. But it may be more useful
to earmark a proportion of the funds specifically for future economic transition
costs, such as an expanded social safety net for job losses. Having these windfall
revenues to draw on during the transition process gives the Arab region a sig-
nificant advantage over other regions in terms of its ability to sustain reform.
Third, another feature the region has in its favor is that it is not in open cri-
sis. To be sure, there are serious and compounding challenges that the region
is facing, but it has not reached the precipice.It can reform ahead of crisis, and
this substantially lowers the total costs of adjustment. In the next 10 years, if
employment grows at the rates averaged during the last decade, unemploy-
ment will have grown to almost 25 percent21 of the labor force. If the region

Breaking the Barriers to Higher Economic Growth100
initiates deeper, broad-based reforms then—with accompanying reductions
in public sector employment, exposure of inefficient firms to greater compe-
tition, and reduced oil resources—all of the costs of adjustment will be borne
at a time when it is significantly more difficult to handle them. So the region
should look at this as an opportunity to embark on deeper reforms, at a time
when the costs are more manageable and the resources exist to mitigate the
adjustment.
Fourth, the region needs to look for ways to build upon successes. The
international experience suggests that growth operates something like a snow-
ball, requiring a moderate degree of success in several areas simultaneously to
have impact. The economies of East Asia were characterized not only by the
breadth of reforms undertaken, but also by the manner in which small suc-
cesses were seized upon and developed. Several countries in the Arab region,
in particular several of the smaller GCC economies, are notable for approach-
ing reform in this more selective fashion—selective not because the reforms
undertaken were easier, but because they aimed at creating momentum
behind some specific agenda which might help to pilot and build up support
for the broader reform agenda. Bahrain, for example, has implemented sever-
al economic and political reforms, in an aim to position itself as a financial
and economic hub in the region. The United Arab Emirates has targeted the
growth of entrepot trade. But in most countries, while successes abound, they
have tended to be isolated and have not spilled over, creating positive feedback
loops of cumulative successes.
Finally, there are the issues of capacity to implement reforms. It is not
enough to make decisions about reform and to design the best policies. These
policies have to be implemented by government and administrative structures
which are competent and not corrupt. This cannot happen when high level
government officials and bureaucrats are badly paid, with much lower salaries
than can be obtained in the most common private sector activity. This cannot
happen when civil servants who are called on to implement new types of rules
and regulations are not well trained and remain with their old mindset of con-
trol and red tape. This cannot happen when government officials are selected
and rewarded according to their loyalties to political patrons rather than their
competence and performance. These problems are pervasive in all Arab coun-
tries, and building and enhancing the governments’ capabilities to implement
reforms is critical in the coming period.
Concluding Comments
Reform in the Arab region has moved to a new path The “what” of policy
reform is well mapped, based on vast international experience. The “how” of
policy reform is a road much less traversed. The Arab region has a large eco-
nomic reform agenda ahead if it is to meet its employment and development

Part I: Growth, Reform, and Governance 101
challenges. Reforms initiated in the past have stalled, primarily the result of
deeply entrenched interests of those benefiting from the status quo, and lack
of information and inability to mobilize among those who would benefit
from reform. For the Arab reform agenda to proceed forward, it is clear that
the region must address critical governance issues that hinder these effective
coalitions for change to unite.
Notes
1. World Bank, 2003a, 2003b, 2003c.
2. World Bank, 2003c.
3. Arab labor force and employment statistics calculated for the period 2004-14.
4. Arab labor force and employment statistics computed for Algeria, Bahrain, Egypt, Jordan, Kuwait,
Lebanon, Morocco, Oman, Qatar, Saudi Arabia, Syria, Tunisia, United Arab Emirates, West Bank and
Gaza, and Yemen.
5. Inflation figures for Algeria, Bahrain, Egypt, Jordan, Morocco, Syria, Saudi Arabia, and Tunisia.
6. Weighted average of external debt/GNI for Algeria, Egypt, Jordan, Lebanon, Morocco, Syria,Tunisia,
Yemen, Kuwait, Oman, and Saudi Arabia.
7. World Bank, 2003a.
8. World Bank, 2003a.
9. Real growth of GDP for Egypt, Morocco, Tunisia, Jordan, Algeria, Syria, Yemen, Iran, Saudi Arabia,
Kuwait, Oman, Bahrain, and the United Arab Emirates.
10. Krueger, Anne, 1993.
11. Koromzay, Val, 2004.
12. Keefer, Philip (2004).
13. Nabli, Keller, and Véganzonès, 2004.
14. Measured at the 4-digit ISIC.
15. Regional averages weighted by GDP.
16. World Bank, 2003b.
17. According to the International Confederation of Free Trade Unions, the Middle East region
remains the most repressive in the world in terms of labor rights.
18. Alfred Tovias; 2000.
19. World Bank, 2003a.
20. Ibid, 2004.
21. Ibid, 2003c.
References
Keefer, Philip. 2004.“What Does Political Economy Tell Us About Economic Development—And Vice
Versa?” Policy Research Working Paper #3250. World Bank, Washington, DC.
Koromzay, Val. 2004. “Some Reflections on the Political Economy of Reform.” Presented at interna-
tional conference on Economic Reforms for Europe: Growth Opportunities in an Enlarged
European Union, Bratislava, Slovakia. March 18.
Krueger, Anne. 1993. Political Economy and Policy Reform in Developing Economies. Cambridge, MA:
MIT Press.

Breaking the Barriers to Higher Economic Growth102
Nabli, Mustapha, Jennifer Keller, and Marie-Ange Veganzones. 2004. “Exchange Rate Management
within the Middle East and North Africa Region: The Cost to Manufacturing Competitiveness.”
Lecture and Working Paper Series No. 1. American University of Beirut: Institute of Financial
Economics.
Tovias, Alfred. 2000. “The Political Economy of Partnership in Comparative Perspective.” Jerusalem,
Israel: The Hebrew University.
World Bank. 2004. Economic Reforms and Growth Experiences: Lessons from the 1990s. Washington, DC:
World Bank.
———. 2003a. Trade, Investment, and Development in the Middle East and North Africa: Engaging with
the World. MENA Development Report. Washington, DC: World Bank.
———. 2003b. Better Governance for Development in the Middle East and North Africa: Enhancing
Inclusiveness and Accountability. MENA Development Report. Washington, DC: World Bank.
———. 2003c. Unlocking the Employment Potential in the Middle East and North Africa: Toward a New
Social Contract. MENA Development Report. Washington, DC: World Bank.

103
Democracy for Better Governance
and Higher Economic Growth in
the MENA Region?
Mustapha K. Nabli
Carlos Silva-Jáuregui*
5
Democracy is valued in itself. The extent to which a citizen is able to live in an
open society and participate in its democratic process affects directly his or
her well-being. But democracy can also affect welfare indirectly, through its
effects on other aspects of the social and economic interactions that influence
the well-being of people.
Democracy can often positively affect the relative rights of social groups, such
as gender-specific groups or minorities. In the economic area, democracy may
affect the distribution of income, with democracies, for instance, tending to pay
higher wages and improve human capital. It may affect volatility of incomes,
with democracies tending to produce fewer recessions. Nobel Prize laureate
Amartya Sen observes that famines have never occurred under democratic
regimes (Sen 1999). Democracy may also affect the rate of economic growth.
To the extent that democratic development reinforces and is reinforced by
these various positive effects, democracy will generally gain more acceptance
and opposition to it will weaken. But what happens if there are trade-offs
between democratic development and any of these positive social and eco-
nomic effects? What happens in cases when a democratic process brings into
power a government that is able to pursue policies that undermine gender
*World Bank.
Published in the proceedings of the International Economic Association World Congress 2006,
Marrakesh, Morocco. Reprinted by permission from Palgrave MacMillan.

Breaking the Barriers to Higher Economic Growth104
equality, or the rights of some minority group? What happens in situations
where democratic development leads, for some reason, to a reduction in
incomes or a reduction in the rate of economic growth?
In these situations, individuals and society may still value democracy despite
the trade-offs. It is likely that most people today would go in such a direction.
Society may also introduce checks and balances, and develop institutional
mechanisms within the democratic process, which would reduce or eliminate
the likelihood of a democratic process producing such negative outcomes.
In view of the complexity of the issues related to democratic development,
the objective of this paper is to discuss only a limited topic: Does democracy
tend to induce higher or lower economic growth? The aim is to help under-
stand the links between democracy and economic growth. But it should be
made clear that findings that show that democracy leads to less growth would
not lead to any presumption that democracy should be sacrificed for the sake
of growth. Such a choice needs to be made by a society (through a democrat-
ic process, preferably!) based on its special circumstances. On the other hand,
a positive link reinforces the strength of arguments for democracy.
The paper will focus more specifically on the Middle East and North Africa
Region (MENA),1given the recent emphasis on democratic development
there. Actually, democracy has risen dramatically on the agenda for and in the
MENA region and countries. It has become an explicit objective of foreign
policy for the United States as well as the G8.Whether it is the primary objec-
tive, and whether it is being pursued effectively, are issues that are subject to
much heated debate. But there is no doubt that promoting democracy has
moved up high on the agenda of both the United States and the EU in the con-
text of its European Neighbourhood Policy. Likewise, if not more important,
there is increased domestic pressure for change from within the region. Civil
society at large has been demanding more political openness over the last few
years. This has been eloquently and forcefully expressed in the United Nations
Development Programme (UNDP) Arab Human Development Reports.
The reasons for the recent drive for democratic change are varied and com-
plex. For the foreign players, they may have to do with the possible or pre-
sumed links between the lack of democracy and “terrorism,”2or between
democracy and “security of borders.” For domestic actors, they may simply
have to do with the people of the region aspiring for more empowerment and
freedom after many decades, if not centuries, of political oppression. But the
paper is not going to delve into those issues.
The paper does not look into the determinants of democratic develop-
ment, either. For instance, there is a large literature and debate about whether
economic growth fosters democratization, as first advanced by Lipset (1959).
Most recently Friedman (2005) argues that over the long run a rising living
standard fosters openness, tolerance, and democracy. He recognizes, howev-
er,that in the short run, economic growth makes more secure whatever polit-

Part I: Growth, Reform, and Governance 105
ical structure may be in place; while economic stagnation and crisis may
undermine a nondemocratic regime. Also, in a recent review, de Mesquita
and Downs (2005) argue that, while economic growth results in higher
income and increases demand for democracy, it may also foster the ability of
autocratic regimes to strengthen their power, as they are able to shape insti-
tutions and political events to their advantage. Acemoglu et. al. (2005) show
that the strong cross-country correlation between income and democracy
does not mean there is causality,and that this correlation can be explained by
historical factors that jointly determine both economic and political develop-
ment paths of various societies. In the most ambitious analytical undertak-
ing to date, Acemoglu and Robinson (2006) develop a general framework
using game theory for understanding how democratic development takes
place and consolidates or not. This work shows that there are problems of
simultaneity between democratic development and economic development
that will be discussed when relevant. These issues go well beyond the scope
of this paper.
This paper considers the possible effects of a “given” democratic process,
without dwelling much on how it may have come about. Its scope is the rela-
tionship between “democratic development” and “economic growth.” From
an economic perspective, the objective is to determine whether there is a well-
defined relationship (or lack thereof) between the two, and more specifically
whether one should expect democracy to help or hinder a stronger economic
performance in general, and in the context of the MENA region in particular.
The next section provides the general context for the democracy–growth
linkages discussion in the MENA region—which has been characterized by
the existence of a democracy deficit, as well as a growth deficit, for the last two
decades. With this background, the third section reviews the empirical litera-
ture on the links between democracy and growth that focuses on direct links
and uses mostly reduced-form type models, and concludes that the nature
and strength of these links are at best ambiguous. More recent work on the
relationship between democracy and growth, surveyed in the fourth section,
pursues a more structural approach and looks at the intermediation channels
and indirect links between democracy and growth, as well as the role of the
nature of the democratic regimes. One main conclusion is that the effect of
democracy on growth, especially in MENA, depends to a large extent on
whether a democratic transition leads to better governance and, therefore, a
better business climate and higher private physical capital accumulation. The
fifth section explores the extent to which better governance is more likely to
be achieved under democratic regimes or nondemocratic ones. The sixth sec-
tion briefly discusses a different approach to looking at the links between
democracy and growth by postulating that, while such a link may not be
established systematically for any country at any time, it may be important for
most MENA countries today. In this case, democratic reform may be needed

Breaking the Barriers to Higher Economic Growth106
to unlock the prevailing status quo—low public accountability and the main-
tenance of prevailing economic policies and networks of privilege—in order
to generate a great political and economic transformationthat could produce
both more democracy and more economic growth in the region. The last sec-
tion concludes that democratic development can be a strong lever for eco-
nomic growth in the MENA region, and that these countries should strive for
democratic regimes that are sustainable, in the sense of having characteristics
that make them more likely to produce good governance. Good “quality
democracy,” which produces quality governance, improves the investment cli-
mate and allows the emergence of a dynamic private sector that can con-
tribute to meeting the current development challenges of the MENA region,
which needs to achieve average growth rates of 6-7 percent per year in order
to absorb the fast-growing, increasingly-educated, and feminized labor force.
The main focus should be on the design of the appropriate democratic insti-
tutionsthat (i) minimize imperfections in the political market, with more
freedom of information and free press, adequate mechanisms to contain
clientelism, and increased credibility of political promises; (ii) introduce safe-
guards and effective checks and balances; (iii) increase the legitimacy of the
democratic transformation; and (iv) in cases where there is significant ethnic
and or religious fragmentation, minimize the risks of social conflicts.
The Democracy and Growth Deficits in the MENA Region
At this point, it is perhaps useful to further explain the context of the discus-
sion of the links between democracy and economic growth in the MENA
region. This context can be summarized by the existence of both a “democra-
cy deficit” and a “growth deficit.” Their simultaneous presence in practically
all countries of the region (even though at different degrees) leads one to
wonder whether any links exist between the two.
Before presenting the evidence about the democracy deficit, it is useful to
provide some basic references about the definition of democracy.3At the most
abstract level, democracy is a system of government (or of exercise of author-
ity) in which effective political power is vested in the people, and where major
decisions of government, and the direction of policy behind these decisions,
rest directly or indirectly on the freely given consent of the majority of the
adults governed. At the more practical level, democracy tends to be defined in
procedural terms, as the body of rules and procedures that regulate the trans-
fer of political power and the free expression of disagreement at all levels of
political life. More concretely, it is defined as a political system where access to
political power is regularly achieved through competitive, free and fair elec-
tions. As stated by Schumpeter (1942), it is “…the institutional arrangement
for arriving at political decisions in which individuals acquire the power to
decide by means of a competitive struggle for the people’s vote.” (p.250).

Part I: Growth, Reform, and Governance 107
Democracy Deficit in the MENA Region
Democracy has gained worldwide acceptance in recent decades. Without
exception, all developed countries maintain democratic systems, and many
developing ones are selecting their leaders through competitive elections, that
is, moving toward more democratic political regimes. In a recent publication
on lessons of the 1990s, the World Bank (2005a) points out that “...a striking
phenomenon of the 1990s was the rise in the number of countries selecting
their leaders through competitive elections. The number rose from 60 coun-
tries in 1989 to 100 in 2000. Among poorer countries (those with less than the
median country’s per capita income), the number nearly tripled, from 11 in
1989 to 32 in 2000; 15 percent of the poorer countries elected their govern-
ments in 1989 and 42 percent in 2000.” This shows a remarkable move toward
democracy, but one that did not spread to MENA as vigorously.
There is now a wide body of evidence on the “democracy gap or deficit” in
the MENA region. It will suffice here to highlight some of this empirical
evidence.
A first piece of evidence can be observed when using the well-known com-
posite Polity index from the Polity IV dataset.4The composite Polity index
(which ranges from –10 for the least democratic regimes to +10 for the most
democratic regimes) shows that the MENA region has consistently lagged
behind the rest of the world, suggesting that there is a persistent democracy
deficit in the region (figure 5.1). The MENA democracy deficit has existed
over the last 40 years, with the average regime in MENA remaining authori-
tarian, according to this metric (negative values for the Polity index). While
the Organisation for Economic Co-operation and Development (OECD)
countries have been consistently democratic, other regions in the world were
traditionally not very democratic. The Polity index shows, however, a clear
tendency toward democratization in developing countries outside MENA
starting around 1977. Developing countries (other than MENA) have their
average Polity index turning positive by 1991, with a gain of 7 points (30 per-
cent of the scale) during the period 1977-2002.
As figure 5.2 shows during the 1960s, on average, the democracy level of
countries in MENA and other developing regions was declining. The democ-
racy gap between MENA and other regions remained, however, relatively sta-
ble, with a small declining trend during the period covering the mid-1960s
and the mid-1970s. The democracy gap reached its lowest point around 1977.
Since then, the gap has increased steadily, accelerating around 1990 as the
Soviet Bloc disintegrated and new democracies emerged, particularly in
Central and Eastern Europe. Since 1994, there has been a decline in the
democracy gap between MENA and other developing regions, but the gap still
remains significantly above the level attained in the 1970s. The democracy gap
between MENA and the OECD countries increased steadily until 1988, when
the average OECD Polity index reached its high plateau. At that point,the gap

Breaking the Barriers to Higher Economic Growth108
was almost 17.5 points, or 88 percent of the scale. Since then the gap between
MENA and the OECD countries has declined as gradual progress toward
democratization has taken place in MENA. Nonetheless, the gap remains at 15
points, or 75 percent of the scale, above the levels attained in the 1960s.
A second piece of evidence comes from analyzing the trends in the
Freedom House Political Rights index. This alternative measure of democra-
cy has been used by several scholars (Acemoglu and Robinson 2006; Barro
1999).
Figure 5.1. Democrac y Trends in MENA and Other Regions
⫺10
⫺8
⫺6
⫺4
⫺2
0
2
4
6
8
10
MENA
OECD
other developing
countries
mean polity index
1960
1962
1964
1966
1968
1970
1972
1974
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
Source: Mean Polity Index 1960-2003.
Figure 5.2. Democrac y GAP between MENA and Other Regions
20
18
16
14
12
10
8
6
4
2
0
OECD
other developing
countries
mean polity index gap
1960
1962
1964
1966
1968
1970
1972
1974
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
Source: Mean Polity Index 1960-2003.

Part I: Growth, Reform, and Governance 109
The original Freedom House index ranges from 1 to 7, with 1 representing
the most political freedom and 7 the least. We use the transformed Freedom
House index (Acemoglu and Robinson 2006), which, following Barro (1999),
is supplemented with data from Bollen (1990, 2001) for 1960 and 1965, nor-
malized and transformed to lie between 0 and 1, with 0 corresponding to the
least democratic set of institutions and 1 to the most.5This facilitates compar-
ison with the other indicators of democracy used in the paper.
Figure 5.3 plots the normalized Freedom House Political Rights index since
1950. The index confirms the low level of democracy in the MENA region, and
a growing democracy gap with OECD and other developing regions in the
world. As the figure shows, the MENA region has a declining trend in the politi-
cal rights index, losing ground particularly between 1960 and 1975, and between
1985 and 1990. The index reveals a small improvement in political rights in the
period 1975-1980, at the height of the oil boom. The index also indicates little, if
any, gain in political rights during the 1990-2001 period, contrary to the trends
of the Polity index. This may indicate that while some elements of democratiza-
tion were implemented, those related to political rights lagged.
An analysis of the democracy gap using the Freedom House index shows
an increasing gap vis-à-vis OECD countries until the early 1990s. With respect
to countries in other developing parts of the world, the democracy gap
declined marginally during the 1960-1980 period. Since 1985, however, it has
increased sharply, as other regions of the world have moved much more rap-
idly toward increasing political rights and advancing democratic reforms.
A third piece of evidence on the gap can be found in the work of
Papaioannou and Siouroumis (2004), who constructed a complete dataset on
Figure 5.3. Freedom House Political Rights Index in MENA and Selected Regions
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
mean Freedom House Political Rights index
1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2001
OECD
MENA
other developing
countries
Source: Transformed Freedom House Index 1950-2001.

Breaking the Barriers to Higher Economic Growth110
democratization in the world since the 1970s. The data used are wide-ranging,
go beyond the various quantitative indicators, and use historical review of
episodes. The authors find that 38 episodes of “full democratization” and 24
episodes of “partial democratization” occurred in the world over the last 30-
40 years. None of these episodes took place in the MENA region.
A fourth piece of evidence on the “democracy deficit” is the absence of a
positive correlation between democracy and level of income in the MENA
region, unlike what is typically found for the rest of the world. Przeworski et
al. (2000) found that democracies are more frequent in developed (wealthier)
countries, while dictatorships are more frequent in poor countries. In typical
charts showing the correlation between level of income and democratic devel-
opment, MENA counties tend to cluster way below the line. In particular, the
many oil producers tend to be less democratic than less wealthy countries. The
Gulf Cooperation Council (GCC) countries have among the lowest scores in
the region using the Polity IV index.
Driving the gap between MENA countries and the rest of the world are
striking weaknesses in external accountabilities and in access to basic political
and civic rights. The World Bank (2003e) constructed an index of public
accountability (IPA). The IPA assesses the process of selecting and replacing
those in authority. It measures the quality of governance according to the
inclusiveness of access to basic political and civic rights, and the relative
strength of external accountability mechanisms. It aggregates 12 indicators
that measure the level of openness of political institutions in a country, and
the extent to which political participation is free, fair, and competitive; civil
liberties are assumed and respected; and the press and voice are free from con-
trol, violation, harassment, and censorship. It also captures the transparency
and responsiveness of the government to its people, and the degree of politi-
cal accountability in the public sphere.
All countries in the MENA region, whatever their income, score well below
the world trend in the IPA (figure 5.4). Some richer MENA countries score
especially low on the IPA—with scores equivalent to those in some of the
poorest countries of the world. Oil seems to matter, as oil-exporting countries
have the worst IPA scores.
Economic growth deficit in the MENA world
The recent history of economic growth in the MENA region can be easily
understood . MENA’s historical model of economic development was based
on state-led development and central planning, economic and social policies
designed for redistribution and equity, and a strong social contract between
governments and the people they represented.
During 1965–85, economic growth per capita averaged 2.9 percent per
year, second only to the East Asia and Pacific region (figure 5.5). Many factors
contributed to this performance, including rapid progress in early-stage

Part I: Growth, Reform, and Governance 111
industrialization; high levels of public employment and spending, especially
on infrastructure; trade protection for domestic producers; and rising oil
prices that yielded large capital inflows, created jobs, and promoted remit-
tance flows into non-oil-producing MENA states.
While this development model paid large dividends in the beginning, par-
ticularly with regard to improvements in social indicators, there were also sig-
Figure 5.4. Public Accountability and per Capita Incomes in MENA
world median
average MENA gap
in public accountability
tted line
for MENA
MENA
rest of the world
tted line for the
rest of the world
index of public accountability
log of per capita GDP
Figure 5.5. Average per Capita GDP Growth, 1965-2004
⫺1
0
1
2
3
4
5
6
7
8
East Asia
and Pacic
Latin America
and the
Caribbean
South Asia Sub-Saharan
Africa
MENA MENA
(oil)
MENA
(nonoil)
percent
1965–85
1986–2000
2001–04
Source: World Development Indicators 2005.
Note: MENA (oil producers) includes Algeria, Bahrain (1986–2004), Iran (1986–2004), Kuwait, Oman, Saudi Arabia, and UAE (1986–2004). MENA
(nonoil producers) includes Egypt, Jordan, Lebanon (1986–2004), Morocco, Syria, Tunisia, and Yemen (1986–2004).

Breaking the Barriers to Higher Economic Growth112
nificant costs. Centralized and hierarchical governments emerged in MENA,
with limited transparency and contestability of representatives or policies.
The MENA development model also created economies that had great diffi-
culty adapting to shocks and economic change.
As early as the late 1970s, the economic systems that had developed in
MENA—and that carried the people in the region through an unprecedented
era of achievements—showed signs of cracking under stress. The high growth
rates were becoming increasingly costly to achieve. Though investments were
at record levels, with the rate of growth of physical capital per worker increas-
ing by more than 80 percent in the 1970s over the 1960s, these investments
were inefficient, delivering increasingly smaller growth payoffs. As a result of
large inefficiencies, total factor productivity growth was lower than in any
other region of the world, and turned negative during the 1970s.
The MENA region entered the 1980s with mounting evidence of strains
and difficulties in sustaining the promise of continued prosperity. Facing
declining public revenues after the oil shock in the mid-1980s, govern-
ments struggled to maintain their growth performance and redistributive
commitments. With a public sector wage bill accounting for as much as 20
percent of gross domestic product (GDP), deficits mounted and debt grew
at an alarming rate. The fiscal strains contributed to large macroeconomic
imbalances. Productivity growth, already declining by the 1970s, plummet-
ed to –1.5 percent a year on average over the 1980s. Real output growth col-
lapsed under the multiple blows of declining public spending, in part a
result of the negative oil shocks, an unattractive private investment climate,
and continuing losses in efficiency. GDP per capita stagnated over the
1980s, growing an almost imperceptible 0.3 percent a year during the
decade.
In the 1990s, several macroeconomic stabilization reforms were imple-
mented, and they began to pay important dividends. By and large, MENA
countries recovered from the instability of the 1980s. Inflation was brought
under control, debt levels declined, and macroeconomic performance turned
positive on average. These were fundamental preconditions for higher private
investment and growth, but the strong growth rates needed to cope with the
demographic transition in the region failed to materialize. Despite the men-
tioned reforms, the effort did not translate into the strong economic recovery
that was anticipated. Though GDP growth improved compared to the crisis-
ridden 1980s, per capita growth remained weak, averaging 1.5 percent per
year in the 1990s. While the declines in productivity growth were arrested,
productivity growth was close to 2 percentage points lower than the world
average and 3.5 percentage points lower than East Asia—at about the same
levels of investment.
With the coming of the new millennium, the region has experienced a new
set of favorable conditions. As a result, the region has achieved exceptional

Part I: Growth, Reform, and Governance 113
growth over the last few years. Accelerating in the early 2000s, economic
growth in the MENA region (excluding Iraq) averaged 5.1 percent a year over
2002-04, the strongest growth rate in a decade, and significantly higher than
the average yearly growth during the 1990s. On a per capita basis, the MENA
region’s 3.2 percent average growth over 2002-04 was its strongest growth per-
formance since the mid-1970s.
Despite the oil-driven growth boom, on a per capita basis the region’s
growth rate over the last few years continues to lag that of most other
regions, a reflection of both the firming of GDP growth rates across devel-
oping regions and the MENA region’s still-high population growth rate,
which continues to be a key development challenge.At the regional level, per
capita growth in East Asia and the Pacific, South Asia, and Europe and
Central Asia all outpaced MENA’s per capita GDP performance in both 2003
and 2004.
In sum, over the past two decades, the MENA region has experienced a
growth deficit, with low per capita income growth. This growth performance
has been weaker than that achieved by most other regions of the world, except
for Sub-Saharan Africa.
Empirical Correlation and Direct Links between Democracy and
Economic Growth
The previous discussion leads one to be tempted to hypothesize that there is
a strong link between the low growth and the democracy deficit in the MENA
region. This section more thoroughly reviews international experience
regarding the empirical relationship between democracy and growth, and
investigates how it applies for the MENA region. International experience,
such as the economic success of many authoritarian regimes, including
Singapore, South Korea, and Indonesia during the 1970s and 1980s, Chile in
the 1980s, and China over the last 20 years,casts some doubt on the existence
of any robust (positive) linear relationship between economic growth and
democracy. Empirical studies using standard growth regression models and
cross-country data have found mixed evidence for such direct links. Reviews
of a large number of these studies (see Borner et al. 1995) found that only
very few show any strong positive relationship, with most showing either
insignificant results or even a negative link. Using the Freedom House indi-
cator for democracy (electoral rights), Barro and Sala-i-Martin (2003) show
that a nonlinear relationship may exist, similar to a Laffer Curve, in which
democratization appears to enhance growth for countries that are not very
democratic, but to retard growth for countries that have already achieved a
high degree of democracy. Przeworski et al. (2000) analyzed the data over a
long period of time and concluded that when countries are observed across
the entire spectrum of conditions, total income grew at about the same rate

Breaking the Barriers to Higher Economic Growth114
for democratic and undemocratic regimes. However, they found that pat-
terns of growth varied between democracies and nondemocracies, particu-
larly in wealthy countries.
The history of the MENA region over the last 50 years shows limited expe-
rience with democracy and political openness. But while no country in the
region has achieved a full transition to democracy, the degree of political
openness, and the use of the election process to choose the government in
power, have varied significantly over time and across countries. The high
income growth rates achieved in the 1970s and early 1980s were not accom-
panied by any significant degree of political openness, while the growth col-
lapse of the 1980s saw some degree of political liberalization in several MENA
countries. During the 1990s, the experiences of low economic growth coun-
tries like Morocco were associated with gains in democratization, while
higher-growth countries like Tunisia were increasingly autocratic. At the same
time, Jordan experienced higher growth and political openness, while low
growth and limited democratization were the norm in most Gulf countries.
Overall, there was little correlation between changes in democracy and
changes in income per capita during 1970-2003 in MENA and the rest of the
world (figure 5.6). Countries that grow faster than others have not become
more democratic.
The lack of strong empirical evidence of direct positive links between
democracy and economic growth, which seems to apply for the MENA region
as well, has led research to move in new directions for the study of the link-
ages between democratic and economic development. We look at some of
them in the next section.
Figure 5.6. Democrac y Growth and Income Growth, 1970–2003
1.0
0.8
0.6
0.4
0.2
0
⫺0.2
⫺0.4
⫺0.6
⫺0.8
⫺1.0
⫺0.5 ⫺0.25 0 0.25 0.5 0.75 1.
0
Tunisia
Syria
Jordan
Algeria
Iran
Morocco Egypt
change in polity index
change in log (GDP per capita, PPP)

Part I: Growth, Reform, and Governance 115
Structural Analysis of the Links between Democracy and Economic Growth
Indirect Links from Democracy to Economic Growth
Recent research has gone in three directions of study for the indirect linkages
between democratization and economic growth.6
The first direction explores more carefully the theoretical and empirical
links between democracy and economic growth, going beyond simple corre-
lations and cross-country regressions, which considered only direct links.
These studies look at the indirect effects of democracy on economic growth
through a number of intermediation channels. They use structural models
that involve income growth, democracy, and variables that represent such
intermediation or transmission channels. Some of these channels show a neg-
ative impact of democracy on economic growth, whileothers show a positive
link. A focus on the inefficiencies of representative government, using the
“median voter model” or the public choice approach, would imply that dem-
ocratic regimes result in larger demand for redistribution and prevalence of
special interest politics. This may result in higher government consumption,
higher taxation, more redistribution, and lower private investment. All these
channels would imply lower growth rates.
But many other channels would lead one to expect a positive impact from
democracy on economic growth. First, redistribution and higher taxation may
result in higher human capital accumulation, through larger subsidies and
dealing with capital market failures. Second, democracy deals better with eco-
nomic instability, through better commitment achieved through the political
process. Political instability is part of everyday life in democracies and does
not affect economic growth as much as in the case of authoritarian regimes.
In nondemocracies, any changes—or expectations of changes—in leadership
negatively affect investment and growth: Whenever dictators are expected to
be removed, growth declines sharply (Przeworwski at al. 2000). Such nonde-
mocratic regimes are successful only if they are stable. Shocks therefore have
large negative impacts on economic growth. In addition, democratic regimes
are better suited for both mediating conflicts among interest groups and
responding to exogenous negative shocks. Countries with higher degrees of
social and ethnic fragmentation and weak democratic institutions are found
to have suffered the sharpest drops in GDP after shocks (Rodrik 1997, 1999).
Tavares and Wacziarg used a full system of simultaneous equations and
panel data for the period 1970-89. They found that democracy fosters growth
because it improves human capital accumulation and, in a weaker way,
because it reduces income inequality. At the same time, they find that democ-
racy hinders growth by reducing physical capital accumulation, and, in a less
robust way, by increasing the government consumption to GDP ratio.
However, no significant impact is found through the channels of political
instability and policy distortions. The overall effect of democracy on growth

Breaking the Barriers to Higher Economic Growth116
is slightly negative, mainly through the large impact on the reduction in the
rate of physical capital accumulation. In a similar vein, Feng (2003) conducts
a wide-ranging empirical study of the impact of political institutions on eco-
nomic growth. He finds that democracy has an insignificant direct effect on
economic growth, but indirect effects that are strong and significant. These
indirect channels include political instability, policy uncertainty, investment,
education, property rights, and birth rates.
The second direction goes beyond cross-country analysis and intensively
uses event analysis and differences in performance for before and after
democratization episodes. Empirical findings from this analysis tend to show
a positive impact of democratization on a given country’s economic growth.
From a theoretical standpoint, the evidence presented in this new branch of
the democracy-growth nexus literature offers direct support for so-called
development theories of democracy and growth that highlight the growth-
enhancement aspects of the democratic process. From a policy perspective,
the results suggest that democratic institutions, if properly introduced and
adapted, can bring substantial growth benefits. They also suggest an impor-
tant role for the international community—to help mitigate the transition
costs, which can be high and can impede the consolidation of democratic rule.
To assess whether a successful democratic transition is associated with
faster growth, Papaioannou and Siourounis (2004) first identify countries and
the exact timing of permanent democratizations in the period 1960-2000.
They employ an event study approach and analyze the evolution of GDP
growth before and after such incidents of political modernization. Using a
dynamic panel with annual observations, and econometric techniques that
address concerns on shortcomings of previous research, the study reveals that
conditional on various growth determinants, global shocks, and business cycle
effects, permanent democratization is associated with approximately a 1 per-
cent increment in real per capita growth. The analysis also reveals a J-shaped
growth pattern. This implies that output growth drops during the democrat-
ic transition, but then fluctuates at a higher rate, suggesting a “short-run
pain,” resulting perhaps from to high transition costs and learning, followed
by “long-run gain,” resulting from higher growth after the consolidation of
democracy.The effect is robust across various model specifications, panel data
methodologies, alternative democratization dates, and the potential endo-
geneity of democratization. The methodology enables quantification of both
the short- and the long-run correlations between political modernization and
growth. The results favor the Aristotelian notion, recently put another way by
Friedrich Hayek, that the merits of democracy will come in the long run.7
That is, stable democracies can foster growth.
The Papaioannou and Siourounis work is related to a new wave of research
that studies the effect of institutions on economic performance.8Their results
suggest that, besides legal norms or property rights protection, the type and

Part I: Growth, Reform, and Governance 117
quality of political institutions correlate substantially with economic growth.
The overall effect of democracy on growth is then positive in these studies.
The third direction of the literature goes beyond the general dualistic spec-
ification of political regimes as democratic or nondemocratic and explores a
number of dimensions. A significant amount of work, mostly on advanced
countries, looks at how the nature and rules of democratic regimes affect out-
comes. Whether these regimes are presidential or parliamentarian, whether
they use majoritarian or proportional representation, does matter for the way
democracy affects economic outcomes.
In recent parallel studies, Persson (2004) shows that income gains follow-
ing democratization are high when the transition leads to proportional repre-
sentation (versus majoritarian) or when it leads to a parliamentary (versus
presidential) system. Giavazzi and Tabellini (2004) document significant
interactions between economic and political liberalization and show that
countries experience substantial growth gains when they liberalize the econo-
my first, and then the polity.
For developing countries, the focus has been on how imperfections in elec-
toral markets tend to make democracy less effective in achieving good govern-
ment than it is in advanced countries (World Bank 2005a). Imperfections in
electoral markets—lack of voter information, the inability of political com-
petitors to make credible promises, and social polarization—are important
factors in understanding policy formulation and explaining differences in
economic performance between rich and poor democracies. Voters in devel-
oping countries tend to be less informed, the role of media weaker, and cam-
paign financing more prone to capture, which result in worse governance out-
comes. At the same time, politicians tend to be less credible, and clientelism
more pervasive, especially as the length of exposure to elections tends to be
shorter. In addition, social polarization and ethnic fragmentation distort elec-
toral processes.
The empirical findings from this strand of the literature tend to condition
the possible impact of democratization on growth on how severe the political
market imperfections are. Differences in economic performance across
democracies can be explained by imperfections in electoral markets.
Numerous imperfections in these markets make it difficult for citizens to hold
politicians accountable for policies. Elected governments are most likely to
make policies favouring narrow segments of the population, at the expense of
the majority, when citizens do not have good information mechanisms, can-
not trust promises made before elections, or are in societies that are deeply
polarized. These factors are three of the most important political market
imperfections that affect policy outcomes. In contrast, elected governments
are most credible, and most likely to respect private property rights, when
they confront checks and balances on their decision making. Thus, accounta-
bility becomes an essential component.

Breaking the Barriers to Higher Economic Growth118
Informed voters are necessary for good political outcomes. Without infor-
mation about what politicians are doing, and how their policies affect citizens’
welfare, or about the attributes of political competitors, citizens cannot easily
identify and reward high-performing politicians. As a result, bad performance
is encouraged and bad political outcomes are likely to occur. In political mar-
kets, information about the characteristics of political competitors and gov-
ernment performance is key. Proxies like newspaper circulation are common-
ly used in empirical analysis for voter information, and reveal, controlling for
income and other factors, that higher newspaper circulation is associated with
lower corruption, greater rule of law, better bureaucratic quality, and greater
secondary school enrolment (Keefer and Khemani 2005).
Credible commitments by politicians are also important for good gover-
nance outcomes. When challengers cannot make credible policy commit-
ments, citizens have no reason to prefer them over incumbents. Even if
incumbents do badly, citizens have no reason to believe that challengers will
do better. This insulates incumbents from competition and diminishes pres-
sure to perform. Politicians may only be able to make credible promises to
some voters, generating clientelism and incentives to politicians to underpro-
vide public goods and extract large rents.
Likewise, social polarization hinders the capacity for political systems to
generate good outcomes. Social polarization undermines the accountability
of government to citizens. In extreme cases, deep divisions among social
groups hinder the capacity of one group to elect a representative of the
other, irrespective of his or her characteristics, political platform, and qual-
ities as a representative. Elected representatives from one group then have
no incentive to address the concerns and solve the problems of citizens in
the other, generating distortions in the provision of public goods. Empirical
studies show that measures of ethnic tension are higher in poorer democra-
cies that in rich ones. The consequences of social polarization can be wors-
ened by all the factors that undermine voters’ ability to hold politicians
accountable.
Implications for MENA Countries?
What can be concluded for MENA countries from the previous brief survey?
Does it mean that democratic reform is unimportant for economic growth?
Since the survey shows a lot of uncertainty about the causal links from
democracy to economic growth, MENA countries should not expect that the
pursuit of democratization can in itself bring quick benefits in terms of eco-
nomic growth. One can be tempted to conclude that the search for higher eco-
nomic growth should focus more on traditional policy and institutional
reforms within an existing political regime. In that case, striving for demo-
cratic development would be a different track, which should remain separate
from the track of economic growth.

Part I: Growth, Reform, and Governance 119
But such a conclusion is premature. In fact, the previous discussion sug-
gests that there are some robust positive effects of democracy on growth,
through higher human capital accumulation. But is such a link likely to be rel-
evant and significant in MENA countries? There is a lot of empirical evidence
that MENA countries have achieved strong gains in terms of human develop-
ment during the last four to five decades—in the presence of nondemocratic
regimes (World Bank 2006). Some of the most impressive improvements in
human development indicators took place in the MENA region between 1960
and 2000, surpassing the performance of countries in other regions with sim-
ilar purchasing power parity (PPP) income levels. For instance, as figure 5.7
shows, average years of education in MENA increased by more than 500 per-
cent, and in the case of women by more than 800 percent, during the 1960-
2000 period. Child mortality decreased from an average of 262 deaths per
1,000 births to an average of 47 deaths per 1,000 births during the same peri-
od. Life expectancy at birth improved from 47 to 68 years, a 45 percent
increase.
Actually, one can argue that regimes in MENA did strive to buy the loyalty
of citizens through strong redistributive programs such as free access to edu-
cation and health care, public sector jobs, and lower prices for basic com-
modities. It is unlikely that the channel of increased investment in education
will be important for increasing growth in MENA countries over the coming
period. On the other hand, there is general recognition of the need for educa-
tion reform, to improve the quality and adequacy of education in response to
changing economic conditions. Governance mechanisms may turn out to be
critical to the success of these reforms.
The previous analysis shows also that there are some robust positive effects
of democracy on growth, through better commitments to policies with a
more credible and predictable political process, and through better interme-
diation of conflicts. Such effects are likely to be relevant in many MENA coun-
tries and situations.
Figure 5.7. Human Development Indicators in MENA Improved Considerably
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
1960 1965 1970 1975 1980 1985 1990 1995 2000
average years of education
per person over 15
deaths per 1,000 live births
MENA10
comparators
0
50
100
150
200
250
300
1960 1965 1970 1975 1980 1985 1990 1995 2000
MENA10
comparators
Source: World Bank (2006). MENA10 refers to Algeria, Egypt, Iran, Jordan, Lebanon, Libya, Morocco, Syria, Tunisia and Yemen. Comparators include
32 countries with similar income levels of MENA countries in PPP terms in 1980.

Breaking the Barriers to Higher Economic Growth120
Another conclusion from this review is that for a number of other links
from democracy to economic growth, the effects are largely contingent on
whether a move toward more democratic government leads to better gover-
nance. In particular, there is uncertainty about the impact of democracy on
capital accumulation and policy distortions. This uncertainty is most likely
related to the extent to which democracy leads to more or less respect for
property rights, to policies that favor broader or narrower interests, and to
adequate or underprovision public goods; which in turn depends on the
strength of politicians’ credibility and the pervasiveness of clientelism.
The main general conclusion is that the net impact of democracy on
growth depends on the severity of political market imperfections—in other
words, whether democracy produces good governance and thereby enhances
physical capital accumulation. This conclusion is strengthened by two addi-
tional and complementary findings from the literature on governance (with-
out reference to democracy), which show direct, strong,positive links between
good governance and economic growth, and between better governance and
higher private investment. The latter linkage is particularly important for the
MENA region.
Governance and Growth
The work of Buchanan, et al. (1980), Evans (1989, 1995), North (1981, 1990),
and de Soto (1989) systematically linked country characteristics such as the
security of property rights directly to the wealth of nations, improving the
understanding of the effects of the nonpolicy characteristics of government per-
formance on economic development and growth. The research stemming from
this strand of the literature brought to the forefront the contribution of previ-
ously under-examined issues like the security of property rights, the rule of law,
expropriation, bureaucratic quality, red tape, and the quality of regulation.
Both empirical evidence and theory support the influence of individual
components of governance, rather than aggregated concepts of governance, on
development and growth. Studies have found that some governance compo-
nents have stronger links than others (Keffer 2004b). The security of proper-
ty rights, the credibility of governments (see Knack and Keefer 1995,
Acemoglu and Johnson 2005, Acemoglu, et al. 2001, Rodrik, et al. 2002, Hall
and Jones 1999), and an honest and efficient bureaucracy emerge as the com-
ponents with the best documented and strongest links to economic develop-
ment and growth. On the other side, causality problems cloud estimates of the
influence of bureaucratic capacity and corruption on development. In addi-
tion, analyses of voice and accountability, while the subject of substantial
attention among researchers, have suffered from a lack of theoretical and
empirical precision, which clouds interpretation.
The theoretical case for secure property rights is based on the basic idea
that economic agents do not invest when the business environment is not

Part I: Growth, Reform, and Governance 121
right and they fear potential confiscation of their assets by government. Still,
there are objections to the theoretical case linking the security of property
rights to growth.9One relates to the fact that often, two important notions of
property rights are confused: the allocation of property rights and the securi-
ty of property rights. Democracy may render property rights less secure
because the introduction of democracy creates opportunities for the poor to
redistribute incomes away from the rich. However, it is not democracy, per se,
that creates insecurity, but the transition to democracy. Once democracy is
established, there is no reason to expect the distribution of property rights to
change further.
Some contributions to the governance literature refer not only to the secu-
rity of property rights, but also to the closely related (but somewhat broader)
concept of “government credibility.” This is the case with Knack and Keefer
(1995), for example. Only credible governments can assure investors that their
assets are safe from expropriation.
In the context of MENA, some empirical evidence shows that better gover-
nance may have a significant positive effect on economic growth. Weaker gov-
ernance in MENA costs 1.0 to 1.5 percentage points in forgone annual GDP
growth (World Bank 2003e). On average, improving the quality of institutions
by one standard deviation—approximately equivalent to raising the average
institutional quality in MENA to the average institutional quality of compara-
ble East Asian countries (Indonesia, Malaysia, the Philippines, Singapore,
Thailand, and Vietnam)—would have resulted in an increase of almost 1 per-
cent in average annual GDP growth for the region as a whole. This figure would
imply an income level that would have been 50 percent higher in a period of 40
years due to compounding. The gain in growth rate from better governance
would have risen to a 1.5 percent difference for the MENA countries with sub-
stantial oil and gas revenues, implying an income level 81 percent higher in the
same comparative period. Similar results have been found by Elbadawi (2002).
Governance and Private Investment
The influence of the quality of governance on growth works primarily
through its effect on private businesses and their capacity to invest. Numerous
studies (see World Bank 2003e) have documented the relationship between
governance and private sector activity. Businesses react to the incentives, costs,
and constraints that form their business environment. Those reactions are, in
turn, influenced by the shaping and implementing of public policies.
Improved governance can produce better business climates that foster
investment, productivity, and growth. It reduces the scope for arbitrary gov-
ernment policymaking, providing mechanisms that help countries minimize
the persistence of policy distortions. By ensuring public accountability of
politicians and bureaucrats, better governance also contributes to the effective
implementation of economic policies that are conducive to growth.

Breaking the Barriers to Higher Economic Growth122
Better governance also improves bureaucratic performance and pre-
dictability, reducing uncertainty and the costs of doing business. This
enhances the business environment. Better governance makes it easier to start
new businesses and to run and expand existing ones. It lowers transaction
costs at all levels (entry, operation, and exit), reduces information asymme-
tries between business and governments, and lowers uncertainties and unpre-
dictability. It does so by protecting and enforcing property rights, curbing
burdensome administrative and judicial rulings, reducing red tape, ensuring
good regulatory quality, and improving access to affordable and reliable
recourse to dispute resolution. By helping ensure more orderly public
accountability processes, better governance also reduces political risk.
Better governance contributes to the effective delivery of public goods that
are necessary for productive businesses. Firms operate in a commercial envi-
ronment that depends on many key public goods. Better governance helps
ensure that such goods are available in a timely, equitable, and cost-efficient
manner. Public goods that are essential for a good business environment
include appropriately regulated public utilities and natural monopolies, a sta-
ble and prudently regulated financial system, public safety and low crime, and
good-quality health and education. Effective delivery of these and other pub-
lic goods boosts the productivity of private investment and leads to faster
growth and development.
A recent empirical study (Aysan, Nabli and Véganzonès-Varoudakis 2006)
provides a quantification of the possible impact of better governance on pri-
vate investment in a few MENA countries (Egypt, Iran, Morocco, Tunisia; fig-
ure 5.8). The study uses panel data and a simultaneous equation model for
private investment and governance indicators to estimate this impact. They
find that improved governance indicators, in terms of quality of public
administration and public accountability, equivalent to one standard devia-
tion of observed variability, would yield about a 3.5 percent increase in private
Figure 5.8. Private Investment in the MENA Region Remains Low
0
2
4
6
8
10
12
14
16
18
20
Egypt Iran Morocco Tunisia
percent
1980
1990
Source: Aysan, Nabli, and Véganzonès-Varoudakis (2006).

Part I: Growth, Reform, and Governance 123
investment-to-GDP ratios. For MENA countries, this would have increased
the ratio of private investment to GDP on average from 12 percent to 15.5 -
percent—a significant impact.
Better Governance Is Critical—But How?
Governance is among the more elastic concepts in the social science and
development lexicons (Keefer 2004a).Definitions tend to include the extent to
which governments are responsive to citizens and provide them with certain
core services, such as secure property rights and rule of law, and the extent to
which public sector institutions and processes give government decision mak-
ers an incentive to be responsive to citizens.
Good governance requires inclusiveness, transparency, and accountability
in the management of public affairs. The governance challenge in the MENA
region is to strengthen the incentives, mechanisms, and capacities for more
accountable and inclusive public institutions, and to expand allegiance to
equality and participation throughout society. Those good governance mech-
anisms are first steps in improving economic policies that are themselves
instruments for better economic growth.
The previous discussion makes a strong argument that democracy can have
a positive impact on economic growth if it leads to better governance. But is
democracy likely to generate good governance—and are there alternative
paths to achieving good governance?
From Democracy to Better Governance?
Researchers have developed various arguments that link democracy to both
greater and lesser quality of governance. The introduction of democratic
institutions, in the form of more political and civil rights, and freedom of the
press, among others, may or may not be associated with improved gover-
nance. On the one hand, democracies allow populations to peacefully and reg-
ularly oust inept, inefficient, and corrupt public administration and to retain
more efficient, successful regimes, thus tending to make the quality of gover-
nance, on average, higher in the long run. On the other hand, a number of
studies have noted that the proliferation of interest groups lobbying for power
or rents under democratic institutions may lead to policy gridlock, pervasive
clientelism, and lack of accountability. This could hinder growth prospects.
The empirical literature appears to confirm that stronger democratic insti-
tutions are positively associated with a higher quality of governance, as well as
with its positive impact on growth (Rivera-Batiz 2002, Keefer 2004a). Stronger
democratic institutions influence governance by constraining the actions of
corrupt and inept officials. They also facilitate the activities of the press, which
can monitor corruption and disseminate information on government officials
to the public, so that they can be held accountable.

Breaking the Barriers to Higher Economic Growth124
But research and experience have shown many situations where democra-
cy is unlikely to produce better governance. For instance, it has been often
observed that younger democracies are more corrupt, exhibit less rule of law
and lower levels of bureaucratic quality. These young democracies spend
more on public investment and government workers. The politicians are less
credible (Keefer 2005),and the inability of political competitors to make cred-
ible promises to citizens leads them to underprovide public goods, overpro-
vide transfers to narrow groups of voters, and engage in excessive rent seek-
ing. Young democracies seem to require time to mature to generate the desir-
able results.
Democracy may also not lead to better governance and higher growth in
MENA, because of the region’s social cleavages and fragmentation, as dis-
cussed above in the context of the importance of political market
imperfections.
The literature on these issues has shown that getting democracy right is
indeed affected by the extent of social cleavages. Elbadawi (2004) tests for the
impact of social cleavages on growth. In his model, social cleavages are meas-
ured by indexes of (ethnic, cultural, or religious) fractionalization and polar-
ization. He finds that several MENA countries have highly ethnically and reli-
giously fractionalized societies, including Djibouti, Jordan, Lebanon., Iraq,
and Syria. Like other authors before (Alesina et al. 2003, Easterly and Levin
1997, and Rodrick 1998), Elbadawi finds that the conflict variable that meas-
ures the social cleavages in his model has a negative effect on growth. The con-
flict variable was highly significant in the case of ethnic fractionalization, and
moderately significant in cases of language, religious, and dominant polariza-
tions In all cases, it was negatively associated with long-term stability of
growth. The results are much less robust in the case of the polarization-based
conflict. However, ethnic and, especially, language polarizations were nega-
tively and significantly associated with long-term growth stability.
This analysis indicates that relatively socially homogenous societies in
MENA (Egypt and Tunisia) may be better suited to get good governance
through democracy, while less homogenous societies in MENA (Iraq,
Lebanon, Syria) may find it more difficult to achieve the needed socio-
political consensus for good governance and good economic policy.
Should the Drive Be Just for “Better Governance”?
If there are risks that democracy does not lead to better governance, that is,
more transparency, better voice and accountability, more secure property
rights, less corruption, more efficient civil service, and more effective public
service delivery, then what are the implications for the MENA region?
One possible implication may be that MENA countries should strive to
achieve “better governance”—and not bother with seeking democracy.
Countries should try to develop features of good governance such as secure

Part I: Growth, Reform, and Governance 125
property rights, rule of law, efficient and less corrupt government and public
administration, predictable rules and laws, and so on. The economic success
of authoritarian regimes, particularly in Asia, like China, Taiwan, Singapore,
and South Korea, and even in some MENA countries, such as Tunisia and
Egypt, that achieved relatively high rates of growth over the long run, suggests
that this is possible, and that some adequate quality of governance can be
secured under such regimes. Some even raise the question whether the very
authoritarian discretion of such regimes actually helped growth, by enabling
the leading parties to push through economic reforms that in a democratic
setting might have been either slower, because of a need for consensus, or
impossible to achieve. This argument would be even more compelling in sit-
uations where there is a high likelihood that democracy would not generate
better governance because of significant social cleavages.
Many argue that this is not the case, and that democracy need not to be sac-
rificed on the altar of development. While East Asian countries have pros-
pered under authoritarianism, many more countries have seen their
economies deteriorate as a result of lack of democracy and accountability, for
example, Zaire, Uganda, and Haiti. Such cases abound in MENA, with the
examples of Iraq, Libya, and Syria. In addition, some of the most successful
economic reforms of the 1980s and early 1990s were implemented under
newly elected democratic governments in many regions, for instance, Latin
America Countries (LAC) (Bolivia, Argentina, and Brazil), and Europe and
Central Asia (ECA) (transition economies like Hungary, the Czech Republic,
Poland, Slovenia, and Slovakia among others).
The critical question is whether “better governance” is more likely to
emerge under democratic or nondemocratic regimes. Some of the previous
discussion indicates that democracy leads to better governance, despite the
risks faced by developing young democracies and countries with significant
social cleavages. What about nondemocratic regimes? Are they able to pro-
duce governance systems that can enhance the quality of the business environ-
ment, leading to higher investment and sustained economic growth?
Rodrik (1999) and Rodrik and Wacziarg (2005) have argued that the per-
formance record for democracies is even better than usually acknowledged.
Claims that democratization leads to disappointing economic results are often
used to justify calls to delay political reforms in poor, ethnically divided coun-
tries until they become “mature enough” for democracy. However, the
hypothesis that democratization is followed by bad economic performance,
particularly in poor, fractionalized countries, is not supported by their
analysis.
In any case, the various cases of “enlightened dictatorship” appear to be the
exception rather than the rule in the recent past. Authoritarian regimes may
provide high-quality governance only randomly, and for each case of a non-
democratic regime that seems to produce better governance and growth out-

Breaking the Barriers to Higher Economic Growth126
comes, one can find many more cases of bad governance and dismal econom-
ic outcomes. In probabilistic terms, democracies appear to be more likely to
generate better governance than nondemocratic regimes.
The Case of Oil-Producing Countries
Oil rents have shielded many MENA countries from economic crises, but they
seem to have also helped to reduce the likelihood of the region becoming
more democratic (Ross 2001). At the same time, the economic record of
mineral-exporting countries over the past few decades, especially oil
exporters, has been disappointing. Studies like Eifert et al. (2003) elaborate
that this performance may be the result of poor public sector governance lead-
ing to poor oil revenue management. This phenomenon has sometimes been
described as political Dutch disease, and was noticed by political scientists in
the context of the MENA region. Several authors (Wantchekon 1999, Ross
2001, and Lam and Wantchekon 2002) find a positive correlation between
resource dependence and authoritarian governments controlling for charac-
teristics such as GDP, human capital, income inequality, and other possible
determinants.
But can such regimes generate better governance in the absence of democ-
racy? In fact, a recent study (Collier and Hoeffler 2005) finds major differ-
ences in the economic performance of autocracies and democracies when
controlling for natural resource endowments. Richly endowed countries, such
as oil producing countries, seem to perform better economically if they are
not restrained by democratic institutions. The underperformance of oil-rich
democracies is explained by economic policy choice, namely by the size and
quality of investment. In view of the finding by Tavares and Wacziarg (2001)
that democracies in general tend to under-invest in physical capital as they
focus more on policies related to human capital development and a more
equitable society, Collier and Hoeffler conclude that oil-rich democracies not
only under-invest but also do so badly, since they face less financial, and con-
sequently, political restraints. Resource-rich countries do not need to tax as
much, which results in less scrutiny on their delivery of public services by
their citizens. Therefore, the key argument is that resource-rich democracies
need a distinctively different design that places more importance on checks
and balances, that is, on instruments that rebalance how power is used, rather
than on mechanisms that determine how power is achieved.
But empirical evidence shows that major oil-producing countries tend to
have lower governance indicators. Though they seem to need more checks
–and –balances, oil-reliant countries have the worst index of public account-
ability scores. Having substantial oil and gas revenues accrue directly to gov-
ernment budgets means that governments can maintain a deficient gover-
nance environment—as long as they do some redistribution and provide pub-
lic goods to the population. In a situation of “no taxation, no representation,”

Part I: Growth, Reform, and Governance 127
governments face little pressure to improve governance to increase economic
development. Substantial revenue from natural resources relieves govern-
ments of the need to tax, thus reducing their obligation to be accountable. In
addition, they are able to redistribute a significant share of their oil revenue
through public employment and broad access to cheap public services. These
two factors—no taxation and some redistribution—help mute demands for
accountability (World Bank 2003e). While the presence of mineral wealth in
a country may not be the cause for a governance deficit, it could make it more
difficult for good governance institutions to emerge.
Eifert et al. (2003) reinforce this conclusion,and the importance of democ-
racy for these countries. The study analyzes the oil-rich regimes of the world,
and, depending on their characteristics, divides political systems into (i)
mature democracies; (ii) factional democracies; (iii) paternalistic autocracies;
(iv) reformist autocracies; and (v) predatory autocracies. The study concludes
that mature democracies have clear advantages in managing oil revenues for
the long term, because of their ability to reach consensus, their educated and
informed electorates, and a level of transparency that facilitates clear decisions
on how to use the oil revenues over a long-term horizon. Reformist and pater-
nalistic autocracies lack transparency and face the risk of oil-led spending
being the legitimizing force behind the state, which tends to foster corruption
and creates problems with political transitions. These countries tend to be
locked in high-spending patterns that are unsustainable in the very long term.
Factional democracies lack an effective political system to create consensus
among competing interests. Finally, predatory autocracies have short-term
horizons and the characteristics of kleptocratic regimes that siphon money
from state coffers, eventually drying up oil wealth.
The “Binding Constraint” to Growth Approach
The previous review of the links between democracy and economic growth
has relied mostly on work that tries to find systematic relationships from
cross-country comparisons. While such research finds complex relationships,
and sometimes uncertain results, it tends to show the existence of a strong
positive relationship between democracy and growth, especially if democrat-
ic institutions are designed in such a way as to lead to better governance, min-
imize the possible negative impact of political market failures, and avoid the
risks from social cleavages.
But even using such an analytical framework to try to find systematic rela-
tionships between democracy and growth, Barro and Sala-i-Martin (2003)
show that a nonlinear relationship may exist. Democratization appears to
enhance growth for countries that are not very democratic, but to retard
growth for countries that have already achieved a high degree of democracy.
This finding, applied to MENA countries, would mean that democracy is

Breaking the Barriers to Higher Economic Growth128
important for growth, given their present low scores in democratic
development.
A completely different approach to looking at the relationship between
democracy and growth is to recognize that expectations of an overall system-
atic relationship between democracy and growth are ill placed. The impact of
democratization on growth should be country- and time-specific; the static
search for a stable relationship may be counterproductive.
One such alternative way to link democracy and economic growth is to use
the recently-developed approach of “binding constraint” proposed by
Hausmann, Rodrik, and Velasco (2004). This approach holds that constraints
to growth are time- and country-specific. It rejects cross-country findings and
one-size-fits-all solutions as useless tools for studying relationships between
reforms and economic growth. In that framework, one has to ask the ques-
tion: at a given time and in a given country, is the “democracy deficit,” com-
pared to other factors, the binding constraint of growth? For MENA coun-
tries, this could imply that democracy did not matter in the past but that it
may be critical now. It may also imply that there is no general answer to this
question for the region as a whole; one has to be country-specific.
The argument for the binding constraint approach for most countries in
the region might go as follows. The low economic growth in the MENA coun-
tries is to the result of low private investment, which is itself to the result of
weak investment climates and poor public sector governance. Major and cred-
ible reforms are needed, especially in terms of public sector governance and
investment climate, in almost all countries of the region, in order to unlock
the growth potential (see World Bank 2003a).10 On the other hand, experi-
ence shows that after 20 years of attempts at reform, the depth and scope of
such reforms, while they vary from country to country, remain limited.
Political economy analysis suggests that political regimes, as they exist, have
been unable to generate the required reforms (Nabli 2005). The existing polit-
ical economy equilibrium favours the status quo: low public accountability
and the maintenance of prevailing economic policies and networks of privi-
lege. In such a situation, democratic reform may be able to unlock this state of
affairs and generate a great political and economic transformation that could
produce both more democracy and more economic growth in the region.
Conclusion: Striving for Democracy in the MENA Region?
Democratization yields benefits in the form of individual freedoms and
empowerment that are valued independently of their consequences in terms
of growth and material wealth. But democratization is also good because
democracies can (a) yield long-run growth rates that are more predictable; (b)
produce greater short-term stability; (c) handle adverse shocks much better;
and (d) deliver better distributional outcomes.

Part I: Growth, Reform, and Governance 129
Our review of the literature on the links between democracy and econom-
ic growth, and its application to the conditions of the MENA region, leads to
the conclusion that MENA countries should strive for democratic regimes
that are sustainable—in the sense of having characteristics that make them
more likely to produce good governance. This means that democratic devel-
opment requires going beyond an electoral process that guarantees free, open,
and competitive elections. These formal democratic processes have to be com-
plemented with a number of reforms that aim at (i) minimizing imperfec-
tions in the political market, with more freedom of information and free
press, adequate mechanisms to contain clientelism, and increased credibility
of political promises; (ii) introducing safeguards and effective checks and bal-
ances; and (iii) increasing the legitimacy of the democratic transformation. In
cases where there is significant ethnic or religious fragmentation, or both, it is
vital that the democratic institutions be designed so as to minimize the risks
of conflict and emergence of unaccountable government.
These should help ensure or maximize the likelihood that democracy leads
to better governance, and therefore to higher economic growth. In such a sit-
uation, one would not have to face short-term trade-offs between democracy
and economic growth.
Such a conclusion is reinforced by the “binding constraint to growth”
approach. Democracy may not be a “binding constraint” to growth, in the
strict sense that if it were achieved today, it would result in higher economic
growth in MENA countries. However, one can argue that progress in democ-
racy is probably critical at this stage of the MENA world’s history for achiev-
ing the required transformation, which would ensure better governance, more
accountability, a better investment climate, and credible policies for increased
private sector investment, employment, and growth.
Finally, one can argue that democracy can lead to better governance, and
therefore, better economic policies and credible reforms, and that even the
design of such economic reforms may in itself enhance democratic develop-
ment. Producing a virtuous circle, where democratic development enhances
governance and economic growth, will itself support consolidation of demo-
cratic development. For instance, more economic openness, which improves
the relative incomes of the owners of human and physical capital, together
with more economic equality, would enhance the development of democracy
(Acemoglu and Robinson 2006).
Notes
1. The MENA region, in the World Bank definition, includes all Arab countries except Sudan,Somalia,
Mauritania, and Comoros, plus Iran.
2. A recent paper by Gause (2005) reviews the question and challenges the view that promotion of
democracy in the Middle East would stop generating anti-American terrorism.

Breaking the Barriers to Higher Economic Growth130
3. The political regime universe has on one side democracy, and on the opposite side dictatorship (or
authoritarian regimes). Dictatorships are defined here as regimes in which political rulers accede to
power and maintain themselves in power by force. They use force to prevent societies from expressing
their opposition to rulers’ decisions. Because they rule by force, they are vulnerable to visible signs of
dissent. These opposing political regimes represent different ways of selecting rulers, processing and
resolving conflicts, and making and implementing public policy. In a sense, they are different ways of
organizing political lives. As such,they are likely to impact people’s lives and welfare in different ways.
4. The Polity IV index is produced by the Integrated Network for Societal Conflict Research Program
of the University of Maryland’s Center for International Development and Conflict Management
(CIDCM)). Polity IV contains coded annual information on regime and authority characteristics for
all independent states (with greater than 500,000 total population) in the global state system, and cov-
ers the years 1800-2003.
5. We thank Daron Acemoglu for providing the transformed Freedom House data.
6. Friedman (2005) provides a useful review and summary of the findings on this issue (Chapter 13).
7. Friedrich Hayek (1960) summarized this point by stating that “...it is in its dynamic, rather than in
its static, aspects that the value of democracy proves itself. As is true of liberty, the benefits of democ-
racy will show themselves only in the long run, while its more immediate achievements may well be
inferior to those of other forms of government.”
8. See Acemoglu et al. (2004).
9. In a recent study by Harber, Razo and Mauer (2003) on the politics of property rights, the authors
challenge the idea that political stability and broader property rights are necessary for economic
growth, based on Mexican historical evidence. They claim that economic growth does not always
require a government that is constrained from preying upon property rights; it only needs a govern-
ment that makes selective credible commitments to a subset of asset holders.
10. The Middle East and North Africa Region of the World Bank produced four major regional reports
on the occasion of the World Bank-International Monetary Fund Annual Meetings in Dubai in
September 2003. These reports-on trade and investment, governance, gender, and employment-are
intended to enrich the debate on the major development challenges of the region at the beginning of
the 21st century.
References
Aysan, A. F., M. K. Nabli, and M. A. Véganzonès-Varoudakis. 2006. “Governance and Private
Investment in the Middle East and North Africa.” Mimeo. World Bank, Washington, DC.
Acemoglu, D., and J. Robinson. 2006. Economic Origins of Dictatorship and Democracy. Cambridge,
U.K.: Cambridge University Press.
Acemoglu, D., and S. Johnson. 2005. “Unbundling Institutions.” Journal of Political Economy 113 (5):
949–95.
Acemoglu, D., S. Johnson, J. Robinson, and P. Yared. 2005. “Income and Democracy.” Mimeo
(February). Acemoglu, D., S. Johnson, and J. Robinson. 2004. “Institutions as the Fundamental
Cause of Long-Run Growth?” Paper prepared for the Handbook of Economic Growth, ed. Philippe
Aghion and Steve Durlauf.
Acemoglu, D., S. Johnson,and J. Robinson.2001. “The Colonial Origins of Comparative Development:
An Empirical Investigation.” American Economic Review 91: 1369–1401.
Alesina, A., A. Devleeschauwer, S. Kurlat, R. Wucziarg, and W. Easterly. 2003. “Fractionalization.”
Journal of Economic Growth 8: 155–94.
Barro, R. J., and X. Sala-i-Martin. 2003. Economic Growth. Cambridge, MA: MIT Press, 2nd edition.
Barro, R. J. 1999. “The Determinants of Democracy.” Journal of Political Economy 107: S158–S183.
Barro, R. J. 1994. “Democracy and Growth.” Working Paper 4909. National Bureau of Economic
Research, Inc., Cambridge, MA.

Part I: Growth, Reform, and Governance 131
Bollen, K. A. 2001. “Cross-National Indicators of Liberal Democracy, 1950-1990.” Computer file, sec-
ond ICPSR version. Chapel Hill, NC: University of North Carolina, Inter-university Consortium
for Political and Social Research.
Bollen, K. A.1990. “Political Democracy: Conceptual and Measurement Traps.”Studies in Comparative
International Development 25: 7–24.
Borner, S., A. Brunetti, and B. Weder. 1995. Political Credibility and Economic Development. London:
Palgrave Macmillan.
Bourguignon, F., and T. Verdier. 2000. “Oligarchy, Democracy, Inequality and Growth.” Journal of
Development Economics (August) 62 (2): 285–313.
Buchanan, J. M., R. D. Tollison, and G. Tullock, ed. 1980. Toward a Theory of the Rent-Seeking Society.
College Station, Texas: A & M University Press.
Collier, P. and A. Hoeffler. 2005. “Oil Democracies.” Mimeo. University of Oxford, U.K., Department
of Economics.
de Mesquita, Bueno B., and G. W. Downs. 2005. “Democracy and Development.” Foreign Affairs
(September-October) 84 (5).
de Soto, H. 1989. The Other Path. New York, NY: Harper and Row.
Easterly, W., and R. Levine. 1997. “Africa’s Growth Tragedy: Policies and Ethnic Divisions.” Quarterly
Journal of Economics 111 (4): 1203–50.
Eifert, B., A. Gelb, and N. B. Tallroth. 2003. “Managing Oil Wealth.” Finance and Development (40) 1.
Washington, DC.
Elbadawi, I. A.2004. “The Politics of Sustaining Growth in the Arab World: Getting Democracy Right.”
Lecture and Working Paper Series No. 2. American University of Beirut, Institute of Financial
Economics.
Elbadawi, I. A., and S. Makdisi. 2004. “Democracy and Development in the Arab World.”A program
initiative from the International Development Research Center.
Elbadawi, I. A.2002. “Reviving Growth in the Arab World.”Working Paper Series 0206. Arab Planning
Institute, Safat, Kuwait.
Evans, P. 1995. Embedded Autonomy: States and Industrial Transformation. Princeton, NJ: Princeton
University Press.
Evans, P. 1989. “Predatory, Developmental, and Other Apparatuses: A Comparative Political Economy
Perspective.” Sociological Forum 4 (4): 561–87.
Feng, Y. 2003. Democracy, Governance and Economic Performance: Theory and Evidence. Cambridge,
MA: MIT Press.
Friedman, B. M. 2005. The Moral Consequences of Economic Growth. New York, NY: Alfred A. Knopf.
Gause, F. Gregory III. 2005. “Can Democracy Stop Terrorism?” Foreign Affairs (September-October) 84
(5).
Giavazzi, F., and G. Tabellini, 2004. Economic and Political Liberalizations. Working Paper 10657.
National Bureau of Economic Research, Cambridge, MA.
Haber, S., A. Razo, and N. Mauer. 2003. The Politics of Property Rights: Political Instability, Credible
Commitments, and Economic Growth in Mexico, 1876–1929. Cambridge, U.K.: Cambridge
University Press.
Hall, R. E.,and C. I. Jones.1999. “Why Do Some Countries Produce So Much More Output per Worker
than Others?” Quarterly Journal of Economics 114 (1): 83–116.
Hausmann, R., D. Rodrik, and A. Velasco. 2005. Growth Diagnostics. Cambridge, MA: Harvard
University Press.
Hayek, F. A. 1960. The Constitution of Liberty. Chicago, IL: The University of Chicago Press.
Keefer, P. 2004a. “Does Democracy Help?” Mimeo. Development Research Group, World Bank,
Washington, DC.

Breaking the Barriers to Higher Economic Growth132
________. 2004b.“A Review of the Political Economy of Governance: From Property Rights to Voice.”
Mimeo. Development Research Group, World Bank, Washington, DC.
________. 2005. “Democratization and Clientelism: Why Are Young Democracies Badly Governed?”
Mimeo. Development Research Group, World Bank, Washington, DC.
Keefer, P., and S. Khemani. 2005. “Democracy, Public Expenditures, and the Poor: Understanding
Political Incentives for Providing Public Services.” The World Bank Research Observer 20 (1).
Knack, S., and P. Keefer. 1995. “Institutions and Economic Performance: Cross-Country Tests Using
Alternative Institutional Measures.” Economics and Politics 7 (3):207–28.
Lam, R., and L. Wantchekon. 2002. “Political Dutch Disease.” Mimeo. New York University.
Lipset, S. M. 1959. “Some Social Requisites of Democracy: Economic Development and Political
Legitimacy.” American Political Science Review 53: 69–105.
Madison, J., A. Hamilton, and J. Jay. 1987. The Federalist Papers. London, U.K.: Penguin Classics.
Nabli, M. K. 2005. “Restarting the Arab Economic Reform Agenda.” The Arab World Competitiveness
Report 2005. World Economic Forum, Houndmills, Basingstoke: Palgrave Macmillan.
North, D. 1990. Institutions, Institutional Change and Economic Performance. Cambridge, U.K.:
Cambridge University Press.
North, D. 1981. Structure and Change in Economic History. New York, NY: W.W. Norton.
Papaioannou, E., and G. Siourounis. 2004. “Democratization and Growth.” London Business School,
Economics Department Working Paper.
Persson, T. 2004. “Forms of Democracy, Policy and Economic Development.” Working Paper. Institute
of International Economics, Stockholm, Sweden.
Przeworski, A., M. E. Alvarez, J. A. Cheibub, and F. Limongi. 2000. Democracy and Development:
Political Institutions and Well-Being in the World, 1950–1990. Cambridge, U.K.: Cambridge
University Press.
Rivera-Batiz, F. L. 2002. “Democracy, Governance, and Economic Growth: Theory and Evidence.”
Review of Development Economics 6 (2): 225–47.
Rodrik, D. 2003. Growth Strategies. In Handbook of Economic Growth, ed. P. Aghion and S. Durlauf.
North-Holland: Harvard University, John F. Kennedy School of Government, forthcoming.
Rodrik, D. 1999. “Institutions for High-Quality Growth: What They Are and How to Acquire Them.”
Paper presented at the International Monetary Fund Conference on Second-Generation Reforms,
Washington, DC. November 8–9.
Rodrik, D. 1998. “Where Did All the Growth Go? External Shocks, Social Conflict, and Growth
Collapses.” Mimeo. Harvard University, John F.Kennedy School of Government, Cambridge, MA.
Rodrik, D. 1997. “Democracy and Economic Performance.” Mimeo. Harvard University, John F.
Kennedy School of Government, Cambridge, MA.
Rodrik, D. and R. Wacziarg, 2005. “Do Democratic Transitions Produce Bad Economic Outcomes?”
Working paper. Center on Democracy, Development and the Rule of Law.
Rodrik, D., A. Subramanian, and F. Trebbi. 2002.“Institutions Rule: The Primacy of Institutions over
Geography and Integration in Economic Development.” NBER Working Paper 9305. National
Bureau of Economic Research, Cambridge, MA.
Ross, M. 2001.“Does Oil Hinder Democracy?” World Politics 53: 325–61.
Sen, A. 1999. Development as Freedom. New York, NY: Alfred A. Knopf.
Schumpeter, J. A. 1942. Capitalism, Socialism and Democracy. New York, NY: Harper and Brothers.
Tavares, J. and R. Wacziarg. 2001. “How Democracy Affects Growth.” European Economic Review 45:
1341–78.
Wantchekon, L. 1999. “Why Do Resource Abundant Countries Have Authoritarian Governments?”
Working Paper 99: 12. Yale University, Leitner Center.

Part I: Growth, Reform, and Governance 133
World Bank. 2003a. Jobs, Growth, and Governance in the Middle East and North Africa. Unlocking the
Potential for Prosperity. Washington, DC: World Bank.
__________. 2003b. Trade, Investment, and Development in the Middle East and North Africa: Engaging
the World. Washington, DC: World Bank.
__________. 2003c. Unlocking the Employment Potential in the Middle East and North Africa: Toward
a New Social Contract. Washington, DC: World Bank.
__________. 2003d. Gender and Development in the Middle East and North Africa: Women in the Public
Sphere. Washington, DC: World Bank.
__________. 2003e. Better Governance for Development in the Middle East and North Africa: Enhancing
Inclusiveness and Accountability. Washington, DC: World Bank.
World Bank. 2005a. Economic Growth in the 1990s: Learning from a Decade of Reform. Washington, DC:
World Bank.
_________. 2005b. Middle East and North Africa: Economic Development and Prospects: Oil Booms and
Revenue Management. Washington, DC: World Bank.
_________. 2006. Sustaining Gains in Poverty Reduction and Human Development in the Middle East
and North Africa, Orientations in Development. Washington, DC: World Bank.

135
The Political Economy of
Industrial Policy in the
Middle East and North Africa
Mustapha K. Nabli
Jennifer Keller*
Claudia Nassif *
Carlos Silva-Jáuregui*
6
The mainstream view on industrial policy has shifted back and forth over the
past half century. During the 1950s and 1960s, industrial policy was wide-
spread. Import substitution was a common strategy used to nurture “infant
industries” throughout the developing world, including the Middle East and
North Africa (MENA). A number of industrial and investment policies were
used throughout the world to expand countries’ industrial bases and develop
key sectors.1
In some cases, industrial policy implemented over the 20th century seemed
to yield astounding results, but perhaps nowhere more so than in a few high-
performing East Asian economies. Interventions varied, including targeting
and subsidizing credit to selected industries; keeping deposit rates low and
maintaining ceilings on borrowing rates to increase profits and retain earnings;
protecting domestic import substitutes; subsidizing declining industries-
establishing and financially supporting government banks; investing in applied
research; establishing firm- and industry-specific export targets; developing
export marketing institutions; and sharing information widely between public
and private sectors. Some industries were promoted while others were not.2
*World Bank.
Published as Egyptian Center for Economic Studies Working Paper no. 110; May 2006.

Breaking the Barriers to Higher Economic Growth136
By the 1980s, however, views on industrial policy had decidedly changed.
Growing evidence found that traditional approaches to industrial policy led
to misallocation of labor and capital across industries and did not improve
long-run growth in total factor productivity, but gave rise to rent seeking.
Even among the high-performing economies of East Asia, it was argued that
industrial policies—notwithstanding the contribution to the growth of the
economies themselves—inflicted significant costs on the economies in the
form of corruption and weak financial systems.3In Europe and the United
States the practice of frequently “picking the loser” in declining industries
such as agriculture, textile, steel, and shipbuilding reinforced this notion.
Other factors contributed to industrial policy falling out of favor in most
development circles, including the wide acceptance of the “Washington
Consensus” reforms for Latin America, the fall of the Soviet Union, and an
increasingly globalized economy.
The diverse experiences with industrial policy in different countries have
shown that its outcome depends enormously on the national context, which
determines how industrial policies are framed and implemented. Ultimately,
the political economy will determine not only what types of industrial policy
will be pursued, it may also determine whether a given set of strategic indus-
trial policies will benefit one country, or specific economic groups therein, or
whether it will harm another.
Despite these changing views of industrial policy, most of the countries in
the MENA region continued to rely extensively on traditional industrial poli-
cy. Throughout the 1980s and 1990s, MENA countries maintained strong
roles for the government and policies of significant government intervention
in production and economic planning. During the same period, countries in
Latin America, Europe and Central Asia that had broadly similar initial con-
ditions undertook sweeping reforms of their economic structures and refo-
cused their strategies into more coherent market-oriented policy packages to
encourage private sector, export-led growth. The MENA economies made rel-
atively less progress toward more market-oriented policies, and today govern-
ments in MENA are still actively implementing traditional industrial policies.
More than in countries of other developing regions, these policies are charac-
terized by vertical elements that protect selected industries and preserve exist-
ing market organizations. They include higher trade protection, higher public
involvement in the production of goods and services, more control of strate-
gic sectors like banking, and more price controls and subsidies.
In this paper, we examine the political economy and consequences of
industrial policy in the MENA region. How can the particular features of
MENA’s industrial policy be explained? And why haven’t industrial policies
in MENA countries followed the evolutionary path of industrial policies of
other countries? Unlike in many other regions, industrial policy in MENA
developed within the context of the region’s strong “social contract” between

Part I: Growth, Reform, and Governance 137
the government and its people. Although industrial development was an
objective, it at times took a back seat to other goals such as social transfor-
mation and economic redistribution—which influenced not only the types
and success of industrial policies adopted, but also the balance of power
among interest groups. Political economy factors are central to understand-
ing the industrial policy experience of the MENA region. These issues have
scarcely been studied in the context of the MENA region, and this paper takes
a broad regional and historical perspective, with the aim of providing a
framework for examining them.
Section two of the paper provides the theoretical framework for under-
standing the industrial policy experience. Starting with a brief survey of the
arguments used to justify industrial policy interventions, and drawing on var-
ious strands of the literature, it provides a review of various mechanisms and
arguments to help understand the factors that determine the emergence and
type of industrial policies observed and how they change. Using this frame-
work, section three reviews the experience of MENA countries from the 1950s
to the 1970s, and the emergence of state-dominated, vertical industrial policy,
where traditional sector-selective and sector-specific policies have been used
extensively.
Section four attempts to explain why industrial policy in MENA failed to
change during the 1980s and 1990s. While most of the developing world has
moved toward more market-oriented policies and production systems, dom-
inated by the private sector and reliant on market signals, MENA has main-
tained many of the old-style industrial policies and the high level of state
intervention that characterized much of the developing world in the past.
Despite the mounting strains on MENA’s economic development models, oil
and strategic revenues, and the lack of a full-fledged economic crisis, have
allowed the region to maintain industrial policies far longer than other
regions. Equally important, the lack of interest groups to emerge and press for
change has hindered the region’s move toward more functional, market-
friendly policies for growth—a phenomenon closely linked to the weaknesses
in governance. In addition, during the initial industrialization stage, MENA
countries used industrial policy to create new activities and support the devel-
opment of new (infant) firms, but during the second stage (1980s-90s), indus-
trial policy played a more passive role—that of preserving existing structures.
From a political economy perspective, the preservation of structures can be
explained by governments’ desire to seek support to remain in power by con-
tinuing to offer rewards to supporters, in order to deter the formation of
opposition groups.
The final section offers some concluding remarks on the likely direction of
industrial policy in the region. As internal and external forces shape the way
industrial policies can be used in the globalized economy, the MENA region’s
old style of industrial policy will need to adjust. The ultimate outcome will be

Breaking the Barriers to Higher Economic Growth138
largely determined by each country’s initial conditions and individual politi-
cal economy factors.
Understanding Industrial Policy
What exactly constitutes industrial policy? In the current vernacular, any poli-
cies or interventions that influence how industries expand are referred to as
“industrial policies,” but distinctions are made between “horizontal” industri-
al policies and “vertical” industrial policies. Interventions that are differential-
ly applied across sectors of the economy are referred to as “vertical policy.”
Likewise, interventions applied across the board are referred to as “horizontal
policy.”
While vertical policies essentially target the economic output of specific
industries and even firms, horizontal policies focus on improving the quality
of inputs in the production process, which would presumably benefit all
firms. Horizontal industrial policies often cited include promoting education
and vocational training, building appropriate and efficient public infrastruc-
ture, encouraging international technology transfers, and fostering research
and development.
Differentiating between policies that would benefit all firms and those that
would benefit a select few may seem appealing, but it is not entirely accurate.
Even horizontal policies may have substantially different impacts among sec-
tors of the economy, and ultimately may be just as distortionary as vertical
industrial policies. For example, if a government provides across-the-board
energy subsidies (for example, in Iran) effectively lowering the unit price of
energy for all consumers, it is technically a “horizontal” policy in that it is
applied across sectors. However, it clearly affects firms differently, providing
greater benefits to more energy-intensive (and often energy-inefficient) firms.
Even a more “virtuous” horizontal policy, such as incentives for promoting
education, will affect sectors of the economy differently depending on the
education being promoted. Indeed, some firms may not benefit at all from a
more educated workforce if education is not in line with the work of the firm.
Thus, the line between “horizontal” and “vertical” industrial policy is often
very difficult to distinguish. Despite these difficulties, this distinction is use-
ful, especially from a political economy perspective, in order to try to under-
stand the strong political appeal of vertical industrial policy compared to
horizontal.
A useful starting point to understanding why countries primarily adopt
vertical rather than horizontal policies is to examine some of the economic
arguments behind both approaches. While the political economy may play a
great role in determining the industrial policy approach adopted, economic
justifications have provided a strong foundation for the road countries have
traveled toward industrial development.

Part I: Growth, Reform, and Governance 139
Economic Justifications of Vertical Industrial Policy
Industrial policy is traditionally justified by market failures that generate sub-
optimal outcomes in resource allocation. Once these market failures are iden-
tified, governments devise policy responses to cope with them in the form of
industrial policy. Incentives and policies can also be used to compensate and
correct government failures.
There have been a variety of economic justifications for the use of vertical
industrial policy. Perhaps the most notable economic justification was the
“infant industry” argument for selective protection. This dynamic
comparative-advantage argument claims that protection is warranted for
newly established firms and industries in countries where production costs
may be initially higher than those of well-established competitors. If, over
time, new domestic producers can reduce costs by learning by doing, then
they can attain the production efficiency of their rivals. Without the initial
protection, however, the domestic industry will never take off. The argument
was used to curtail competition with both domestic and foreign firms. Many
countries in MENA used this argument during the initial stages of industrial-
ization after WWII, particularly regarding establishing trade barriers to pro-
tect their “infant” industries from competition from abroad.
But many other economic justifications for vertical industrial policy have
been advanced as well. A second argument is that coordination problems of
either upstream or downstream investments may hinder the development of
otherwise competitive industries. This is exemplified by the development of
the orchid industry in Taiwan,4where potential orchid growers contemplat-
ing investment in greenhouses needed to be assured that a variety of fixed
investments were in place—including an electrical grid, irrigation, logistics
and transport, and pest control measures. At the same time, all of these serv-
ices had high fixed costs and were unlikely to be undertaken by the private sec-
tor without assurances that there were a sufficient number of greenhouses to
demand their services. In this case, the Taiwanese government’s upstream
investments aimed to coax the downstream investments in greenhouses.
In the case of MENA, the argument has been used to support the creation
of industrial zones (for example, in Morocco), or large investments in the
water and irrigation sector to support agriculture (for example, in Egypt). In
Jordan, a joint public-private initiative led to the establishment of “Cyber
City” in 2003, which provides transportation, logistics, insurance, technology,
and other commercial services to attract investments in the IT sector. In the
service sector, specialized hubs have been created in the Gulf to support the
development of financial services, education, health, trading, and transport.
Qatar, for instance, is positioning itself to become a regional education and
health services hub.
A third argument is based on information externalities that may restrict a
country’s capacity to determine which activities have costs low enough to be

Breaking the Barriers to Higher Economic Growth140
profitable. Unlike innovation, which can be protected with patents, the costs-
(risks)-to-benefits ratio is high for starting a domestic industry that is already
well established internationally. As Rodrik (2004) points out, entrepreneurs
who figured out that Colombia was a good environment for flowers,
Bangladesh for t-shirts, Pakistan for soccer balls, and India for software gen-
erated large social gains for their economies, but could keep very few of the
gains for themselves. For this reason, selective government interventions may
be required as a means of determining a country’s areas of cost advantage. In
MENA, information externalities arguments have been used in the develop-
ment of export promotion agencies that played an important role in the
Tunisian and Moroccan textile sectors.
The presence of market failures provides powerful arguments for govern-
ments to intervene, using vertical industrial policies to ensure that countries
can determine their areas of advantage and generate spillovers to other sec-
tors. In some instances, government failure, in the form of ineffective imple-
mentation of policies, leads to undesirable outcomes that have justified the
implementation of compensatory industrial policies.
Despite theoretically sound arguments for vertical industrial policy, inter-
national success with industrial policy has been far from glowing, confirming
that governments can “get it wrong.” Well motivated or not, vertical industri-
al policies have often either prevented the emergence of dynamic, competitive
enterprises or led to significant unintended consequences, not only when the
policies were in place, but for a long period afterward.
The socialist industrial policy pursued by India in the 1950s included pro-
tectionist policies that increased the cost of unskilled labor. As a result, the
country was shut out of the global market for manufacturing products with
unskilled labor for many years after it ultimately opened up.5Korea’s use of
credit allocation as an industrial policy instrument is widely blamed for the
financial system crisis that emerged in the late 1990s. Commercial banks were
urged to lend to firms in preferred sectors or to the large and powerful con-
glomerates (“cronies”). Consequently, banks incurred weak balance sheets
due to the low profitability of these firms. In addition, the policy nearly obvi-
ated bank skills for project monitoring and evaluation. As a result, the huge
capital inflows that began in the mid-1990s were channeled through institu-
tions that had suffered a serious erosion of skills and discipline.6
After Algeria nationalized nearly its entire economy in 1966, the govern-
ment invested heavily in the creation of basic capital-intensive industries (for
example, hydrocarbon, steel, plastic, and fertilizers) and in prioritized indus-
tries relevant to processing and import substitution (for example, construc-
tion materials, metal products, consumer goods). In many respects Algeria
succeeded, and in less than a decade created a strong industrial base virtually
from scratch. However, the rapid industrialization resulted in severe ineffi-
ciencies in the production capacity of various industries and generated fiscal

Part I: Growth, Reform, and Governance 141
imbalances in the mid-1970s, which constrained Algeria’s ability to continue
its expansionary industrial policy.7
In addition to possibly “getting it wrong,” vertical industrial policy is sub-
ject to two potential (and common) damaging side effects: rent seeking and
corruption. Wherever the government makes selective interventions that
could contribute to the development of one sector or firm over another, there
is the potential for interest groups to attempt to sway policy in their direction
and utilize it for personal gain. The ability of countries to control corruption
and rent seeking, in fact, is a key difference between countries in which indus-
trial policy could be used effectively and those where the strategy floundered.
However, the types of interest groups that emerge in a country, the mecha-
nisms they can use to influence public policy, and how far they can shift
industrial policy from “good” to “bad” can vary enormously from country to
country. Attempting to deal with these government failures often leads to
more selective policies, making industrial policies increasingly complex.
Economic Justifications of Horizontal Industrial Policy
While market failures, information asymmetries, externalities, and problems
of coordination and learning are among the most commonly cited reasons for
government intervention, addressing these issues from a horizontal policy
perspective rather than a sector- or firm-specific one has been gaining accept-
ance by policy makers in the developing world.
Horizontal industrial policies may have many advantages over vertical poli-
cies. On one hand, horizontal policies tend to reduce the distortions generat-
ed by the use of vertical industrial policy. They approach industrial policy
from an angle that is closer to competition policies, while still actively sup-
porting the economy. Horizontal policies are applied across the board. As
such, they tend to level the playing field across firms, industries and sectors,
rather than giving a privileged position to some groups. Because of their “uni-
versal” nature, they also tend to reduce rent-seeking incentives and to limit
corruption opportunities. This reduces the development of pressure and lob-
bying groups that seek to benefit from policies targeted to their interest.
Horizontal industrial policies also increase transparency, by eliminating the
need for backroom politics, and promote social cohesion as they can be seen
as spreading the benefits across society as a whole instead of concentrating
them within specific groups. In addition, they can promote efficiency and
competition among firms, industries, and sectors in the economy, as each of
these agents will have greater incentives to internalize the most from the hor-
izontal policies.
On the other hand, horizontal industrial policies can serve to reduce the
problem of state capture and government failure. Because these policies are
applied across the board, they can eliminate some of the problems that help
perpetuate the use of vertical policies. The distribution of benefits is more

Breaking the Barriers to Higher Economic Growth142
clearly defined with horizontal policies and the problems of non-neutrality
are in principle eliminated (or at least reduced). However, it is important to
note that while policies can be horizontal in design and nature, their effect
may not be horizontal. The energy subsidy example above clearly indicates a
case in which a horizontal policy may have a “vertical” outcome.
Finally, compared to traditional vertical policy, horizontal industrial policy
adjusts more easily to changing market conditions, since its benefits are not
captured by special groups, with vested interests, that would lobby to main-
tain the status quo. This builds a much needed dynamism in the policy itself,
facilitating change and adjustment when needed.
Political Economy of Industrial Policy
The traditional approach to political economy posits that economic policies
are shaped in the political market. The main actors in the market are interest
groups or distributional coalitions who are rent seekers pursuing politically
mediated gains. Any policy change is bound to produce winners and losers,
who may or may not organize to block these policies and/or support other
policies. Finally, policy makers mediate these pressures and determine their
choices in view of their own objectives.
Sometimes policy makers may determine that it is in their interest to
undertake fundamental policy reforms that may disrupt existing political
coalitions and dislocate privileged economic actors. The old elites become los-
ers in the reformed economic environment and a new set of winners emerges
from the process. For example, reforms may aim at shifting the economic
environment from cronyism, patronage, corruption, and rent seeking to
transparency, accountability, and clear property rights. Such reform measures
are risky because they disrupt the status quo and remove the dominant elites.
In such situations reformers and policy makers distance themselves from pre-
vailing interest groups, as they may be attempting to reach some longer-term
objectives and mobilize alternative coalitions. The existing distributive coali-
tions do not shape policy reforms. The interests of winners and losers are
essentially fixed by their position in the pre-reform political economy. Elites
react to the reforms by attempting to block them or undermine the reform
process to avoid losing their privileged position.8
So what influences the industrial policy choices countries make? Or more
to the point, why, despite the frequent failures and the risks of vertical indus-
trial policy, do political systems predominantly tend to produce and perpetu-
ate vertical policies rather than horizontal? Despite the criticism leveled at
selective policy interventions, most countries—both developing and
developed—maintain something of a “halfway house” between mainstream
free-market policy measures and vertical interventions to encourage specific
industries. A number of arguments, rooted in the principles of collective
action and interest groups, are relevant.

Part I: Growth, Reform, and Governance 143
First, the benefits from horizontal policies tend to be diffused across groups
and sectors, and it is not often easy to mobilize “winners” from horizontal
policies. Take, for example, the case of education, skills development, and
technological innovation policies. Although many businesses might benefit
from skills upgrading, the extent to which a given firm could reasonably
expect to benefit from across-the-board education policies might be small. In
fact, a great deal of the resources might be devoted to the development of
skills not used intensively in its own business. It is unlikely that collective
action will emerge to support these generalized policies. On the other hand,
within a particular sector, the types of skills used and technological knowledge
needed are often highly specific. As a result, firms in a given sector are proba-
bly more likely to be able to develop lobbying power to pursue specific educa-
tional objectives.
Second, some of the most important horizontal policies are long term in
nature; it often takes a long time for results to materialize. This is particularly
the case with education and non-sector-specific research and development.
Interest groups are unlikely to pursue and push such policies when the bene-
fits to them are less visible and spread out over a long period of time.
Therefore, it is up to governments to actively define and support such long-
term development policies.
Third, in many instances, market and coordination failures are sector spe-
cific, and cannot be adequately addressed through horizontal approaches.
Take the case of the tourism sector, in which coordination problems are per-
vasive. The emergence and/or expansion (new zones, new business niches) of
the sector require coordination on a number of fronts: (i) development of
basic infrastructure, such as zones with adequate water supply, sanitation,
access to international transport, and similar core needs; (ii) development of
specific technical and managerial skills; (iii) development of joint or support
activities such as travel agencies, entertainment businesses, restaurants, and
related activities; and (iv) information on markets, advertising, and opening
of new markets. Given the number of factors that must come together for
development of the sector, it is unlikely that market forces and mechanisms
would be able to resolve all of the coordination problems. Typically, govern-
ments have to be involved. This was partly the case with the development of
the tourism industry in Egypt when the government invested heavily in
improving the tourism infrastructure near historic sites during the 1980s.9
Finally, arguments used by sector lobbies can often be couched in terms of
benefits to other groups, such as workers and consumers, in strengthening
their political power. Tariff protection for a specific sector is often couched in
terms of domestic jobs protected. Protection of agricultural production and
subsidies for European farmers are couched in terms of preserving the envi-
ronment and livelihood of weak farmers. Protection of domestic production
of many food products, such as meat, is couched in terms of health safety for

Breaking the Barriers to Higher Economic Growth144
consumers. These arguments cannot be made for blanket horizontal policies.
They are intrinsically product- and sector-specific, and they have a powerful
political thrust.
Lessons from reforms in the developing world (including the MENA
region) indicate that in many instances, pre-reform elites are resilient.
Privileged economic actors have been able to keep their positions after
reforms. The process of reform has been one of reorganizing opportunities for
rent seeking rather than eliminating them. Reforms have produced outcomes
that continue to provide significant opportunities for privileged economic
actors to capture rents from a set of regulatory arrangements and economic
institutions.10
These lessons and experiences suggest that the effectiveness of political sys-
tems to generate welfare-enhancing policies would depend on a number of
factors that can be usefully summarized as good governance. These factors
include: (i) the extent to which policy makers are held accountable for the
choices they make and the extent to which they pursue public vs.private inter-
est; (ii) the transparency of the decision-making process and political influ-
ence; (iii) the ease with which interest groups can organize to pursue their
interests openly; and (iv) the quality of information and analysis available to
support and enlighten decision making.
The Emergence of State-Dominated Vertical Industrial Policy in MENA
from the 1950s to the 1970s
Industrial policy in the MENA region developed to some extent as an offshoot
of the region’s social contracts that emerged in the 1950s. Partly to correct for
a legacy of inequities and poverty in the region, the MENA countries adopted
models of development based on heavy state intervention and redistribution.
The social contract was designed to align with the norms and expectations of
social groups to the legitimate claims they had on state resources, as well as to
determine the demands state actors may legitimately make on society. It had a
number of distinctive features, including institutional arrangements, public
policies, legitimating discourses, and modes of state-society interrelations.
Core attributes of the social contract—some of them directly linked with
industrial policies of choice—included: (i) the preference of states over mar-
kets in managing national economies; (ii) a reliance on state planning in deter-
mining economic priorities; (iii) a penchant for redistribution and equity in
economic and social policy; (iv) an encompassing vision of the role of the state
in the provision of welfare and social services; and (v) a vision of the political
arena as an expression of the organic unity of the nation, rather than as a site
of political contestation or the aggregation of conflicting preferences.11
Another key marker in the rise of the interventionist-redistributive social
contract and its associated industrial policy was the emergence of centralized,

Part I: Growth, Reform, and Governance 145
hierarchical, and tightly regulated corporatist structures of interest groups in
the first decade after independence. These arrangements provided the blue-
print for the organization of relations between the state and a wide range of
stakeholders, including firms, laborers, students, and women, in addition to
various professional associations. It was through these arrangements that cor-
poratism created possibilities for agency, bargaining, and negotiation for the
groups it was designed to contain; the so-called privileged networks that are
the center of the new theories of political economy of reform. These struc-
tures also helped to determine the modalities of industrial policy used in the
region, and facilitated state capture and corruption.
Institutionally and in terms of economic and social policies, these elements
were consolidated through broadly similar strategies in a number of MENA
states, including Algeria, Egypt, Iraq, Syria, and—to a lesser degree—Jordan,
Morocco, Yemen, and Tunisia. These strategies included (i) the rise of a dom-
inant single-party or ruling-party government; (ii) new, postindependence
constitutions that enshrined interventionist and redistributive principles in
the basic laws; (iii) a wave of agrarian reform programs to redress inequalities
in the rural economy; (iv) the centralization of trade unions and professional
associations; and (v) programs for state provision of social services, including
education, housing, health care, food subsidies, and other benefits.
Understanding the reasons for the widespread “acceptance” of this type of
state-dominated social contract in MENA countries is beyond the scope of
this paper. However, it is clear that the political system underlying this social
contract is different from the kind of governance structure that is typically
assumed when arguments are made about the likely emergence of a vertical
industrial policy. Such arguments assume a governance structure where polit-
ical competition is prevalent, the market system is predominant, and there is
a large degree of openness and availability of information and analysis. The
different system prevalent in MENA would have significant implications for
the type of industrial policies that emerged during the 1950s and 1960s. The
attributes of the social contract and governance structure led to four mecha-
nisms/features of this model, which modified and/or reinforced standard
political economy arguments:
• First, sweeping nationalizations of industry, trade, and agriculture in the
late 1950s and early 1960s produced a dramatic expansion in the scale of
public sectors and reduced the role of the private sector; the relationship
between state-owned enterprises and decision makers allowed them to
influence policies more effectively.
• Second, the state captured (either wholly or predominantly) the banking
and insurance sectors.
• Third, price controls and subsidies were chosen as the predominant mode
of regulation based on the need to protect the poor and pursue a social
agenda of redistribution.

Breaking the Barriers to Higher Economic Growth146
• Fourth, the role of oil wealth—both oil resources for oil-producing
economies, and oil-related revenues12 for resource-poor economies—
underwrote much of the region’s emerging social contracts, and the public
sector (both governments and state-owned enterprises) became a key vehi-
cle for redistribution through employment. While this trend may be more
apparent in the oil sectors, the role of the state in most of the MENA coun-
tries has been dominant, comparable at times to the command economies
of the former Soviet bloc.13 As a result of these trends, by the 1960s the
commanding heights—the means of production—were mostly in the
hands of the state in many MENA countries.
Within the predominant role of the state and the use of central planning as
the main vehicle for resource allocation, these mechanisms and characteristics
led to an industrial policy that was bound to be sectoral/vertical and highly
preferential, thus creating an environment of “winners” and “losers.”
The Public Enterprise Sector and Industrial Policy
To a large extent, industrial policy was structured to support public sector
enterprises. This was partly the result of the waves of nationalization imple-
mented in several countries of the region. Like many countries of the world
that embraced the infant-industry concept in the 1950s, MENA implemented
an inward-looking model with protection from external competition. Import
tariffs, licenses, prohibitions and other forms of nontariff barriers were used
as the direct instruments of choice to support public sector enterprises, but
other policies supported public sector production (and employment), includ-
ing credit rationing, subsidies, and foreign exchange policies.
The nature of public enterprises and their relation to decision makers
made the pursuit of sectoral interests relatively easy and the implementation
of incentive schemes straightforward, as most of the instruments could be
shaped accordingly. MENA’s trade policies of high import tariffs and nontar-
iff barriers, which limited competition from abroad, echoed strategies adopt-
ed elsewhere, protecting industries that would, it was hoped, flourish and
compete later on.14 When, despite trade protection, these firms incurred loss-
es, it was easy to cover them directly from the budget, or more commonly,
from the banking system.
In situations where the private sector continued to play a role, it often faced
distorted incentives and the negative impact of the presence of public enter-
prises, such as higher cost and lower quality. Under such circumstances, the
only way for private businesses to pursue their interests was, in addition to
dealing with externalities, to obtain specific incentives in the form of special
protection, access to finance, or other subsidies. This process enhanced the
vertical and complex nature of industrial policy.

Part I: Growth, Reform, and Governance 147
Financial Sector and Industrial Policy
While this instrument of industrial policy is typically not available or of lim-
ited use in capitalist economies, it has emerged as one of the major mecha-
nisms of sector-specific interventions in MENA countries. In combination
with pension and social security funds, bank and nonbank financial institu-
tions were used to collect sizable resources that were managed by the state.
Savings were collected at low cost through administered interest rates that
were usually negative in real terms. This generated implicit subsidies that were
transferred to the privileged priority firms and sectors. Credit was directed by
central bank command. Countries including Algeria, Egypt, Jordan, Morocco,
and Tunisia created industrial development banks to provide foreign
exchange loans for the imported capital goods necessary for investment.
Government control of the banking systems made it possible to pursue verti-
cal industrial policy. Resources of the banking system were directly allocated
to selected activities, with quota allocations by sectors and preferential access
Box 6.1. Industrial Policy in Tunisia
After gaining independence in 1956, Tunisia’s manufacturing and mining sector was
small, and in the years that immediately followed, there was little or no growth. Most
Tunisian businessmen continued the long commercial tradition of the country, and
there was no net inflow of capital from foreign industrial investors, partly as a result
of uncertainty about the government’s attitude to private investment and political
tensions with France. In the late 1950s the government began to take steps to
assure the industrial development of the country.
In the first half of the 1960s, the government launched an initiative involving a
number of government offices and financial institutions, including the National
Company for Investment (SNI), to establish several industrial enterprises that were
either directly or indirectly controlled by the state. From 1959-60, the major share of
industrial financing was provided by the government either directly, through equi-
ty participation, or indirectly through its banking and insurance subsidiaries lending
or taking equity positions in industrial companies.
By the end of the 1960s virtually every industry was primarily controlled by the
government. The tourism sector was the one area in which the government had
been hesitant to invest, mostly because it regarded the investment potential as too
risky and because there was considerable opportunity for private activities.
Convinced that the tourism industry had great growth potential, the government
actively promoted it with measures such as tax holidays for new hotels, a rebate on
loan interest rates for tourism investments, preferred access to land, and subsidies
for hotel construction. These policies led to high growth rates in the tourism sector,
which developed into one of the main growth engines of the Tunisian economy
during the 1970s.
Source: World Bank (1966, 1969, 1971, 1972).

Breaking the Barriers to Higher Economic Growth148
by public enterprises. This is epitomized by Morocco, which created the
Banque Nationale Pour le Développement Economique (BNDE) in 1959, with
the sole purpose of providing loans to investment projects in selected indus-
tries that had experienced insufficient growth. In line with its mandate, the
BNDE significantly contributed to the expansion of the tourism sector, the
agrofood industry, and the textile industry. In addition, monetary policy was
conducted through direct credit allocation and refinancing.
Consumer Subsidies and Industrial Policy
MENA countries also used subsidies as an active policy choice. In part, subsi-
dies and other artificial supports were a necessary part of industrial policy in
MENA because of the externalities of the instruments of redistribution.
Administered prices prevailed throughout MENA economies and damaged
the link between prices and production costs, making compensation mecha-
nisms necessary.
Among initial subsidies, the most pervasive were for consumer goods,
especially foodstuffs. These subsidies-cum-price-controls meant that specific
sector policies were needed for agricultural and food products in order to
compensate for the weakened production incentives. As a result, the agricul-
tural sector required further policy instruments of trade protection, access to
preferred bank financing and subsidies. The incentives were justified as criti-
cal for employment, protecting the poor and maintaining social peace.
Oil Wealth and Industrial Policy
In addition to supporting the social contract, in many MENA countries
industrial policy developed in reaction to the influence of oil wealth.
Recognizing the impact that Dutch disease had on the competitiveness of
tradables, governments instituted a range of compensatory policies to miti-
gate the adverse effects. In a real sense, industrial development in the MENA
economies could not take place without some direct government interven-
tions. This is most obvious for the oil-producing economies, but even in the
resource-poor economies of the region, the exchange rate appreciation that
occurred from the massive inflow of aid and remittances contributed to the
view that vertical industrial policies were needed for industrial development.
Additionally, the abundance of oil revenues has given rise to interest groups
who have sought to retain a disproportionate flow of the rent and evaluate
reform policies based on their capacity to be captured.15 This has perpetuat-
ed and motivated the use of more vertical industrial policies.
The Economic “Outcomes”from MENA’s Vertical Policies
Some of the economic costs of the industrial policies adopted in the MENA
region have been widely acknowledged. The continued strategies of import
protection and inward orientation in MENA have resulted in significantly

Part I: Growth, Reform, and Governance 149
weakened trade, with trade-to-GDP growth at about half of the world’s pace
since the 1980s. The region’s exports are dominated by oil, with only the small
number of resource-poor and labor-abundant economies developing fairly
well-established nonoil export sectors. The entire MENA region, with a pop-
ulation close to 320 million, has fewer nonoil exports than Finland or
Hungary, countries with populations of 5 and 10 million, respectively.16
Also, the region did not sustain high levels of productivity after the 1960s.
MENA experienced two decades of strong economic growth during the 1960s
and 1970s. In fact, MENA’s economic growth performance in the 1960s was the
strongest in the world. During the decade, productivity growth was strong, in
part as a result of the industrial policies adopted, which allowed the region to
utilize underused capacities and provide the early boost of industrialization.
But by the 1970s, productivity deteriorated sharply as massive investments
were generating increasingly poor payoffs in terms of growth. Over the 1970s
and 1980s, productivity growth in the MENA region was negative.17
The MENA industrial policy strategy has also been less successful than
expected. Egypt adopted an active industrial policy from the early 1960s until
the early 1990s, when some elements of vertical interventions were phased out
in the wake of structural reforms that did not prove very successful. One could
argue that if industrial policy had been effective the Egyptian economy should
be more diversified.A recent assessment of the impact of industrial policies in
Egypt revealed, however, that this is not the case. In contrast, the manufactur-
ing industry has become more concentrated over time, particularly between
the 1980s and 1990s.18
Moreover, full employment—a virtual mainstay of the social contract—
could not be maintained. Between the 1980s and 2000, the unemployment
rate climbed from an average of less than 8 percent to 15 percent.19 By 2000,
the MENA region’s unemployment rate was higher than every other region of
the world except Sub-Saharan Africa.
Beyond the economic failures, the industrial policies adopted in MENA
powerfully influenced the emergence and control of interest groups in the
region, a fact that heavily contributed to the continued use of vertical policies
well beyond their justification.
The Failure to Change Industrial Policy in the 1980s and 1990s
Deteriorating budget deficits and the lack of economic growth prompted a
handful of economies in the region—including Jordan, Morocco, and
Tunisia—to embark on programs of macroeconomic stabilization and struc-
tural reform in the mid-1980s. The programs aimed to restore macroeconom-
ic balances and promote private sector-led development. By the late 1980s and
early 1990s most governments followed suit, adopting some form of econom-
ic stabilization. Policies varied, but included cutting subsidies, reducing pub-

Breaking the Barriers to Higher Economic Growth150
lic spending, liberalizing trade, reforming exchange rate regimes, encouraging
exports, easing restrictions on foreign investment, privatizing state enterpris-
es, and strengthening the institutional foundations of a market-led economy,
including consolidation of the rule of law. Many governments joined interna-
tional trade-promoting institutions and signed trade agreements to spur the
domestic economy.20
The use of vertical industrial policy diminished in MENA, but compared
with other regions it has remained in place to a large extent. In the area of trade
protection, tariffs in MENA countries have been slow to decline, in contrast to
the rapid decrease observed in other developing countries. In 2004, half of the
countries in the region had average tariff protection21 higher than that of
developing countries as a whole. Ranking countries worldwide according to
average tariffs revealed that on average, MENA countries are in the bottom 35
percent in terms of tariff protection, second only to Sub-Saharan Africa22 (see
figure A1 in the Appendix).23 Only in the last several years has the region
achieved some progress in lowering barriers to trade. Nontariff protection—
which can constitute a variety of measures, including quantitative restrictions,
rules for valuations of goods at customs, rules of origin, and price control
measures—has been reduced, but it is today still higher in MENA countries
than in most countries of Latin America, East Europe,or South Asia. According
to IMF classification, only three countries show a low incidence of nontariff
barriers, namely Djibouti, Qatar, and the UAE, while all others show either an
intermediate or high incidence, such as Syria, Iran, and Libya (see tables A1 and
A2). Significant distortions in the tariff schedule continue to exist, especially in
Iran, Syria, Tunisia, and Morocco (see table 6A.1).
Box 6.2. Financial Sector—Islamic Republic of Iran
The banking sector in Iran remains essentially dominated by the state. There are 10
large state banks in Iran. Six of them are general commercial banks that take deposits
from the public and make loans to both the public and private sectors. The other four
are specialist banks that lend to particular sectors: One lends exclusively to finance
housing, a second lends to the agricultural sector, a third to industry and mining; the
fourth specializes in export finance. These four specialist banks obtain most of their
funds from the general commercial banks, the central bank, and other public sources.
The instruments that banks can use for borrowing and lending are governed by
a 1982 law on Islamic banking. The rates of return on both loans and deposits set by
the central bank have been generally less than the rate of inflation over the last
decade. The controlled lending rates vary with the term and the sector receiving the
loan. Every large loan must be approved by the central bank, which also sets the
minimum percentages of each bank’s loan portfolio that must be lent to various
broad sectors such as housing, agriculture, exports, and state-owned enterprises.
Source: World Bank (2003e).

Part I: Growth, Reform, and Governance 151
Inefficiencies and distortions of financial sector policies became apparent,
and in many countries of the region, reforms have taken place to liberalize
financial systems, and in a limited number of cases even privatize them.24 In
fact, the financial system might well be the policy sphere that has experienced
the biggest reduction in government intervention. All countries, Syria and
Iran aside,25 unified their currency rates either before or in the early 1990s.
Around the same time, most countries moved away from the administration
of interest rates (see tables A3 and A4). But several MENA countries continue
to use these distorted policies today. Access to credit by the private sector
remains limited in Algeria, Iran,Libya, Qatar,Syria, and Yemen, where the pri-
vate sector receives less than 35 percent of all domestic credit extended (see
table 6A.5).
MENA countries also used subsidies as an active policy choice to support
their industries (and consumers). Subsidies were greatly reduced as part of
macroeconomic stabilization programs; cash subsidies to industries in partic-
ular were reduced by almost 50 percent. Nonetheless, the levels of subsidies
remain high. Studies indicate that subsidies in the form of direct cash trans-
fers to enterprises are significant in MENA countries, albeit not as high as the
European Union or in Europe and Central Asia (see figure A2).26 However,
because of the difficulties of measuring subsidies (rates of effective protection
and effective assistance are not easily available), cash transfers to industries
may be only a weak proxy for the actual level of subsidies, which might be sig-
nificantly higher.
Despite the MENA region’s lack of success with industrial policy, transition
from “bad” vertical industrial policies toward those that are more horizontal
has proceeded slowly, especially in comparison to the transitions in other
regions of the developing world, such as Latin America and the Caribbean,
East Asia and the Pacific, and Europe and Central Asia. Several factors have
contributed to this slow transition.
The first factor, perhaps ironically, has been the fact that the results of
industrial policy have not been “bad enough.” There is a line of thinking
among those who study reform that deep economic reform movements only
result from fundamental change, either from leadership change (regime
change, or a shift in governing coalition), or from dramatic economic
change—in other words economic crisis. In many Latin American countries,
for example, economic reforms were undertaken only when the “economic
conditions had deteriorated sufficiently so that there emerged a political
imperative for better economic performance.”27 In other words, reform often
is only adopted “once the possibilities of throwing money at the problem are
foreclosed.”28 Crises elsewhere—Latin America in the early 1980s and Europe
and Central Asia in the late 1980s—generated pressure for both political and
economic change. As a result, countries in those regions moved from com-
mand to market economies with less state intervention.

Breaking the Barriers to Higher Economic Growth152
The MENA region, although it has experienced a significant decline in
growth and employment from the gradual exhaustion of its economic mod-
els, has not experienced economic crises in a systemic way.29 The substantial
revenues from oil, which declined throughout the 1980s and 1990s, have
remained large (and of late increased significantly), giving MENA govern-
ments and the public a temporary sense of economic health. This, along with
foreign aid and strategic rents, has permitted MENA governments to maintain
damaging vertical industrial policies.
Second, the slow pace of change in industrial policy in the region reflects
either a lack of power among those interest groups that might be instrumen-
tal in lobbying for a move toward more horizontal industrial policy, or the
gradual creation of privileged networks that are influencing policy to retain
their rent-seeking capacity.
The private sector, rather than challenge the status quo, has adapted to the
prevailing industrial policy that has protected state-owned enterprises. Private
sector activity is concentrated in a small number of large firms that have ben-
efited from protective policies, along with a number of microenterprises,
which account for a large percentage of employment but have little access to
formal finance, markets, or government support programs. This behavior of
Box 6.3. Strategic Crops in the Arab Republic of Syria
One example that illustrates how industrial policy is used as an instrument for
achieving both economic as well as social objectives is agriculture policy in Syria.
Agriculture plays a key role in Syria’s economy by contributing substantially to
domestic production, employment generation, and export revenues. As in many
other countries in the world, the agriculture sector in Syria has been enjoying a high
level of public protection and support. Interventions are targeted toward the crops
that the government considers “strategic,“ namely wheat, barley, lentils, chickpeas,
cotton, and sugar, either because they provide significant export earnings or osten-
sibly ensure food supply.
The case of cotton shows to what extent these interventions result in systemic
distortions that are propagated throughout the economy. In order to protect farm-
ers from price volatility, the government buys cotton from farmers above world mar-
ket prices through its central marketing organization, which subsequently exports it.
While this is done at world market prices, the domestic industry is required to pur-
chase cotton at the state fixed price without the chance to obtain cheaper import-
ed cotton (high duties are levied on cotton imports). Consequently, cotton-based
textile manufacturing has been deprived from taking full advantage of the low
wage level in Syria, which otherwise would have helped to develop a real compar-
ative advantage in the textile industry. The government interventions in the agricul-
tural sector have, thus, been counterproductive and hampered growth in the textile
sector.
Source: FAO (2003) and World Bank staff.

Part I: Growth, Reform, and Governance 153
the emerging private sector has reduced the likelihood of faster reform and
policy change.
Trade unions, which could also be an effective vehicle for change, are tight-
ly controlled in the region and lack real independence from the political sys-
tem. Thus, trade unions have not been effective in organizing the labor force
to press for reforms. Trade union membership as a share of employment in
MENA averages about 9 percent, compared to 34 percent in the OECD, 43
percent in Europe and Central Asia, and 15 percent in Latin America and East
Asia. Only South Asia and Sub-Saharan Africa have lower trade union
membership.
The industrial export sector, which is highly competitive and would likely
lobby for horizontal policies, is grossly underdeveloped and generally scat-
tered among diverse product groups. A recent study examined the lobby
power of the manufacturing sector30 and found that MENA ranks last com-
pared to the developing regions of the world (table 6.1).
Other groups that would benefit from more horizontal industrial policies
that could stimulate growth and employment—the many young and educat-
ed who are unemployed,31 small businesses and young entrepreneurs who
seek to enter protected markets and have difficulty accessing finance, and
small farmers—have limited ability to unite and lobby the government for
change in industrial policy.
Table 6.1. Lobby Power of the Manufacturing Export Sector, 2001
Country/Region Lobby power of the manufacturing sector
Bahrain 0.27
Algeria 0.01
Egypt 0.18
Iran 0.03
Jordan 0.14
Lebanon 0.35
Libya 0.00
Morocco 0.36
Oman 0.03
Qatar 0.06
Saudi Arabia 0.02
Syria 0.05
Tunisia 0.48
Middle East and North Africa 0.13
Sub-Saharan Africa 0.19
East Asia and the Pacific 0.34
Europe and Central Asia 0.42
Latin America and the Caribbean 0.31
South Asia 0.52
High-Income/OECD 0.51
Source: Nabli, Keller, and Véganzonès (2002).

Breaking the Barriers to Higher Economic Growth154
To press for more horizontal policies, these groups require certain rights—
access to information to formulate choices, the ability to mobilize, the ability
to contest policies that are poor—which are only weakly present in the MENA
region. Government information is not fully accessible to the public. Activities
of the press are carefully monitored and freedom of the press circumscribed
in most countries. There are restrictions on civil society, and the ability to
contest government policies is weak. More generally, the MENA region suffers
from fundamental weaknesses in governance, in terms of inclusiveness, pub-
lic accountability, and strength of civil society. In a ranking of more than 142
countries according to some 19 indicators of governance ( in terms of both
the quality of public administration and public sector accountability), the
MENA region on average ranked in the bottom third of countries
worldwide—lower than every other region of the world (see table 6A.6). This
has hindered the development of coalitions for change and helped perpetuate
ineffective policies. Analysis suggests that the existing political economy equi-
librium favors the status quo of little or no public accountability and main-
taining prevailing economic/industrial policies and networks of privilege. In
such a situation industrial policy reforms are likely to come slow. Democratic
reform may be able to change this state of affairs and generate a great politi-
cal and economic transformation that could produce more democratic gov-
ernments, more effective policies and more economic growth in the region
(Nabli 2005).
The move toward horizontal policies is far from straightforward. It affects
the balance of power between actors in society; at its core it involves finding
the economic rents that have built up over the years and cutting them back;
and it attacks the economic privilege that some have enjoyed for generations.
Thus, it is hardly surprising that MENA has found moving toward a more
horizontal industrial policy a profoundly difficult task.
In many cases prereform economic elites have proven to be resilient to the
reform process even when policy reforms were designed to reduce their rent-
seeking opportunities—like those shifts aimed at moving toward horizontal
industrial policy.The existing sectoral interests are better able to preserve their
privileges, which leads to a passive industrial policy.32 The governance struc-
tures did not help new pressure groups to emerge or move toward more effec-
tive reforms.
Where Does MENA Go from Here?
The claim that industrial policy is a thing of the past is largely exaggerated.
Industrial policies continue to be used throughout the world, but their modal-
ities and focus have been changing to reflect the reality of the new global econ-
omy and rapid technological change, as well as to acknowledge the costly mis-
takes made in the past with traditional industrial policy.

Part I: Growth, Reform, and Governance 155
One of the greatest realizations is that vertical industrial policies almost
always breed dependency. An “infant industry” seldom feels it has grown up
and asks for government support to stop.33 Because of this, the modalities of
industrial policy have to change. At the same time, there is practically no coun-
try in the world without a foreign direct investment promotion policy or an
export-oriented focus—a clear signal that industrial policy is alive and well.
The challenge for MENA in the 21st century is to recognize that the kinds
of industrial policy needed in the current international setting are clearly dif-
ferent from the traditional forms of inward-looking, paternalist-state indus-
trialization strategies of the past. What MENA needs is not more industrial
policy but better,more sensible industrial policy. This may imply reducing the
scope of intervention in the region. But it also implies moving away from ver-
tical to horizontal policies, and from “choosing winners” (or more often “pro-
tecting losers”) to policies that make sense in the current global environment.
There is therefore a compelling argument for MENA to reconsider the types
of industrial policies it uses. The rapid growth in information and communi-
cation technologies, the acceleration of technical change and the intensifica-
tion of global integration require countries to focus on efficiency gains in the
value chain. Economic development is increasingly linked to the economy’s
ability to acquire and adapt new technical and socioeconomic knowledge.
Comparative advantages are coming less from abundant natural resources or
cheaper labor, and more from technical innovations and the competitive use
of knowledge. Moreover, the speed with which economies are able to dissem-
inate and apply knowledge increasingly determines their level of competitive-
ness and their chances of succeeding in the global arena.
MENA countries, like other developing countries in the world, operate in
an environment that is much different from two or three decades ago. The
global crises of the 1980s and 1990s led to the tendency to impose greater dis-
cipline in national economies. External mechanisms, including multilateral,
regional, and bilateral agreements, structural adjustment programs, and
political pressure from the industrialized world, have all reduced the scope of
industrial policies that MENA countries can use. Developing countries are
members of a number of international organizations and have signed agree-
ments that limit the countries’ capacity to use distortionary policies to pro-
mote particular sectors. For example, the new formal trade rules under the
auspices of the WTO limit selective interventions in trade; Basel core bank-
ing principles restrict direct lending; articles of agreement with internation-
al financial institutions impose market-driven procompetition policies in a
large array of areas, eliminating or reducing food subsidies, trade distortions,
restrictions on FDI, and regulating the use of monetary, foreign exchange,
and fiscal policies; free trade agreements eliminate tariff barriers and nontar-
iff barriers, enforce intellectual property rights, and regulate rules of origin
and services; and codes of conduct increase transparency and accountability.

Breaking the Barriers to Higher Economic Growth156
All these external pressures are shaping the changes in MENA’s industrial
policies.
Internal forces are also playing a role in reshaping MENA’s industrial poli-
cies. Although the region has not suffered from the deep crises that forced
other developing regions in the world to move rapidly out of their tradition-
al industrial policies, the picture in the MENA region is not homogeneous.
Wealth has allowed the oil-rich economies of the region to slow down the pace
of reforms. However, in countries like Jordan, Morocco, and Tunisia, reforms
have moved faster. Other internal forces are also putting pressure on the sta-
tus quo. The growth and employment challenges that MENA countries are
facing—in light of the region’s demographic transition—are rapidly revealing
the weaknesses of MENA’s economic model. Industrial policies are failing to
generate the promised results, and the social contract is not being honored.
Several factors are intensifying the need for reform, including labor market
pressures, rising expectations for improved standards of living, and the need
for efficient production models that will allow the region to deal with compe-
tition from world markets.
In addition, the MENA region is facing sociopolitical pressures to improve
inclusiveness and accountability, as well as to increase the transparency and
contestability of public policy. These forces are manifested in public demands
for greater individual and social freedoms, more democratization through
open and fair electoral processes, greater female participation in the social
sphere, better public services, and enhanced public sector governance. These
internal forces are also shaping the path of MENA’s reforms and the future of
its industrial policies.
But how is MENA going to move from the policies of the past to the poli-
cies of the future? The processes that will determine the path of change are
likely to depend on each country’s initial conditions and individual political
economy factors. All roads may lead to Rome, but which road each MENA
economy takes is an open question. The priorities for change in policies,
industrial or otherwise, vary with resource endowments, governance institu-
tions, and reform progress to date, as will the paths to the target.
From the political economy point of view, differences in economic per-
formance across countries can be explained by imperfections in electoral
markets—such as uninformed voters, noncredible political competitors, and
social polarization—which make it difficult for citizens to hold politicians
accountable for policies and outcomes. These imperfections offer powerful
insights into the development of particular (industrial) policies and the
underperformance of many economies.
Clientelism is a dominant characteristic of public policy in many countries.
One explanation for its existence derives from the struggle to make credible
promises to citizens. Patron-client relations drive politicians to focus on tar-
geted favors and goods rather than broad public goods and effective public

Part I: Growth, Reform, and Governance 157
policy: insofar as only clients believe patron promises (given the absence of
well-developed political parties, for example), political competition mainly
concerns targeted transfers to clients rather than public policy issues more
generally (World Bank 2005c).
Given the complexity of MENA’s political economy, political market incen-
tives must be changed in favor of better policies. One avenue is through gov-
ernance reform. The imperfections in political markets can be reduced, but
require improving the availability of information, transparency, checks and
balances, accountability, space for civil societies, and contestability of the
political market. These elements are needed for improved governance and
could increase the likelihood of effective design and implementation of pub-
lic policies, including industrial policies, and that could lead to better eco-
nomic outcomes for the region.
Notes
1. While the term “industrial policy” was originally used to describe policies relating specifically to the
industrial sector, today the term has become more broadly recognized to include policies that encour-
age any sector (for example, agriculture or tourism) not only the industrial sector.
2. World Bank (1993).
3. Noland and Pack (2003, 2005).
4. Rodrik (2004).
5. As it turns out, the country found its niche in sunrise industries with high knowledge content.
However, the best intentions with industrial policy have often produced the most disastrous outcomes.
6. Borensztein and Lee (2005) and Pack (2000).
7. World Bank (1975).
8. See, for example, Olson (1971, 1982) on collective action and rent-seeking behavior, North (1990)
on economic institutionalism, Waterbury (1993) on public enterprise reform, and Hellman (1998) on
“winners take all.”
9. World Bank (1980).
10. This led Heydemann (2004) to propose a new approach to the political economy of policy reform
that focuses on the so-called networks of privilege.
11. World Bank (2003a).
12. Including capital flows and labor remittances.
13. It is important to note that while socialist ideas had an important influence on the economic model
implemented in the region, there is significant heterogeneity among countries and the economic
polices they used in the past and continue to use today.
14. This was not a MENA-specific problem. Other regions that adopted these policies (for example,,
Latin America and the Caribbean) experienced this problem as the domestic industry had less access-
but particularly less incentive—to implement the latest technologies and management practices.
15. Auty (2004).
16. Muller-Jentsch (2005).
17. Nabli, Keller, and Véganzonès (2002).
18. Galal and El-Megharbel (2005).
19. World Bank (2003b).

Breaking the Barriers to Higher Economic Growth158
20. World Bank (2003c).
21. Measured by unweighted average tariffs.
22. World Bank (2005b).
23. Ibid.
24. Jbili, Enders, and Treichel (1997).
25. Iran in the early 2000s.
26. Schwartz and Clements (1999).
27. Krueger (1993).
28. Koromzay (2004).
29. Some countries in the region, like Jordan, have experienced deep economic crisis and as a result
have moved forward the reform agenda at a faster pace than the region’s average.
30. Nabli, Keller, and Veganzones (2002). Lobby power is measured as an interactive between the size
of manufacturing exports in total exports, and the share of manufacturing exports among the top four
export categories at the three-digit ISIC.
31. Unemployment in the region, which averages about 15 percent of the labor force, is about two-
three times higher for those under the age of 30.
32. As mentioned by Heydemann (2004), reforms in the MENA region as well as in other parts of the
developing world have provided opportunities for the networks of privilege to survive. The elites con-
tinue to be elites in their sectors even after reform. In other regions privatization processes are good
examples of where the networks of privilege exert their influence.
33. Krueger (1993).
References
Auty,Richard M. 2004. “Economic and Political Reform of Distorted Oil-Exporting Economies.” Paper
prepared for the workshop “Escaping the Resource Curse: Managing Natural Resource Revenues
in Low-Income Countries,” held at the Center for Globalization and Sustainable Development,
Earth Institute at Columbia University. February 26.
Borensztein, Eduardo, and Jong-Wha Lee. 2005. “Financial Reform and the Efficiency of Credit
Allocation in Korea.” Journal of Policy Reform 8 (1): 55–68.
Creane, Susan, Rishi Goyal, A. Mushfiq Mobarak, and Randa Sab. 2004.“Financial Development in the
Middle East and North Africa.” IMF Working Paper 04/201. Washington, DC: International
Monetary Fund.
FAO (Food and Agriculture Organization). 2003. “Syrian Agriculture at the Crossroads.” Agricultural
Policy and Economic Development 8. Rome, Italy: FAO.
Galal, Ahmed, and Nihal El-Megharbel. 2005. “Do Governments Pick Winners or Losers? An
Assessment of Industrial Policy in Egypt.”Working Paper 108. Cairo, Egypt: The Egyptian Center
for Economic Studies.
Hellman, Joel S. 1998.“Winners’ Take: The Politics of Partial Reform in Postcommunist Transitions.”
World Politics 50 (January): 203–234.
Heydemann, Steve. 2004. Networks of Privilege in the Middle East: The Politics of Economic Reform
Revisited. New York: Palgrave Macmillan.
International Monetary Fund (IMF). 2005. Annual Report on Exchange Arrangements and Exchange
Restrictions. Washington, DC: International Monetary Fund.
Jbili, Abdelali, Klaus Enders, and Volker Treichel. 1997. “Financial Sector Reforms in Morocco and
Tunisia.” Finance and Development IMF Working Paper 97/81. Washington, DC: International
Monetary Fund.

Part I: Growth, Reform, and Governance 159
Koromzay, Val. 2004.“Some Reflections on the Political Economy of Reform.” Paper presented at inter-
national conference on “Economic Reforms for Europe: Growth Opportunities in an Enlarged
European Union.” Bratislava, Slovakia. March 18.
Krueger, Anne. 1993. Political Economy of Policy Reform in Developing Economies. Cambridge, MA:
MIT Press.
Muller-Jentsch, Daniel. 2005. “Deeper Integration in Trade Services in the Euro-Mediterranean
Region: Southern Dimensions of the European Neighborhood Policy.” Draft working paper.
Washington, DC: World Bank.
Nabli, Mustapha K., Jennifer Keller, and Marie-Ange Veganzones. 2002.“Exchange Rate Management
within the Middle East and North Africa Region: The Cost to Manufacturing Competitiveness.”
Working Paper 2004:1. Beirut: American University of Beirut.
Nabli, Mustapha K. 2005. “Restarting the Arab Economic Reform Agenda.” The Arab World
Competitiveness Report 2005. World Economic Forum. Houndmills, U.K.: Palgrave Macmillan.
Noland, Marcus,and Howard Pack. 2003. Industrial Policy in an Era of Globalization: Lessons from Asia.
Washington, DC: Institute for International Economics.
———. 2005. The East Asian Industrial Policy Experience: Implications for the Middle East. Washington,
DC: Institute for International Economics.
North, Douglass C. 1990. Institutions, Institutional Change and Economic Performance.New York:
Cambridge University Press.
Oliva, Maria Angels. 2000. “Estimation of Trade Protection in Middle East and North African
Countries.” IMF Working Paper 00/27. Washington, DC: International Monetary Fund.
Olson, Mancur. 1971. The Logic of Collective Action: Public Goods and the Theory of Groups. Cambridge,
MA: Harvard University Press.
———. 1982. The Rise and Decline of Nations.Economic Growth, Stagflation, and Social Rigidities.New
Haven, CN: Yale University Press.
Pack, Howard. 2000. “Industrial Policy: Growth Elixir or Poison?” The World Bank Research Observer
15 (1): 47–67. Washington, DC: World Bank.
Rodrik, Dani. 2004. “Industrial Policy for the Twenty-first Century.” Unpublished paper. Harvard
University, Cambridge, MA. Available at: http://ksghome.harvard.edu/~drodrik/UNIDOSep.pdf
Schwartz, Gerd, and Benedict Clements.1999. “Government Subsidies.” Journal of Economic Surveys 13
(2): 119–147.
UNCTAD (United Nations Conference on Trade and Development). 2005. Trade Analysis and
Information System (TRAINS) database. Available at: http://r0.unctad.org/trains/
Waterbury, John. 1993. Exposed to Innumerable Delusions: Public Enterprise and State Power in Egypt,
India, Mexico and Turkey. New York: Cambridge University Press.
World Bank. 1966. Tunisia—Second Societe Nationale d’Investissement Project. Staff appraisal report.
Washington, DC: World Bank.
———. 1969. Tunisia—Third Societe Nationale d’Investissement Project. Staff appraisal report.
Washington, DC: World Bank.
———. 1971. Tunisia—Fourth Societe Nationale d’Investissement Project. Staff appraisal report.
Washington, DC: World Bank.
———. 1972. Tunisia—Fifth Societe Nationale d’Investissement Project. Staff appraisal report.
Washington, DC: World Bank.
———. 1975. Algeria—Industrial Credit Project. Staff appraisal report. Washington, DC: World Bank.
———. 1990. Egypt—Tourism Project. Project completion report. Washington, DC: World Bank.
———. 1993. The East Asia Miracle: Economic Growth and Public Policy. New York: Oxford University
Press.

Breaking the Barriers to Higher Economic Growth160
———. 2003a. Unlocking the Employment Potential in the Middle East and North Africa: Toward a New
Social Contract. MENA Development Report. Washington, DC: World Bank.
———. 2003b. Trade, Investment, and Development in the Middle East and North Africa: Engaging with
the World. MENA Development Report. Washington, DC: World Bank.
———. 2003c. Jobs, Growth and Governance in the Middle East and North Africa. MENA Development
Report. Washington, DC: World Bank.
———. 2003d. Knowledge Economies in the Middle East and North Africa. Toward New Development
Strategies. Washington, DC: World Bank.
———. 2003e. “Iran: Medium Term Framework for Transition. Converting Oil Wealth to
Development.” Country economic memorandum. Washington, DC: World Bank.
———. 2005a. “Technology and Growth Series: Chilean Salmon Exports.” PREM Notes 103.
Washington, DC: World Bank.
———. 2005b. MENA Economic Developments and Prospects 2005: Oil Booms and Revenue
Management. Washington, DC: World Bank.
———. 2005c. Economic Growth in the 1990s: Learning from a Decade of Reform. Washington, DC:
World Bank.
World Economic Forum. 2003. Arab World Competitiveness Report 2002–2003.

Part I: Growth, Reform, and Governance 161
Data Appendix
Figure 6A.1. Average Nominal Tariffs by Region, 1980–85
⫺10 0 10 20 30 40 50 60 70 80
South Asia
Latin America and the
Caribbean
East Asia and Pacic
Sub-Saharan Africa
MENA
Europe and Central Asia
industrialized economies
percent
Source: World Economic Forum, Arab World Competitiveness Report 2002–2003.
Figure 6A.2. Cash Transfers to Industries (% of GDP)
2.59
3.4
1.53
1.74 1.46 1.24
4.12
12.02
2.48
3.57
1.61 1.14
2.68
8.44
0
2
4
6
8
10
12
14
all EU developing
countries
Africa Asia MENA ECA
1975–82
1983–90
Note: Based on SNA database, national authorities, and authors’calculations. The category Developing Countries does not include Israel and
South Africa, although they are included in their geographical country groups. Here, subsidies are defined as cash transfers to industries.

Table 6A.1. Current Average Tariffs and Standard Deviation for Selected Countries
Average Standard Average Standard
tariff deviation tariff deviation
MENA High-Income
Algeria (2003) 18.63 10.37 European Union 4.4 5.5
Djibouti (2002) 30.71 7.65 Japan (2004) 3.13 7.06
Egypt (2002) 19.88 16.19 United States (2004) 3.7 11.27
Iran (2004) 19.88 20.14 Africa
Jordan (2003) 13.7 16.1 Nigeria 27.21 23.51
Lebanon (2002) 6 10.75 South Africa (2001) 8.15 11.62
Morocco (2003) 28.87 20.25 Asia (EAP + SA)
Saudi Arabia (2004) 6.11 6.65 China (2004) 10.46 7.92
Syria (2002) 14.64 21.6 India (2004) 29.02 13.79
Tunisia (2004) 25.96 26.81 Indonesia (2004) 7.05 15.53
Yemen (2000) 12.78 6.06 Malaysia (2003) 8.48 21.07
LAC Europe and Central Asia
Argentina (2004) 12.64 7.58 Hungary (2002) 9.5 13.49
Bolivia (2004) 9.28 2.5 Poland (2003) 13.39 32.14
Brazil (2004) 13.57 6.79 Russian Federation (2002) 9.62 5.13
Chile (2004) 5.99 0.71 Turkey (2003) 7.11 20.22
Mexico (2004) 17.43 14.99
Source: UNCTAT (2005).
Table 6A.2. Ranking of Nontariff Measures, 1999
Low Intermediate High
incidence incidence incidence
Algeria Syria
Djibouti Egypt Iran
Qatar Jordan Libya
U.A.E. Kuwait
Lebanon
Morocco
Oman
Saudi Arabia
Tunisia
Source: Oliva (2000), according to IMF classification.
Breaking the Barriers to Higher Economic Growth162

Part I: Growth, Reform, and Governance 163
Table 6A.3. Overview of Exchange Rate Systems in MENA Countries
Country Exchange rate system Exchange rate structure Exchange tax/subsidy
Algeria Managed Float Unified No/No
Djibouti Currency Board Unified No/No
Egypt Managed Float Unified No/No
Iran Managed Float Unified No/No
Iraq Peg Unified No/No
Jordan Peg Unified No/No
Lebanon Peg Unified No/No
Libya Peg Unified No/No
Morocco Peg Unified No/No
Saudi Arabia Peg Unified No/No
Syria Peg Multiple No/No
Tunisia Managed Float Unified No/No
UAE Peg Unified No/No
Yemen Independent Float Unified No/No
Source: IMF (2005).
Table 6A.4. Overview of Interest Rates and Credit Controls in MENA
Interest rate All credit
Country liberalized? controls removed? Notes
Algeria Yes, de jure. Yes, de jure. Ceilings Interest rates were liberalized in 1990 (as well as deposit rates
removed in 2000. and ceilings on lending rates).
Bahrain Yes Yes.
Djibouti Yes Yes
Egypt Yes, de jure. Yes Interest rates were liberalized in 1991.
Iran No No Through the lending instruments of the state-owned banks,
credit allocation is decisively used to support certain sectors in
Iran’s economy. The lending instruments that banks can use are
governed by a 1982 law on Islamic banking with the rates of
return on both loans and deposits set by the central bank, gen-
erally less than the rate of inflation over the last decade. The con-
trolled lending rates vary with the term and the sector receiving
the loan. Every large loan must be approved by the central bank,
which also sets the minimum percentages of each bank’s loan
portfolio. The preferred sectors are, roughly speaking, agriculture
and housing before export, industry, and trade and services.
Jordan Yes Largely. Preferential In the early 1990s, Jordan fully liberalized interest rates. In 1993,
credit facilities remain the Central Bank of Jordan moved away from direct instruments
for agriculture, handi- of monetary control by issuing its own certificates of deposit to
crafts and export sectors mop up excess liquidity. In 1996, the central bank’s rediscount
subsidies and preferential credit facilities were eliminated, except
for small specialized banks that extended credit to the agricul
tural, handicrafts, and export sectors.
Lebanon Yes Yes
Morocco Yes Yes Steady steps of liberalization and elimination of credit subsidies
since the1980s. Interest rates liberalized in 1991, full liberalization
of ceilings, etc. by 1996.
Qatar Yes Yes Specialized banks offer subsidized loans to small companies.
Saudi Arabia Yes Yes
(Table continues on the following page.)

Breaking the Barriers to Higher Economic Growth164
Table 6A.4. Overview of Interest Rates and Credit Controls in MENA (continued)
Interest rate All credit
Country liberalized? controls removed? Notes
Syria No No Until interest rates were adjusted in 2003; the bank’s discount
interest had remained unchanged at 7-9 percent since 1981
(7 percent for the public sector) irrespective of liquidity condi-
tions or inflation. As a result, real interest rates were negative in
times of high inflation (over much of the 1980s and until 1995,
when inflation averaged 19 percent), and very high for much of
the rest of time, particularly in the late 1990s when prices were
contracting. Lending priority is given to the public sector, with
many loans often insufficiently serviced by public institutions.
Private companies often find it difficult to obtain loans through
the banks, and resort to the unofficial market (or offshore banks)
where rates are often as high as 20 percent.
Tunisia Partial. Some Yes, de jure. However, Gradual liberalization except for interest rates on lending in pri-
deposit rates lending is still en- ority sectors. In 1987, interest rates on short term deposits were
remain couraged to certain liberalized. Lending rates, except for those to priority sectors,
regulated. sectors through were allowed to be set freely within a set of 3 percentage points
preferential access. above TMM. In 1990, preferential rates for all priority sectors were
increased, albeit only moderately for agriculture. In 1994 and
1996, interest rates were liberalized for all sectors.
UAE Yes Yes.
Yemen Partial A minimum benchmark
rate for saving deposits
is set administratively.
Source: Creane et al. (2004).
Note: Authors’notes based on various World Bank resources.
Table 6A.5. Share of Domestic Credit Directed to the Private Sector, 2003
Share of Share of
domestic domestic
Country credit (%) Region credit (%)
Algeria 11.5 Middle East and North Africa 55.3
Iran 35.4 Latin America and the Caribbean 30.5
Syria 10.1 Europe and Central Asia 45.6
Yemen 6.9 Sub-Saharan Africa 73.8
Libya 18.0 High-income/OECD 123.0
Egypt 61.5 East Asia and the Pacific 137.9
Jordan 71.7
Lebanon 82.0
Morocco 55.1
Tunisia 69.0
Saudi Arabia 58.2
United Arab Emirates 55.9
Bahrain 65.4
Oman 38.6
Kuwait 73.8
Qatar 30.5
South Asia 31.7
Source: World Bank World Development Indicators online (August, 2005).
Note: Regional averages weighted by total domestic credit.

Table 6A.6. Governance Indicators in MENA
Quality of Public sector Overall
Country public administration accountability governance
Algeria 44.1 42.3 42.5
Bahrain .. 37.2 ..
Djibouti .. .. ..
Egypt 46.1 38.3 40.9
Iran 37.2 36.5 35.9
Iraq .. .. ..
Jordan 54.7 53.1 53.1
Kuwait 51.1 44.8 47.2
Lebanon .. .. ..
Libya 34.5 7.7 17.5
Morocco 57.8 47.9 51.1
Oman 51.7 32.8 40.6
Qatar .. 31.0 ..
Saudi Arabia 50.8 21.0 32.6
Syria 37.0 25.8 29.7
Tunisia 58.4 37.1 44.3
United Arab Emirates 51.1 33.6 40.8
West Bank Gaza .. .. ..
Yemen, Republic 40.6 34.8 36.5
MENA 47.3 35.1 39.4
Sub-Saharan Africa 42.6 55.1 49.1
East Asia and the Pacific 43.1 57.3 50.6
Europe and Central Asia 49.0 69.5 60.3
Latin America and Caribbean 47.4 75.7 63.3
OECD 79.6 97.5 90.2
South Asia 47.2 58.2 52.9
LMIC (excluding MENA) 45.7 61.1 53.9
Source: World Bank staff calculations.
Notes: Governance indices range from 0 to 100; higher values reflect better governance standing compared with other countries. Regional and
subregional aggregates are simple averages of relevant country values. The indices are constructed using principal component analysis (PCA), an
aggregation technique designed to linearly transform a set of interrelated variables into a new set of uncorrelated principal components that
account for all the variance in the original variables. LMIC refers to low- and middle-income countries.
Part I: Growth, Reform, and Governance 165

Part II
Labor Markets and
Human Capital

169
The Macroeconomics of Labor
Market Outcomes in MENA
How Growth Has Failed to Keep Pace with a
Burgeoning Labor Market
Jennifer Keller*
Mustapha K. Nabli
7
Perhaps the greatest single development issue facing the economies of the
Middle East and North Africa (MENA)1is the challenge of employing its peo-
ple in good jobs. While the region is heterogeneous in terms of developments
in the labor market, the majority of the region has been characterized by high
levels of unemployment, and in some cases by declining real wages, as well.
The problem of job creation for the MENA region is enormous. A 2003 World
Bank study on the region estimated that between 2000 and 2020, some 100
million new jobs will need to be created within the MENA countries, more
than doubling the number of jobs in existence, if the region is to both absorb
future labor force entrants and eliminate current unemployment.2To g i ve
some perspective on this challenge—a 100 percent increase in the number of
jobs over two decades—one need only look at some of the countries in East
Asia, which managed the highest rates of sustained employment growth in
modern history. In China, employment growth over the two decades begin-
ning in the late 1970s averaged 40 percent. Korea’s employment growth over
the 1980s and 1990s was slightly less than 60 percent. Malaysia’s employment
*World Bank. The authors are grateful to Sebastien Dessus, Carlos Silva-Jauregui, Dipak Dasgupta, and
Marie-Ange Véganzonès for their comments and suggestions.
Earlier version published as Egyptian Center for Economic Studies Working Paper 77, August 2002.

Breaking the Barriers to Higher Economic Growth170
growth was highest in East Asia over the 1980s and 1990s, with a 90 percent
increase in jobs.3These countries have experienced some of the greatest spurts
in employment growth over a sustained period of time. And employment
growth in the MENA region needs to exceed this pace.
Unemployment rates in the region are among the highest in the world,
averaging 14.0 percent of the labor force.4The problem of unemployment
affects virtually every country in the region, even several oil-exporting Gulf
economies that traditionally had to import expatriate laborers to supplement
the national work force. In a few countries, the unemployment rate reaches
close to 20 percent or higher,including Algeria (23.75percent), Morocco (19.3
percent), and the West Bank and Gaza (25.6 percent).
These high levels of unemployment imply a substantial loss of human cap-
ital to the economy. The MENA region has made considerable progress over
the last decades in increasing access to basic education. The educational
attainment of the adult population in MENA increased by an average of more
than 5 percent a year between 1960 and 1990,6higher than any other region
of the world. But just as these human capital achievements should be having
their greatest payoff, in terms of economic growth, a considerable portion of
these resources are left idle.
Much of the story behind the MENA region’s lost decade of growth, and
the consequences on the labor market, is understood. Declining oil prices had
a major impact on the region, both for the oil-exporting nations and for much
of the region, through the impact on remittances and external financial flows.
Additionally, the region was marked by macroeconomic instability and struc-
tural inefficiencies that prevented the emergence of a strong private sector. At
least half of the MENA economies suffered from some degree of macroeco-
nomic instability during the 1985-95 period. Public sector ownership was
extensive, yet while large investments were taking place with the oil windfalls,
there were few policies in place to make these investments competitive. Trade
regimes were protective.Regulation limited the entry of the private sector into
most sectors. Financial sectors were geared to serving public enterprises, and
institutions were not in place to facilitate a vibrant private sector. As a conse-
quence, when oil prices collapsed, the engine for growth in the economies of
MENA stalled, and there was limited ability to absorb the burgeoning labor
force.
What was of far greater concern in MENA was that despite macroeconom-
ic stabilization, and at least some structural reforms undertaken throughout
most of the region over the late 1980s and early 1990s, a strong economic
recovery remained elusive. GDP per capita growth over the region averaged
1.6 percent a year over the 1990s, higher than the anemic growth over the
1980s (averaging only 0.4 percent a year7), but hardly the rebound one would
have desired following a decade of stagnation. Outside the Gulf economies,
growth averaged a slightly higher 1.8 percent per year, far from robust. As a

Part II: Labor Markets and Human Capital 171
result, following two decades of poor or lackluster economic performance, by
2000 the MENA region faced unemployment rates averaging 15 percent of the
labor force, higher than in every other region in the world but Sub-Saharan
Africa. Improving labor market opportunities has become among the highest
priorities for policy makers in the region.
In this paper, we analyze the major labor market trends that developed in
the MENA region over the 1990s. We then examine the failure of growth to
materialize following widespread structural reform throughout the region. By
decomposing growth over the 1990s between factor accumulation and pro-
ductivity growth, we find that productivity growth improved for the majority
of countries in the region over the 1990s, with an average increase in TFP
growth of 1.4 percent from the 1980s. At the same time, despite these positive
improvements in productivity in MENA, economic growth remained anemic
(with average growth in GDP per laborer over the 1990s virtually unchanged
from the 1980s), in great part because of the collapse in investment that has
occurred in virtually every economy in the region.
We then offer some reasons why private sector investment did not materi-
alize as dynamically as hoped, despite widespread macroeconomic and struc-
tural policy reforms instituted throughout the region in the early part of the
decade. Despite the region’s achievements over the 1990s in terms of macro-
economic stabilization and policy reform, MENA’s progress with structural
reform has been incomplete. Financial sectors remain weak. Trade liberaliza-
tion remains incomplete, with continuing high protection levels. Public own-
ership remains high. And, the regulatory framework and supportive institu-
tions for private sector investment have not materialized.
Finally, we offer policy recommendations for improving labor market out-
comes. Pushing forward with more complex and politically challenging
“second-generation”reforms may be mandatory if the region is ever to ensure
the higher and sustainable economic growth that is needed to ensure better
labor market prospects in the region.
The Disappointing Labor Market Outcomes in MENA
According to national statistics, MENA’s unemployment rates are among the
highest in the world, averaging 14 percent of the labor force, second only to
Sub-Saharan Africa (figure 7.1).8In a number of countries in the region, such
as Algeria, Morocco, and the West Bank and Gaza, close to 20 percent or more
of the labor force are without jobs.
Unemployment in the region worsened throughout the 1990s, contribut-
ing to the high unemployment rates currently observed. Table 7.1 compares
labor force growth to employment growth over the 1990s. Because of discour-
aged workers leaving the labor force, it is not always possible to interpret
employment growth’s outpacing labor force growth as necessarily a reduction

Breaking the Barriers to Higher Economic Growth172
in “unemployment.” However, when the labor force growth exceeds the
growth of employment, it is indicative of worsening unemployment. From
that table, the rate of growth of the labor force exceeded the rate of growth of
employment in Algeria, Iran, Egypt, and Morocco, which together account for
approximately 70 percent of the entire region’s labor force, and in which cur-
rent unemployment rates now average 15 percent. In Tunisia and Jordan,
where the rate of unemployment was already moderately high, the rate of
growth of employment remained about on par with the rate of growth of the
labor force.
Worker productivity—which over the long term forms the basis for
increases in real wages—has generally increased throughout MENA, but
remains low by international standards (table 7.2). Over the last decade, the
growth of GDP per worker was lower in the MENA region than any other
region of the world but Europe and Central Asia (heavily influenced by
Russia), averaging only 0.5 percent a year. Productivity actually declined over
the 1990s in Algeria, Saudi Arabia, and Jordan.
Figure 7.1. Worldwide Comparison of Unemployment Rates, 2004a
0
2
4
6
8
10
12
14
16
18
20
percent of labor force
developing country average
East Asia
and Pacic
Europe
and
Central
Asia
High-Income
OECD
Latin
America
and the
Caribbean
Middle East
and
North Africa
South Asia Sub-Saharan
Africa
Source: World Bank data.
a. Or most recent year available.
Table 7.1. Labor Force Growth versus Employment Growth over the 1990s
Size (labor
Unemployment force as %
Labor force Employment (latest year of total regional
Country growth (%) growth (%) Time period available) labor force)
Algeria 3.8 2.2 1989–2000 23.7 11
Iran 2.2 1.9 1991–2000 13.2 21
Egypt 2.8 2.4 1988–2000 9.9 26
Morocco 2.5 1.8 1987–2000 19.3 12
Tunisia 2.8 2.8 1989–2000 14.3 4
Jordan 5.5 5.9 1990–2001 14.5 2
Source: World Bank staff estimates from country sources.

Part II: Labor Markets and Human Capital 173
Understanding the Poor Labor Market Outcomes in MENA
Under most comparisons, MENA’s labor market outcomes over the 1990s
were disappointing. Why weren’t enough jobs created? Why were laborers
who found jobs unable to watch their wages grow?
The simplest answer is that economic growth has been insufficient, given
the region’s labor force growth. Labor force growth in MENA is exceptional,
the result of both rapid population growth and increasing rates of labor mar-
ket participation (particularly for females). At an average rate of growth of 3
percent a year, MENA’s labor force is growing at a higher rate than in any other
region of the world.
At the same time, before the recent upturn in growth from rising oil prices,
the region’s labor force growth was barely matched by economic growth. High
labor force growth, of course, need not be an automatic recipe for poor labor
outcomes. It could very easily contribute to high GDP growth, as was the case
in East Asia during its high-growth years. But, in MENA, high labor force
growth rates have been accompanied by only marginal growth of real output.
In table 7.3, labor force growth rates and real GDP growth rates in East Asia
in the 1970s are compared with labor force growth rates and real GDP growth
Table 7.2. Employment, Growth, and Productivity Growth in MENA versus Other Regions,
1990–2004a
Employment GDP growth Worker productivity
Country growth (%) (%) Time period growth (%)
Algeria 4.2 2.5 1990–2004 ⫺1.6
Iran 3.0 4.2 1990–2002 1.2
Egypt 2.9 4.3 1992–2003 1.5
Morocco 2.1 2.6 1990–2001 0.5
Tunisia 3.0 5.0 1989–2000 2.0
Jordan 5.9 5.1 1990–2001 ⫺0.7
Syria 3.7 4.6 1991–2002 0.8
Saudi Arabia 5..7 2.7 1990–2000 ⫺2.0
MENA 0.5
East Asia and Pacific
(including China) 8.0
EAP (excluding China) 2.5
Latin America and
the Caribbean 0.8
Europe and Central Asia 0.2
South Asia 4.5
High-Income/OECD 1.5
World (excluding China) 2.4
Source: MENA employment growth: World Bank staff estimates from country sources; GDP: World Bank World
Development Indicators; Employment outside of MENA economies: World Development Indicators, International
Monetary Fund International Financial Statistics online.
Note: Worker productivity growth calculated as log growth of GDP/laborer for the period noted.
a. Or closest year available.

Breaking the Barriers to Higher Economic Growth174
rates of MENA economies in the 1990s. There is no difference between the
two regions’ labor force growth—both were exceptionally high, at 3.1 percent
a year, on average.
The real difference between the regions, of course, is that East Asia’s labor
force growth was accompanied by enormous increases in real output not wit-
nessed in the MENA economies. Real GDP growth in East Asia averaged 7.8
percent a year between 1970 and 1980—more than double its labor force
growth rate for the same period. In MENA, in comparison, economic growth
during the 1990s has only averaged about 3.6 percent a year —only marginal-
ly higher than the growth rate of its labor force, and implying virtual stagna-
tion in productivity per potential laborer for the region as a whole.
To better understand the importance of the growth of output per laborer
in improving labor market outcomes, we can refer to the simple accounting
framework below. Creating employment for those who want to work is equiv-
alent to increasing the ratio of employed persons to the total labor force (c).
Increasing productivity (the basis for wage growth, at least over the long term)
is equivalent to increasing output per employed person (b). The sum of these
two objectives results in growth in output per laborer (a). The higher the real
Table 7.3. Labor Force Growth and Real GDP Growth
MENA 1990s versus East Asia in the 1970s (%)
Labor force GDP GDP per
Country/Region growth growth worker growth
MENA, 1990–2000 3.1 3.6 0.4
Algeria 4.1 1.7 ⫺2.3
Iran 2.9 4.0 1.1
Syria 4.3 5.1 0.8
Yemen 4.0 5.5 1.5
Egypt 2.9 4.5 1.6
Jordan 7.0 5.1 ⫺1.8
Lebanon 3.2 10.3 6.9
Morocco 2.4 2.2 ⫺0.2
Tunisia 3.1 4.7 1.6
Bahrain 3.5 4.6 1.1
Kuwait 0.4 3.4 3.1
Oman 4.2 4.6 0.4
Saudi Arabia 2.3 2.7 0.5
United Arab Emirates 6.3 3.2 ⫺2.9
East Asia, 1970–80 3.1 7.8 4.6
Hong Kong 4.2 9.3 4.9
Indonesia 2.9 7.9 4.9
Korea 3.1 7.2 4.0
Malaysia 3.5 7.8 4.1
Singapore 4.4 8.9 4.3
Thailand 3.4 6.9 3.4
Source: World Bank, World Development Indicators.

Part II: Labor Markets and Human Capital 175
output per laborer growth, in turn, the greater is the scope for the economy to
either reduce unemployment and/or increase productivity (and wages). In
short, output per laborer growth provides a snapshot of the labor market out-
comes that will arise.9Strong growth means that there is room for both
unemployment reduction and wage increases. In MENA, output per laborer
growth has been only 0.4 percent per year on an average basis. As a result,
almost any reductions in unemployment have had to come at the expense of
wages. There has been limited scope for simultaneously lowering unemploy-
ment and realizing real wage increases.
Output per laborer has grown in MENA at an average annual rate of only 0.4
percent a year, with actual deteriorations in output per laborer in Algeria,
Jordan, Morocco, and the United Arab Emirates. In only five countries—Yemen,
Egypt, Lebanon, Tunisia, and Kuwait—did output growth per laborer exceed
1.5 percent a year (and the strong growth in output experienced in Lebanon was
primarily the result of massive reconstruction efforts that took place following
the 15-year civil war; the strong growth in output per laborer in Kuwait, mean-
while, reflects the growth rebound following the first Gulf War).
Does high growth guarantee good labor market outcomes? No. It is possi-
ble that employment problems will still persist with high economic growth, if
that growth is primarily capital-intensive (rather than employment-
intensive). Looking at the MENA economies, however, there does not appear
to be an issue with past growth being employment-unfriendly. On the con-
trary, for the countries in which there exist both employment growth and eco-
nomic growth estimations (table 7.2), the employment elasticity of output
growth averaged 0.7 over the 1990s. As a comparison, during the height of
their employment creation in the high-performing East Asian economies, the
employment elasticity of growth rarely exceeded 0.6. The process by which
output growth in MENA has led to employment (or more accurately, with
which employment has expanded strongly despite low levels of growth) is a
reflection of the nature of employment creation in the region, where public
sector employment has been used to as a refuge for large portions of the labor
force. While this type of employment creation is unlikely to be sustainable
over the longer term (and employment will inevitably have to emerge from
the private sector), there is still little evidence that the MENA region’s growth
has a poor employment-generating capacity.
But more important, what is clear is that employment cannot emerge with-
out growth. High employment growth cannot coexist over a sustainable peri-
od with low levels of economic growth. Paramount to improving the region’s
labor market outcomes, then, is improving the region’s growth prospects.
Output Output Employment
Labor Force Employment Labor Force
Growth = Growth + Growth
(a) = (b) + (c)

Breaking the Barriers to Higher Economic Growth176
In the end, policy makers should have two basic goals for what happens in
labor markets: (i) that those who want to work can find work, and (ii) that
wages increase. In MENA, lack of growth of output per laborer has prevented
both goals from transpiring simultaneously in the majority of countries. If
one goal has been achieved (such as a reduction in unemployment), it has had
to come at the expense of the other (real wage loss).
The story of employment outcomes in MENA is clear from an arithmetic
standpoint: Output growth has been insufficient. With output growth just
keeping pace with growth in the labor force, it is impossible to achieve simul-
taneous objectives of growth in wages and reduction in unemployment.
Within MENA, that tradeoff is apparent—the region has, as a whole, experi-
enced slight or no reductions in unemployment rates over the last decade, but
output per worker has declined as well. If the region wants to achieve both
higher employment growth and higher wages, much higher output growth will
be required. It is well established,and backed by a wealth of empirical evidence,
that rapid output growth brings with it rapid growth in employment. Periods
of buoyant GDP expansion are almost invariably associated with rising job
numbers while, conversely, slowdowns bring growing unemployment.10
What Explains MENA’s Poor Growth Performance?
Over the last decade, MENA countries took a number of steps to overcome the
macroeconomic imbalances and structural impediments that prevailed
throughout the 1980s. Starting in the late 1980s, several countries in the
region—Morocco and Tunisia, and soon after, Jordan and Egypt—embarked
on extensive programs of macroeconomic stabilization and policy reform. By
the 1990s, nearly all of the non-Cooperation Council for the Arab States of the
Gulf (GCC) countries in the region followed suit, as did several of the Gulf
economies. While there was considerable variance among economies in terms
of both the speed and depth of these reforms, the overall change in policy
throughout the region would seem to have been a significant step forward in
creating an environment in which the private sector could emerge and
become an engine for higher and sustainable growth. Despite this, strong
growth failed to emerge.
In order to understand why, we have examined the region’s economic
growth in a growth accounting framework, in which economic growth occurs
as the result of factor accumulation (either physical or human) and increases
in total factor productivity (see Annex for methodology and description of
the data).
Total factor productivity (TFP) growth is something of a mixed bag. It is
the residual of what cannot be explained by investments, if we assume those
investments (both physical and human) earn a reasonable rate of return. TFP
growth is often thought of as “technical progress,” but in fact, as the residual

Part II: Labor Markets and Human Capital 177
of a growth-accounting estimation, it not only embodies the differences
across countries in their progress in the adoption of better technology, but
also reflects a host of nontechnological differences, including changes in the
use of both capital and labor, changes in schooling quality, and changes in the
overall efficiency with which factors are allocated in the production process.
Our interest is to explore how MENA’s overall growth has improved or dete-
riorated since it began its structural reform process, to better understand what
has prevented the region from achieving the rates of growth needed to
improve its labor market outcomes.
In the MENA region, accumulation and productivity have often gone in
opposite directions, such as during the period of massive public sector invest-
ments. Examining growth alone will mask these very different effects, and the
somewhat anemic growth that has characterized the region since reform may
be more a reflection of significantly lower investments than of continuing
poor productivity performance.
In table 7.4, estimates of total factor productivity growth over the 1960-
2000 period are presented by region and decade. TFP growth has been calcu-
lated as the simple residual between output growth and the growth of factor
inputs (capital and labor), assuming those factors earn a reasonable rate of
return.11 From that table, MENA economies exhibited a pattern of high TFP
growth in the 1960s, declining dramatically over the 1970s and continuing to
decline throughout the 1980s.12 Understanding these developments, however,
requires a more detailed look at growth, accumulation, and productivity.
In the 1960s, MENA’s economic growth performance was the highest in the
world, averaging 6.3 percent per year (4.2 percent per year per laborer).
Beginning in the 1960s, the region began a two-decade period of massive pub-
lic investment in infrastructure, health, and education, which in this early
period of development was able to translate into high growth. In addition to
high levels of accumulation spurring growth, TFP growth over the 1960s was
also high, with large-scale public investments in critical infrastructure gener-
ating a significant growth response.
This is not to say that all of the investments undertaken during the 1960s
were exceptionally productive. Along with investments in large infrastructure
projects, the regional also invested heavily into protected state industries. But
in the 1960s, even the region’s overall strategy of industrial and agricultural
protectionism, supported by trade barriers and encouraged by publicly subsi-
dized energy, water, and agrochemicals, was initially successful, as it allowed
the region to utilize underused capacities and provide the early boost to
industrialization.
In the 1970s, going by growth figures alone, MENA was still in the middle
of a growth “heyday,” with GDP growth averaging 5.8 percent a year. But the
underlying conditions spurring growth in the 1970s represented a serious and
negative departure from the previous decade of high growth and productivity.

Breaking the Barriers to Higher Economic Growth178
Table 7.4. GDP per Laborer Growth and Growth of Accumulation and Productivity by Region,
1960–2000a(%)
Average Average Average
annual annual annual
GDP per growth of growth of Average
laborer human capital fixed capital annual
Region/country growth per laborer per laborer TFP growth
Sub-Saharan Africa
1960s 1.6 0.2 5.0 ⫺0.5
1970s 1.1 0.5 2.9 ⫺0.3
1980s 0.2 0.6 1.3 ⫺0.7
1990s 0.5 0.5 0.8 ⫺0.1
East Asia and Pacific
1960s 2.0 0.9 1.2 1.0
1970s 3.8 0.9 5.2 1.2
1980s 6.1 1.0 6.0 3.1
1990s 7.2 0.7 8.5 3.3
Latin America and the Caribbean
1960s 2.8 0.6 3.1 1.2
1970s 2.9 0.7 3.9 0.9
1980s ⫺1.7 0.9 0.2 ⫺2.3
1990s 0.8 0.9 0.7 0.0
Middle East and North Africa
1960s 4.2 0.6 4.6 2.0
1970s 2.6 1.1 7.2 ⫺1.0
1980s 0.2 1.4 1.9 ⫺1.5
1990s 0.6 1.3 -0.5 0.0
High-income/OECD
1960s 4.3 0.5 5.7 1.7
1970s 1.9 1.4 3.7 ⫺0.4
1980s 1.9 0.3 2.4 0.7
1990s 1.6 0.6 2.3 0.3
South Asia
1960s 2.3 0.6 4.1 0.3
1970s 0.7 0.9 1.9 ⫺0.7
1980s 3.4 0.9 3.2 1.6
1990s 3.1 0.9 3.7 1.1
World
1960s 2.8 0.7 3.4 1.0
1970s 2.4 1.0 4.0 0.2
1980s 3.3 0.8 3.6 1.3
1990s 3.8 0.8 4.5 1.5
World (minus China)
1960s 3.2 0.6 4.6 1.0
1970s 1.9 1.1 3.8 ⫺0.3
1980s 1.9 0.7 2.7 0.4
1990s 1.9 0.8 2.7 0.4
a. Regional averages weighted by initial period population. Eastern Europe included in world averages.

Part II: Labor Markets and Human Capital 179
To begin, the 1970s were marked by an increase in the rate of physical capital
accumulation per laborer of more than 50 percent, and almost a doubling of
the rate of human capital accumulation per laborer.Over the 1970s, the MENA
region realized the highest rate of growth of physical capital per laborer in the
world and the second highest rate of growth of human capital per laborer in
the world. Despite this immense increase in accumulation, on a per laborer
basis, growth actually declined, on average by 1.6 percent per year. Thus, the
1970s represented two large and yet conflicting growth dynamics for the
region, where investments were being undertaken in record levels (all things
equal, increasing the region’s growth potential) at the same time as the invest-
ments were having increasingly poor payoffs, in terms of growth (figure 7.2).
While MENA’s investments in needed infrastructure during the 1960s gen-
erated a significant payoff in terms of a growth response, by the 1970s, the
public sector’s sphere of comparative advantage in investment began to
shrink, and the limits of the MENA region’s strategy of protection of both
public and private industries began to be realized.
The first countries to experience the pattern of higher levels of accumula-
tion partnered with declining productivity were the oil-producing economies.
Saudi Arabia saw a four-fold increase in the rate of growth of fixed capital per
laborer—with physical capital per laborer increasing by an average of 18 per-
cent a year over the 1970s (versus 3.5 percent a year over the 1960s). Despite
this, total factor productivity growth plummeted from an average of 4.5 per-
cent per year to negative 0.8 percent a year. In Algeria, despite doubling the
rate of growth of human capital accumulation and a four-fold increase in the
rate of growth of fixed capita per laborer, TFP growth went from around 2.1
Figure 7.2. Growth of Output versus Growth of Factor Inputs per Laborer, 1960s versus 1970s
⫺1
0
1
2
3
4
5
average yearly growth of GDP per laborer
(broken down by factor accumulation
and productivity), %
left bar: 1960s
right bar: 1970s
accumulation
TFP growth
Middle East
and
North Africa
East Asia
and
Pacic
Latin America
and the
Caribbean
South Asia High-Income
OECD
Sub-Saharan
Africa
Source: See Annex.
Note: GDP growth equals bar length when TFP growth positive. Otherwise, GDP growth can be determined by tak-
ing accumulation and deducting that portion of growth lost t o negative TFP growth.

Breaking the Barriers to Higher Economic Growth180
percent a year over the 1960s to negative growth of about 0.1 percent a year
over the 1970s. And in Iran, despite a 50 percent increase in the rate of growth
of human capital per laborer over the 1970s (and relative maintenance of the
growth of fixed capital per laborer), the average growth of TFP over the
decade fell by more almost 8 percentage points (averaging 2.3 percent a year
over the 1960s and negative 5.5 percent over the 1970s).13 Nonoil economies,
on the other hand, by and large maintained positive productivity growth over
the 1970s—the limits of state-led planning and investment not being felt until
the following decade.
By the 1980s, as international oil prices slumped in the wake of global over-
production, the region’s economic gains became unsustainable and much of
the region witnessed slow, or even negative, per laborer growth rates. By the
1980s, most of the nonoil economies in the region saw TFP growth turn neg-
ative. With eroding macroeconomic balances and growing debt burdens, and
despite both heavy external assistance (which permitted spending for several
more years) and a strong social contract (which hindered the government’s
abilities to retract from commitments), investments declined dramatically,
with the rate of growth of the physical capital stock per laborer declining by
almost three-quarters from the previous decade (figure 7.4).
The decline in accumulation was almost without exception, with every
country in the region but Egypt and Kuwait experiencing a dramatic down-
turn in accumulation between the 1970s and 1980s (and in Kuwait, factor
accumulation did not exactly grow, but rather the negative accumulation per
Figure 7.3. Growth of Output versus Growth of Factor Inputs per Laborer: MENA Economies, 1960s versus 1970s
⫺8
⫺6
⫺4
⫺2
0
2
4
6
8
average yearly growth of GDP per laborer
(broken down by factor accumulation
and productivity), %
left bar: 1960s
right bar: 1970s
accumulation
TFP growth
Egypt Jordan Morocco Tunisia Algeria Iran Syria Kuwait Saudi
Arabia
Source: See Annex.
Note: GDP growth equals bar length when TFP growth positive. Otherwise, GDP growth can be determined by taking accumulation and deduct-
ing that portion of growth lost t o negative TFP growth.

Part II: Labor Markets and Human Capital 181
laborer that characterized the 1970s merely slowed over the 1980s), and
almost every economy experiencing a like decline in TFP. Only Morocco, Iran,
and Kuwait saw actual improvements in total factor productivity between the
1970s and 1980s (although for Iran and Kuwait, this improvement in TFP
growth was only a reflection of continuing TFP declines slowing over the
1980s). Negative productivity growth was most prevalent in the oil-producing
economies of the region—within the GCC economies as well as Algeria.
Because our TFP estimates are a reflection of factor efficiency, the degree to
which capital is underutilized will be heavily reflected in the ensuing TFP
growth measurements. This feature is of particular importance for these
economies, since, as oil prices collapsed in the 1980s, there was a significant
effort on the part of oil producers to prop up oil prices by holding down pro-
duction. Nevertheless, even in the non-oil-producing economies, there were
widespread declines in productivity for almost every country. With both mas-
sive declines in accumulation and corresponding declines in TFP for most
countries, the MENA region experienced a collapse of economic growth per
laborer.
By the late 1980s, the “lost decade of growth” prompted a handful of coun-
tries in the region—Morocco and Tunisia, and soon after, Jordan, to embark
on programs of macroeconomic stabilization and policy reform. By the 1990s,
nearly all of the non-GCC countries in the region followed suit, as did sever-
al of the Gulf economies. The reasoning, of course, was to create an environ-
ment in which the private sector could emerge and become an engine for
higher and sustainable economic growth, crucial for employment creation.
Figure 7.4. Growth of Output versus Growth of Factor Inputs per Laborer: MENA Economies, 1970s versus 1980s
⫺8
⫺6
⫺4
⫺2
0
2
4
6
8
average yearly growth of GDP per laborer
(broken down by factor accumulation
and productivity), %
left bar: 1970s
right bar: 1980s
Egypt Jordan Morocco Tunisia Algeria Iran Syria Kuwait Saudi
Arabia
accumulation
TFP growth
Source: See Annex.
Note: GDP growth equals bar length when TFP growth positive. Otherwise, GDP growth can be determined by taking accumulation and deduct-
ing that portion of growth lost t o negative TFP growth.

Breaking the Barriers to Higher Economic Growth182
How did the region fare over the 1990s? To understand the developments
over the 1990s, we turn to a growth decomposition. In table 7.5, we present the
MENA region’s changes to accumulation, productivity, and growth over the
decade. For clarity, the table does not present GDP and TFP growth over the
1990s, but rather the change in average GDP, factor, and TFP growth between
the 1980s and 1990s (thus if an economy moved from an average GDP per
laborer growth of 2 percent a year in the 1980s, to 5 percent a year over the
1990s, the change in GDP growth per laborer over the decade is 3 percent).
The countries are presented in order of the change in their average TFP
growth per laborer between the 1980s and 1990s. At the top of the list of
improved productivity growth is Saudi Arabia, which stemmed the enormous
productivity declines (averaging minus 5.9 percent a year) that characterized
the 1980s. Also exhibiting strongly improved productivity relative to the 1980s
is Syria, which in the 1990s benefited from both increased oil production and
agricultural performance, an aid windfall during the Gulf War (which allowed
it to undertake key growth-enhancing infrastructure investments, such as the
purchase of power stations and a telephone network), as well as some limited
liberalization reforms. And three of the four countries14 we would term the
‘early reformers’ (specifically, Jordan, Tunisia, and Egypt) also experienced
improvements in their average TFP growth between the 1980s and 1990s.
Overall, TFP growth in the region actually improved in all but one econo-
my (Morocco, which suffered from recurring droughts over the 1990s). At the
same time, however, as a result of large declines in accumulation within most
of MENA (particularly accumulation of physical capital), the improvements
in factor allocation and efficiency have not translated into significant increas-
es in GDP growth.
What are the implications of this exercise? The message that emerges in
terms of the region’s failure to improve its labor market outcomes is that
Table 7.5. Change in MENA’s per Laborer GDP Growth, Factor Accumulation, and TFP Growth
between 1980s and 1990s (%)
Change in GDP
per laborer Change in HK Change in K
growth: per laborer per laborer Change in
Country 1990s versus 1980s growth growth TFP growth
Saudi Arabia 6.7 ⫺0.5 0.8 6.8
Syria 1.7 ⫺0.7 ⫺7.8 5.3
Jordan 2.3 ⫺0.6 ⫺3.0 3.9
Kuwait 3.1 ⫺0.5 1.6 2.8
Iran 1.7 0.4 0.6 1.3
Tunisia 0.8 0.2 ⫺1.1 1.1
Egypt ⫺1.3 ⫺0.5 ⫺5.2 1.0
Algeria ⫺1.3 ⫺0.3 ⫺3.8 0.4
Morocco ⫺1.5 0.1 ⫺0.6 ⫺1.4

Part II: Labor Markets and Human Capital 183
improving the region’s labor market outcomes must come from substantial
increases in investment. Improving employment creation in the region can
come from only two sources: enhancing the employment-creation capacity of
growth (the employment elasticity), or higher economic growth itself.
Let’s consider the first notion—that improving labor market outcomes
could be achieved through increasing the employment intensity of growth.
Employment elasticities reflect the percent change in employment that is
associated with some percent change in real output. International evidence
would suggest that the long-run elasticity of output with respect to employ-
ment falls somewhere between 0.4 and 0.8. In countries that are highly capi-
tal-intensive, employment elasticity is likely to be closer to 0.4, while in labor-
intensive production structures, employment elasticity is likely to be closer to
0.8. In our estimates, we assume an employment elasticity of 0.7, which is rel-
atively healthy by international standards, and is unlikely to be improved. In
table 7.6, we estimate the level of output growth necessary to create sufficient
jobs to fully absorb the growing labor force, given relatively high rates of
employment creation.15 We then compare that GDP growth rate with the
observed growth rate over the 1990s. The difference reflects the gap between
needed growth and observed growth.
In four cases (Egypt, Tunisia, Kuwait, and Iran) the rates of growth that
were observed were sufficient, under optimistic employment generation
assumptions, to create the levels of employment to meet the growth of the
labor force, without any increases in either productivity or capital accumula-
tion. For the other countries in the region, however, the needed increases in
output growth in order to reach growth levels consistent with the desired rates
of employment growth are substantial: in Jordan, output would have needed
to grow by almost 3 percent more a year, and in Algeria, by more than 4 per-
cent more a year.
Table 7.6. GDP Growth Consistent with Desired Employment Creation Rates, 1990–2000 (%)
Needed GDP growth
employment consistent with that
growth over 1990s of employment Observed GDP growth
Country (= labor force growth) growth GDP growth gap
Egypt 2.9 4.2 4.5 0
Jordan 5.6 8.0 5.1 2.9
Morocco 2.4 3.5 2.2 1.2
Tunisia 3.1 4.4 4.7 0
Algeria 4.1 5.8 1.7 4.1
Iran 2.9 4.1 4.0 0.1
Syria 4.3 6.2 5.1 1.0
Kuwait 0.4 0.5 3.4 0
Saudi Arabia 2.3 3.2 2.7 0.5

Breaking the Barriers to Higher Economic Growth184
Thus, even with employment-intensive growth, the level of economic
growth itself has prevented the employment creation rates needed to absorb
the growing labor force. The remaining potential for improving employment
creation in the region, then, is higher economic growth. We saw, throughout
the 1990s, a substantial improvement in the region’s productivity growth.
While there is always potential for even greater productivity improvements,
there are also limits to what can be achieved. With economic growth so sub-
stantially below what would be needed to be consistent with full labor absorp-
tion, the considerable improvement in economic growth rates must come pri-
marily from substantial increases in investment.
Over the 1990s, however, almost every economy in MENA experienced an
actual decline in the amount of physical capital per laborer, and the region
went from increasing its physical capital per laborer by 1.9 percent a year in
the 1980s to experiencing actual declines in physical capital per laborer of 0.5
percent a year in the 1990s (Table 4). MENA stands at the bottom in terms of
physical capital accumulation during this period.
Interpreting the Decline in Investment
It is difficult to definitively interpret the substantial declines in accumulation
throughout the 1990s, without reliable investment data, broken down
between the public and private sectors. Public sector investments have almost
certainly dropped off. So, in the midst of an overall factor accumulation dete-
rioration, it is possible that private sector factor accumulation is actually
improving, but not sufficiently to counteract the large declines in public
investment.
However, it is also possible that while productivity and factor allocation
efficiency have improved significantly over the 1990s, they have failed to gen-
erate a comparable private sector investment response. Understanding the
lack of increased private investment is complex. Much of the private sector
investment that occurred in the 1980s in the region was domestically demand-
driven. The private sector developed under the patronage of governments. It
flourished not so much by being dynamic in a competitive environment, but
often by supplying protected domestic markets and generally “living off the
state.” Thus, the investments that took place by the private sector during the
1980s were largely focused on serving the domestic market rather than on
export expansion. For example, the share of construction in value added dur-
ing the 1980s, at 7.1 percent (relative to a world average of 5.6 percent), sug-
gests that a larger than average share of the region’s investment constituted
new buildings rather than retooling or investments in (new or high-tech) sec-
tors (see table 7). Over the 1990s, we saw a decline in share of construction in
value added (table 7.7), which may signal a change in the types of investment
the private sector is undertaking. So, even while we saw declines in private sec-

Part II: Labor Markets and Human Capital 185
tor investments over the 1990s, it could be the case that the investments cur-
rently taking place are more externally demand-driven and, hence, over the
long term, more sustainable.
Then there is the possibility that private sector investment, in addition to
dropping off in the domestic demand markets, has not significantly improved
in the tradable goods sectors, either. Why would the reform process,which has
clearly produced an impact on the region’s productivity, fail to generate a pri-
vate sector investment response in the external-oriented sectors? While the
comprehensive macroeconomic and structural reform programs espoused by
many of the MENA economies created an exuberant boost in their economic
outlooks, most of the region has failed to complete the reform process.
Reforms have generally been limited to the “stroke of the pen” type, easily exe-
cuted but in the absence of other, more serious, and challenging reforms, with
limited effect.
Policy Implications
Understanding why private investment has failed to respond to the improved
productivity and reforms in the region is essential for realizing the rates of
investment necessary for high and sustained economic growth. In order to
significantly increase the growth of private investment in the region, the pri-
vate investment climate must be improved. While this paper cannot definite-
ly establish the chief factors that have inhibited an enabling private investment
climate, several likely possibilities are offered.
To begin, the region has substantial work to do in terms of creating an
enabling macroeconomic environment through exchange rate management.
The preoccupation with macroeconomic stability during the 1990s often
Table 7.7. Share of Construction in Value Added, 1980s and 1990s (%)
Share of construction in value added
Country 1980s 1990s Change
Algeria 13.7 10.5 ⫺3.2
Egypt 5.0 5.0 0.0
Iran 6.9 3.8 ⫺3.1
Jordan 7.6 5.1 ⫺2.5
Kuwait 3.5 3.3 ⫺0.2
Morocco 5.9 4.7 ⫺1.2
Oman 5.2 3.2 ⫺2.0
Saudi Arabia 11.7 8.9 ⫺2.9
Syria 5.9 4.1 ⫺1.8
Tunisia 5.2 4.5 ⫺0.8
Regional Averagea7.1 5.3 ⫺1.8
World Averagea5.6 5.7 0.1
a. Unweighted average.

Breaking the Barriers to Higher Economic Growth186
relied on maintaining nominally fixed and stable exchange rates by virtually
all countries in the region (Algeria, Tunisia, Yemen, and more recently Iran
and Egypt are notable exceptions), which has meant that one important tool
to make exports more profitable was surrendered. This is the opposite of the
policies that the successful exporting countries such as the “East Asian Tigers”
have pursued over the last three decades. We know from experience that grow-
ing export of manufactured and nontraditional goods also creates dynamism
in the domestic economy with very significant spillover effects. The issue now
is to find ways to “exit” the pegs, in an orderly manner, to push economic
growth.
Next, there is the size of the public sector. On average, the public sector
accounts for about 50 percent of GDP and about a third of employment in the
region.16 The large role of the state—a sector that essentially has low produc-
tivity and limited inherent potential for productivity gain—is a drag on
growth in most economies in the region. Efforts have begun in many coun-
tries to reduce the public sector through rationalization of public employ-
ment, to improve its performance through better incentives and institutions,
and to privatize production of goods and services that could be produced
more efficiently in the private sector. But, by and large, these efforts have been
slow and half-hearted to date.
Third, the private sector is deterred from development because of the sys-
tems of governance that pervade the region. The World Bank’s report on gov-
ernance in the MENA region17 highlights the major governance challenges.
On the administrative side, MENA countries fall short of other countries at
similar income levels. In areas such as the efficiency of the bureaucracy, the
rule of law, the protection of property rights, the level of corruption, the qual-
ity of regulations, and the mechanisms of internal accountability, MENA
countries have, individually and on average, lower levels of quality of admin-
istration in the public sector than would be expected for their incomes.18
But even more important, countries across the region exhibit a pattern of
limited government accountability and inclusiveness, reflected in an index of
public accountability. In the area of openness of political institutions and par-
ticipation, respect of civil liberties, transparency of government, and freedom
of the press, the MENA region falls far short of the rest of the world.
Fourth, there is the issue of trade reform. Trade policy in the region
remains among the most restrictive in the world, with low level and speed of
integration into the world economy. Tariff rates remain high and the extent of
nontariff barriers large. A recent evaluation of the trade policy within the
MENA region, relative to other regions, found MENA economies to be among
the most restrictive in the world, second only to Sub-Saharan Africa and
South Asia.19 The importance of export orientation in growth is well estab-
lished in the empirical literature. High and sustainable growth simply does
not occur without a substantial outward orientation. A number of policy

Part II: Labor Markets and Human Capital 187
moves across the region are expected to lead to greater trade openness, stim-
ulating integration, and, it is hoped, growth. Most notable are the EU
Association Agreements on preferential trade (currently in force in Jordan,
Egypt, Algeria, Lebanon, the West Bank and Gaza, Tunisia, and Morocco, and
with an agreement signed by Syria in 2004).
Among the most important but lagging reforms is that of the banking sec-
tor.While the economies in the GCC, Jordan, and Lebanon have fairly sophis-
ticated financial sectors, with high bank and nonbank financial sector devel-
opment and generally good regulation and banking supervision, much of the
region’s private sector still has limited access to market finance. Banks domi-
nate the financial system, but in general they play a limited role in financial
intermediation. Much of the banking sector remains primarily in government
hands, inextricably linked to state-owned enterprises (SOEs), and subject to
government intervention in its lending and credit allocation policies to SOEs.
This intervention has led to a crowding out of the private sector where it is
permitted to operate, especially in Algeria, Libya, Syria, and Yemen.20
And finally, the region needs a virtual overhaul of its system of property
rights, as well as better legal systems and improved contract enforcement
mechanisms.
Unless the private sector begins to see itself as an independent source of
growth and productivity in the economy, and society begins to underpin this
change economically and politically, it is unlikely that any of the past econom-
ic reforms in themselves will be adequate. The public sector’s role in improv-
ing labor market outcomes in the region is important, but unlike in the 1960s
and 1970s, better labor market outcomes cannot be guaranteed through pub-
lic employment. The government’s role has distinctly changed. Now the pub-
lic sector must find ways to improve the investment climate and promote eco-
nomic growth, which remains the single most important way to ensure better
labor market outcomes in the future.
Notes
1. The countries of the Middle East and North Africa region included in this analysis (depending upon
data availability) are: Morocco, Algeria, Libya, Tunisia, Egypt, Jordan, Lebanon, Syria, Iran, Iraq,
Yemen, Bahrain, Kuwait, Oman, Qatar, Saudi Arabia, the West Bank and Gaza, and the United Arab
Emirates.
2 .World Bank 2003.
3. World Bank World Development Indicators database (April 2004 update).
4. 2005 estimate, based upon unemployment data for Algeria, Morocco, Tunisia, Egypt, Iran, Jordan,
Lebanon, Syria, West Bank and Gaza, Yemen, Bahrain, Kuwait, Oman, Qatar, Saudi Arabia, and West
Bank and Gaza.
5. Unemployment in Algeria was close to 30 percent in 2002. With recent rising oil revenues and as
part of the country’s Economic Recovery Program, however, large-scale temporary employment
schemes have dramatically lowered the rate of official unemployment (to 23.7 percent currently),
although the longer-term sustainability of these jobs is questionable.

Breaking the Barriers to Higher Economic Growth188
6. Based upon growth in average educational attainment of the adult (15+) population between 1960-
90 in Algeria, Bahrain, Egypt, Iran, Jordan, Kuwait, Morocco, Saudi Arabia, Syria, and Tunisia.
Educational attainment data from Barro and Lee (2000). Regional average population weighted.
7. Based on GDP per capita growth in constant US$ for Algeria, Bahrain, Egypt, Iran, Jordan, Kuwait,
Morocco, Oman, Qatar, Saudi Arabia, Syria, Tunisia, and the United Arab Emirates (does not include
Lebanon, Libya, Djibouti, Iraq, or Yemen, due to data availability for both periods). GDP per capita
weighted by population for regional average.
8. This estimate does not include Iraq, where as much as 50 percent of the workforce may be unem-
ployed.
9. Of course, over the short term, wages may not move in tandem with worker productivity increases.
Additionally, employment growth may arise without real output growth. But over the long run, sus-
tainable increases in employment and wages depend upon increases in real output per laborer.
10. Boltho and Glyn, 1995.
11. In our case, the elasticity of output with respect to capital is exogenously assumed to be 0.4, which
is based upon both international evidence and our own estimations.
12. See Bosworth, Collins and Chen (1995) for similar findings.
13. See Annex Table 1.
14. Jordan, Morocco, Tunisia, and Egypt, which all embarked upon structural reform programs from
the mid-1980s to early 1990s.
15. Of course, the process is circular: Just as employment creates output growth, output growth in
some sense “creates” employment, in that—in order to sustain that level of output growth—it requires
continuing increases in employment. Thus, rather than think of growth generating employment, we
can think of certain levels of output growth consistent with a given level of sustainable employment
creation.
16. World Bank 2005.
17. World Bank 2003b.
18. World Bank 2003b.
19. As measured by average tariffs on imports. From World Bank, 2005.
20. World Bank, 2005.
References
Barro, Robert J., and Jong-Wha Lee. 2000. “International Data on Educational Attainment: Updates
and Implications.” CID Working Paper No. 42. Center for International Development at Harvard
University.
Boltho, Andrea, and Andrew Glyn. 1995. “Can Macroeconomic Policies Raise Growth?” International
Labor Review 134 (4–5): 451–70.
Bosworth, Barry, Susan M.Collins, and Yu-chin Chen. 1995. “Accounting for Differences in Economic
Growth.” Paper prepared for conference on “Structural Adjustment Policies in the 1990s:
Experience and Prospects,” organized by the Institute of Developing Economies, Tokyo, Japan.
Easterly, William, and Ross Levine. 2001. “It’s Not Factor Accumulation: Stylized Facts and Growth
Models.” World Bank Economic Review 15 (2): 177–219.
Fajnzylber, Pablo, and Daniel Lederman. 2000. “Economic Reforms and Total Factor Productivity
Growth in Latin America and the Caribbean, 1990–95: An Empirical Note.” World Bank Working
Paper 2114. The World Bank, Washington, DC.
Griliches, Zvi. 1979. “Issues in Assessing the Contribution of Research and Development to
Productivity Growth.” Bell Journal of Economics 10 (1): 92–116.

Part II: Labor Markets and Human Capital 189
Griliches, Zvi, and Frank Lichtenberg. 1984.“R&D and Productivity Growth at the Industry Level: Is
There Still a Relationship?” In R&D, Patents, and Productivity, ed. Zvi Griliches. Chicago, IL:
University of Chicago Press.
Islam, Nazrul. 1999. “International Comparison of Total Factor Productivity: A Review.” Review of
Income and Wealth 45 (4).
Lefort, Fernando, and Andres Solimano. 1994.“Economic Growth after Market-Based Reform in Latin
America: The Cases of Chile and Mexico.” World Bank Macroeconomics and Growth Division
Working Paper.World Bank,Washington, DC.
Nehru, Vikram, and Ashok Dhareshwar. 1993. “A New Database on Physical Capital Stock: Sources,
Methodology, and Results.”Revista de Análisis Economico 8 (1): 37–59.
Rama, Martin. 1998.“How Bad is Unemployment in Tunisia? Assessing Labor Market Efficiency in a
Developing Country.” World Bank Research Observer 13 (1): 59–77.
World Bank. 2003. Unlocking the Employment Potential in the Middle East and North Africa: Towards a
New Social Contract. MENA Development Report. World Bank. Washington, DC.
———. 2003b. Better Governance for Development in the Middle East and North Africa: Enhancing
Inclusiveness and Accountability. MENA Development Report. World Bank. Washington, DC.
———. 2004. World Development Indicators database. April 2004 update.
———. 2005. MENA Economic Developments and Prospects 2005: Oil Booms and Revenue
Management. World Bank, Washington, DC.

Breaking the Barriers to Higher Economic Growth190
Annex: Measuring Growth, Accumulation, and TFP Growth
To examine how the MENA region’s growth has changed since it began its
comprehensive structural reform process, we made simple calculations of the
change in rate of accumulation and total factor productivity growth.
TFP growth is the residual of what cannot be explained by investments if
we assume those investments (both physical and human) earn a reasonable
rate of return. TFP growth is often thought of as “technical progress,” but in
fact, as the residual of a growth accounting estimation, it not only embodies
the differences across countries in their progress in the adoption of better
technology, but also reflects a host of non-technology-related differences,
including changes in the utilization of both capital and labor, changes in
schooling quality, and changes in the overall efficiency with which factors are
allocated in the production process. Because of the many other factors that
can potentially affect the growth residual, much empirical work has focused
on reducing those elements of the residual (TFP) which do not reflect actual
shifts in technology-related opportunities in the economy. For example,
adjustments for the business cycle have been introduced to account for the
short-term fluctuations in capacity utilization (Griliches 1979; Lefort and
Solimano 1994; Fajnzylber and Lederman 2000). An alternative procedure
employed by Griliches and Lichtenberg (1984) has been to estimate growth
over five-year periods,and to allow the TFP series only to increase or stay con-
stant (resetting any values to the previously observed peak level), to maintain
the assumption that “true” productivity can only improve and that measured
reductions in TFP can only reflect short-term fluctuations.
For our purposes, we have adopted a more casual approach about our
measurements. Our interest is to explore how MENA’s overall growth has
improved or deteriorated since it began the structural reform process. In the
end, growth will be determined by accumulation of both physical and human
capital, as well as the overall manner in which those factors are put to produc-
tion. For the MENA region, things such as improved capacity utilization of
capital and human capital by the region are precisely the elements we believe
may be heavily affected by structural reform, and thus we would like to have
this effect reflected in our estimates. At the same time, as we discuss in the
subsequent section, we have controlled for global shocks.
Under many circumstances, the environment created to encourage invest-
ment would also correspond to an environment in which those investments
could be productive. But in the MENA region, accumulation and productivi-
ty have often gone in opposite directions, such as during the period of mas-
sive public sector investments, which yielded rates of return well below inter-
national norms. Examining growth alone will mask these very different
effects, and the somewhat anemic growth that has characterized the region
since reform may be more a reflection of significantly lower public invest-

Part II: Labor Markets and Human Capital 191
ments than of continuing poor productivity performance. From the stand-
point of evaluating the impact of the region’s structural reform, it is precisely
TFP growth which we would expect to be most influenced by changes in
national policies that enhance the efficiency of capital and labor.
Data and Methodology
TFP growth estimates were made utilizing panel data of capital stock accumu-
lation, human capital stock accumulation, and GDP growth from 1960 to
2000. Estimates of the physical capital stock for a sample of 83 economies
from 1960 to 1990 come from Nehru and Dhareshwar (19931), which was cre-
ated by a perpetual inventory method from investment rates from 1950 for-
ward, with initial assumptions about the capital/output ratio, and assuming a
common fixed annual geometric depreciation rate of 0.04. These capital stock
data were extended to 2000 using the growth rates of constant price local cur-
rency investment from the World Bank’s World Development Indicators data-
base,2and applying similar assumptions on the depreciation rate. Capital
stock estimates for another 12 economies, including 4 economies in the
MENA region of particular interest to us, were created according to a similar
methodology, using investment rates from 1960 forward.
Real GDP in constant local currency also come from World Bank data. The
human capital-augmented labor stock was estimated, using both labor force
estimates from the World Bank’s World Development Indicators, and esti-
mates of the educational attainment of the adult population from Barro and
Lee.3The functional form of human capital-augmented labor has been
assumed as:The functional form ofhuman capital-augmented labor has been assumed as:
H = L e (r * S)
where L is the labor force, S is the average years of schooling of the adult pop-
ulation, and r is the rate of return to schooling. According to international evi-
dence, a reasonable approximation of that rate of return is 10 percent, which
we have assumed for the purposes of our analysis.
TFP growth was calculated over 10-year periods from 1960 to 2000, rather
than on an annual basis, to minimize the error that is inherent in current cap-
ital stock measurements. National accounts would attribute any investment
expenditures made over the year, even the last day of the year, to that year’s
capital stock. However, it is unlikely that that investment expenditure would
contribute to economic growth immediately, but rather would only create the
potential to contribute to growth into the future. To reduce this lag effect that
physical capital exhibits, we calculated TFP growth based on 10-year averages.
Production was assumed to follow a Cobb-Douglas specification with con-
stant returns to scale between physical and human capital-augmented labor:
Yt= A (t) * Ktα*Ht(1 – α)

Breaking the Barriers to Higher Economic Growth192
where Y is output, A is an index of total factor productivity, and K and H are
the stocks of physical and human-augmented labor, respectively. Dividing
both sides by the work force, taking logs,and first-differencing, growth of out-
put per laborer can be related as follows:
ln (yi/yi–1,) = αln (kt/ kt – 1) + (1 – α)ln(ht/ ht – 1) + ln (At/ At – 1)
To determine the coefficients on capital and human capital-augmented
labor, αand (1 – α), the average annual rate of GDP per capita growth over
the decade, was regressed on average growth of physical capital per worker
and human-capital per worker with a least squares trend over the entire peri-
od of availability (1960–2000).
From our estimation, the elasticity of output of physical capital was esti-
mated to be 0.49, somewhat higher than the average estimated coefficient
from previous research, but within the range of accepted parameters. This
may be due to the inclusion of several more developing countries than in the
original Nehru-Dhareshwar physical capital stock dataset, made possible
using World Bank data. At the same time, our purpose here is not to break
new ground in measuring TFP, but to evaluate the region’s performance in
factor allocation and efficiency. Thus, we have calculated the TFP using three
distinct calculations of factor shares—αk=0.3, αk=0.4, and αk=0.5—to check
the sensitivity of the region’s growth performance to the assumptions made
on the output elasticities. The resulting sets of TFP growth estimations for the
full sample of countries are presented in Annex Table 1. Within the text of the
paper, TFP calculations are based on an elasticity of capital assumption of 0.4
across countries. Regional averages were calculated by weighting growth rates
by initial population.

Part II: Labor Markets and Human Capital 193
Annex Table 1. TFP Estimates under Various Assumptions on Elasticity of Output with Respect to Physical Capital (α),
1960–2000
TFP TFP TFP
Region Country Decade (α= 0.3) (α= 0.4) (α= 0.5)
Africa Botswana 1960s .. .. ..
Africa Cameroon 1960s ⫺0.66 ⫺0.95 ⫺1.24
Africa Ghana 1960s ⫺2.14 ⫺2.37 ⫺2.60
Africa Kenya 1960s 1.34 1.45 1.56
Africa Lesotho 1960s ⫺2.38 ⫺4.38 ⫺6.37
Africa Malawi 1960s 0.10 ⫺0.72 ⫺1.53
Africa Mali 1960s 0.62 0.42 0.21
Africa Mauritius 1960s 0.01 0.26 0.50
Africa Mozambique 1960s 1.58 1.16 0.73
Africa Rwanda 1960s .. .. ..
Africa Senegal 1960s ⫺0.37 ⫺0.34 ⫺0.31
Africa Sierra Leone 1960s 1.26 0.66 0.07
Africa South Africa 1960s 2.43 2.10 1.77
Africa Sudan 1960s ⫺2.99 ⫺3.76 ⫺4.52
Africa Tanzania 1960s 3.16 2.98 2.81
Africa The Gambia 1960s .. .. ..
Africa Togo 1960s 1.61 0.47 ⫺0.67
Africa Uganda 1960s 0.33 0.07 ⫺0.18
Africa Zambia 1960s 0.54 0.46 0.37
Africa Zimbabwe 1960s 3.32 3.39 3.45
E. Asia China 1960s 1.06 1.13 1.19
E. Asia Indonesia 1960s 0.73 0.73 0.72
E. Asia Korea, Rep. of 1960s 1.74 0.90 0.07
E. Asia Malaysia 1960s 0.78 0.21 ⫺0.37
E. Asia Papua New Guinea 1960s 3.94 3.62 3.31
E. Asia Philippines 1960s 0.40 0.09 ⫺0.23
E. Asia Singapore 1960s 2.04 0.75 ⫺0.53
E. Asia Taiwan, China 1960s 1.84 0.87 ⫺0.10
E. Asia Thailand 1960s 2.36 1.33 0.31
ECA Bulgaria 1960s .. .. ..
ECA Hungary 1960s 3.50 3.44 3.39
ECA Portugal 1960s 3.33 2.66 1.98
ECA Romania 1960s .. .. ..
ECA Turkey 1960s .. .. ..
LAC Argentina 1960s 0.66 0.39 0.12
LAC Bolivia 1960s 0.40 0.03 ⫺0.34
LAC Brazil 1960s 1.61 1.33 1.05
LAC Chile 1960s 1.48 1.26 1.04
LAC Colombia 1960s 2.40 2.23 2.05
LAC Costa Rica 1960s 1.20 0.91 0.62
LAC Dominican Republic 1960s 0.68 0.62 0.56
LAC Ecuador 1960s 0.81 0.62 0.43
(Table continues on the following page.)

Breaking the Barriers to Higher Economic Growth194
Annex Table 1. (continued)
TFP TFP TFP
Region Country Decade (α= 0.3) (α= 0.4) (α= 0.5)
LAC El Salvador 1960s 0.56 0.35 0.14
LAC Guatemala 1960s 1.62 1.40 1.17
LAC Guyana 1960s 0.95 0.82 0.69
LAC Honduras 1960s 0.86 0.63 0.39
LAC Mexico 1960s 1.52 1.14 0.76
LAC Nicaragua 1960s 1.51 1.08 0.65
LAC Panama 1960s 2.62 1.97 1.31
LAC Paraguay 1960s 0.69 0.50 0.30
LAC Peru 1960s 1.34 1.22 1.09
LAC Uruguay 1960s 0.52 0.59 0.66
LAC Venezuela 1960s 1.86 1.92 1.99
MENA Algeria 1960s 2.16 2.08 2.00
MENA Bahrain 1960s .. .. ..
MENA Egypt 1960s 1.81 1.52 1.23
MENA Iran 1960s 3.29 2.34 1.39
MENA Jordan 1960s ⫺2.97 ⫺3.49 ⫺4.02
MENA Kuwait 1960s ⫺0.16 0.72 1.61
MENA Morocco 1960s 1.98 1.84 1.71
MENA Saudi Arabia 1960s 4.85 4.53 4.20
MENA Syria 1960s 2.11 2.04 1.98
MENA Tunisia 1960s 1.64 1.24 0.83
OECD Australia 1960s 1.27 0.98 0.69
OECD Austria 1960s 3.04 2.20 1.37
OECD Belgium 1960s 2.26 1.90 1.55
OECD Canada 1960s 1.94 1.78 1.63
OECD Denmark 1960s 1.39 0.73 0.07
OECD Finland 1960s 2.06 1.67 1.28
OECD France 1960s 2.40 1.76 1.11
OECD Greece 1960s 4.13 3.17 2.22
OECD Iceland 1960s 0.69 0.46 0.24
OECD Ireland 1960s 2.20 1.67 1.15
OECD Israel 1960s 3.30 3.06 2.81
OECD Italy 1960s 3.00 2.42 1.84
OECD Japan 1960s 4.95 3.64 2.34
OECD Netherlands 1960s 0.27 0.00 ⫺0.26
OECD New Zealand 1960s 0.88 0.69 0.50
OECD Norway 1960s 1.26 1.17 1.07
OECD Spain 1960s 3.30 2.58 1.86
OECD Sweden 1960s 2.00 1.56 1.12
OECD Switzerland 1960s 0.67 0.24 ⫺0.18
OECD United Kingdom 1960s 0.94 0.44 ⫺0.07
OECD United States 1960s 0.79 0.72 0.64
S. Asia Bangladesh 1960s 1.27 1.08 0.89
S. Asia India 1960s 0.52 0.20 ⫺0.13
S. Asia Myanmar 1960s 0.43 0.32 0.20

Part II: Labor Markets and Human Capital 195
TFP TFP TFP
Region Country Decade (α= 0.3) (α= 0.4) (α= 0.5)
S. Asia Pakistan 1960s 0.81 ⫺0.20 ⫺1.22
S. Asia Sri Lanka 1960s 1.25 1.20 1.15
Africa Botswana 1970s 9.21 8.86 8.52
Africa Cameroon 1970s 2.01 1.35 0.68
Africa Ghana 1970s ⫺2.18 ⫺2.15 ⫺2.11
Africa Kenya 1970s 3.09 3.10 3.11
Africa Lesotho 1970s 6.05 5.57 5.10
Africa Malawi 1970s 0.60 ⫺0.03 ⫺0.66
Africa Mali 1970s 1.66 1.59 1.52
Africa Mauritius 1970s 1.10 1.11 1.12
Africa Mozambique 1970s ⫺3.64 ⫺3.74 ⫺3.84
Africa Rwanda 1970s 0.42 ⫺0.03 ⫺0.47
Africa Senegal 1970s ⫺1.26 ⫺1.24 ⫺1.21
Africa Sierra Leone 1970s ⫺0.19 ⫺0.22 ⫺0.25
Africa South Africa 1970s 0.41 ⫺0.06 ⫺0.52
Africa Sudan 1970s ⫺0.82 ⫺1.14 ⫺1.46
Africa Tanzania 1970s ⫺0.06 ⫺0.36 ⫺0.67
Africa The Gambia 1970s .. .. ..
Africa Togo 1970s ⫺1.42 ⫺2.14 ⫺2.85
Africa Uganda 1970s ⫺10.30 ⫺10.18 ⫺10.06
Africa Zambia 1970s ⫺1.92 ⫺1.76 ⫺1.61
Africa Zimbabwe 1970s ⫺0.44 ⫺0.60 ⫺0.75
E. Asia China 1970s 1.83 1.47 1.12
E. Asia Indonesia 1970s 1.45 0.77 0.09
E. Asia Korea, Rep. of 1970s ⫺1.41 ⫺2.32 ⫺3.23
E. Asia Malaysia 1970s 1.11 0.51 ⫺0.10
E. Asia Papua New Guinea 1970s ⫺0.37 ⫺0.50 ⫺0.63
E. Asia Philippines 1970s 0.29 0.03 ⫺0.22
E. Asia Singapore 1970s 0.96 0.00 ⫺0.96
E. Asia Taiwan, China 1970s 2.11 1.05 ⫺0.01
E. Asia Thailand 1970s 1.28 0.69 0.10
ECA Bulgaria 1970s .. .. ..
ECA Hungary 1970s 2.67 2.07 1.47
ECA Portugal 1970s 0.23 0.03 ⫺0.18
ECA Romania 1970s .. .. ..
ECA Turkey 1970s 0.06 ⫺0.49 ⫺1.03
LAC Argentina 1970s ⫺0.04 ⫺0.30 ⫺0.56
LAC Bolivia 1970s 1.96 2.09 2.22
LAC Brazil 1970s 3.20 2.58 1.96
LAC Chile 1970s ⫺0.05 0.10 0.25
LAC Colombia 1970s 0.80 0.77 0.76
LAC Costa Rica 1970s ⫺0.71 ⫺1.04 ⫺1.36
LAC Dominican Republic 1970s 1.32 0.75 0.18
LAC Ecuador 1970s 1.00 0.84 0.68
LAC El Salvador 1970s ⫺2.21 ⫺2.54 ⫺2.87
(Table continues on the following page.)

Breaking the Barriers to Higher Economic Growth196
Annex Table 1. (continued)
TFP TFP TFP
Region Country Decade (α= 0.3) (α= 0.4) (α= 0.5)
LAC Guatemala 1970s 1.17 0.91 0.65
LAC Guyana 1970s ⫺1.47 ⫺1.43 ⫺1.39
LAC Honduras 1970s 0.77 0.56 0.36
LAC Mexico 1970s 0.71 0.44 0.17
LAC Nicaragua 1970s ⫺4.10 ⫺4.25 ⫺4.40
LAC Panama 1970s ⫺1.62 ⫺1.97 ⫺2.32
LAC Paraguay 1970s 2.68 2.10 1.53
LAC Peru 1970s ⫺0.96 ⫺0.91 ⫺0.86
LAC Uruguay 1970s 1.60 1.43 1.27
LAC Venezuela 1970s ⫺3.72 ⫺3.49 ⫺3.25
MENA Algeria 1970s 0.28 ⫺0.14 ⫺0.55
MENA Bahrain 1970s .. .. ..
MENA Egypt 1970s 1.98 1.51 1.04
MENA Iran 1970s ⫺4.72 ⫺5.52 ⫺6.33
MENA Jordan 1970s 3.18 2.38 1.58
MENA Kuwait 1970s ⫺6.95 ⫺6.33 ⫺5.72
MENA Morocco 1970s 0.04 ⫺0.32 ⫺0.68
MENA Saudi Arabia 1970s 0.82 ⫺0.83 ⫺2.48
MENA Syria 1970s 2.12 1.17 0.22
MENA Tunisia 1970s 1.61 1.43 1.25
OECD Australia 1970s 0.04 ⫺0.19 ⫺0.42
OECD Austria 1970s 1.22 0.71 0.20
OECD Belgium 1970s 1.88 1.50 1.13
OECD Canada 1970s ⫺0.53 ⫺0.53 ⫺0.53
OECD Denmark 1970s ⫺0.51 ⫺0.81 ⫺1.11
OECD Finland 1970s 0.95 0.71 0.47
OECD France 1970s 0.21 ⫺0.15 ⫺0.51
OECD Greece 1970s 0.62 0.18 ⫺0.26
OECD Iceland 1970s 2.01 1.83 1.65
OECD Ireland 1970s 1.57 1.14 0.71
OECD Israel 1970s 0.72 0.49 0.25
OECD Italy 1970s 1.55 1.22 0.89
OECD Japan 1970s 0.09 ⫺0.63 ⫺1.35
OECD Netherlands 1970s 0.21 0.01 ⫺0.19
OECD New Zealand 1970s ⫺1.85 ⫺1.84 ⫺1.83
OECD Norway 1970s 1.43 1.30 1.17
OECD Spain 1970s 0.21 ⫺0.21 ⫺0.63
OECD Sweden 1970s ⫺1.15 ⫺1.21 ⫺1.27
OECD Switzerland 1970s ⫺1.55 ⫺1.71 ⫺1.87
OECD United Kingdom 1970s 0.09 ⫺0.16 ⫺0.40
OECD United States 1970s ⫺1.13 ⫺1.02 ⫺0.90
S. Asia Bangladesh 1970s ⫺2.11 ⫺1.93 ⫺1.75
S. Asia India 1970s ⫺0.64 ⫺0.76 ⫺0.87
S. Asia Myanmar 1970s 1.88 1.78 1.67
S. Asia Pakistan 1970s 0.58 0.42 0.25
S. Asia Sri Lanka 1970s 0.09 ⫺0.33 ⫺0.75

Part II: Labor Markets and Human Capital 197
TFP TFP TFP
Region Country Decade (α= 0.3) (α= 0.4) (α= 0.5)
Africa Botswana 1980s 4.04 3.66 3.28
Africa Cameroon 1980s ⫺1.56 ⫺2.15 ⫺2.74
Africa Ghana 1980s ⫺0.37 ⫺0.07 0.24
Africa Kenya 1980s 0.96 1.17 1.39
Africa Lesotho 1980s 0.98 0.51 0.04
Africa Malawi 1980s ⫺0.35 ⫺0.19 ⫺0.03
Africa Mali 1980s ⫺1.87 ⫺1.86 ⫺1.84
Africa Mauritius 1980s 3.03 3.01 3.00
Africa Mozambique 1980s ⫺1.32 ⫺1.32 ⫺1.33
Africa Rwanda 1980s ⫺3.22 ⫺3.77 ⫺4.31
Africa Senegal 1980s 0.45 0.45 0.45
Africa Sierra Leone 1980s ⫺1.13 ⫺1.03 ⫺0.93
Africa South Africa 1980s ⫺2.64 ⫺2.47 ⫺2.31
Africa Sudan 1980s ⫺1.01 ⫺1.09 ⫺1.18
Africa Tanzania 1980s ⫺0.14 ⫺0.05 0.04
Africa The Gambia 1980s ⫺1.91 ⫺2.29 ⫺2.67
Africa Togo 1980s ⫺2.36 ⫺2.31 ⫺2.26
Africa Uganda 1980s 0.40 0.75 1.10
Africa Zambia 1980s ⫺0.74 ⫺0.22 0.30
Africa Zimbabwe 1980s ⫺1.16 ⫺0.73 ⫺0.30
E. Asia China 1980s 4.40 3.92 3.45
E. Asia Indonesia 1980s 1.06 0.37 ⫺0.32
E. Asia Korea, Rep. of 1980s 2.09 1.40 0.71
E. Asia Malaysia 1980s 0.20 ⫺0.24 ⫺0.69
E. Asia Papua New Guinea 1980s ⫺1.67 ⫺01.61 ⫺1.55
E. Asia Philippines 1980s ⫺2.12 ⫺2.23 ⫺2.33
E. Asia Singapore 1980s 1.81 1.28 0.75
E. Asia Taiwan, China 1980s 3.92 2.30 2.49
E. Asia Thailand 1980s 2.63 2.19 1.74
ECA Bulgaria 1980s ⫺0.02 ⫺0.32 ⫺0.63
ECA Hungary 1980s 0.92 0.54 0.16
ECA Portugal 1980s 1.05 0.86 0.66
ECA Romania 1980s 0.52 0.92 1.31
ECA Turkey 1980s 1.35 1.21 1.08
LAC Argentina 1980s ⫺3.43 ⫺3.29 ⫺3.16
LAC Bolivia 1980s ⫺1.95 ⫺1.59 ⫺1.24
LAC Brazil 1980s ⫺2.42 ⫺2.38 ⫺2.34
LAC Chile 1980s 0.66 0.72 0.78
LAC Colombia 1980s ⫺0.75 ⫺0.75 ⫺0.75
LAC Costa Rica 1980s ⫺1.64 ⫺1.59 ⫺1.54
LAC Dominican Republic 1980s ⫺1.80 ⫺2.08 ⫺2.37
LAC Ecuador 1980s ⫺1.19 ⫺1.17 ⫺1.14
LAC El Salvador 1980s ⫺2.87 ⫺2.69 ⫺2.51
LAC Guatemala 1980s ⫺1.67 ⫺1.59 ⫺1.51
LAC Guyana 1980s ⫺4.09 ⫺3.94 ⫺3.79
LAC Honduras 1980s ⫺1.77 ⫺1.58 ⫺1.39
(Table continues on the following page.)

Breaking the Barriers to Higher Economic Growth198
Annex Table 1. (continued)
TFP TFP TFP
Region Country Decade (α= 0.3) (α= 0.4) (α= 0.5)
LAC Mexico 1980s ⫺3.04 ⫺2.92 ⫺2.80
LAC Nicaragua 1980s ⫺4.59 ⫺4.48 ⫺4.38
LAC Panama 1980s ⫺2.69 ⫺2.46 ⫺2.23
LAC Paraguay 1980s ⫺1.91 ⫺2.20 ⫺2.49
LAC Peru 1980s ⫺3.70 ⫺3.66 ⫺3.62
LAC Uruguay 1980s ⫺1.96 ⫺1.76 ⫺1.55
LAC Venezuela 1980s ⫺1.68 ⫺1.50 ⫺1.33
MENA Algeria 1980s ⫺2.45 ⫺2.40 ⫺2.36
MENA Bahrain 1980s ⫺5.28 ⫺5.24 ⫺5.20
MENA Egypt 1980s 0.01 ⫺0.31 ⫺0.63
MENA Iran 1980s ⫺1.26 ⫺1.10 ⫺0.93
MENA Jordan 1980s ⫺4.45 ⫺4.43 ⫺4.41
MENA Kuwait 1980s ⫺5.12 ⫺4.61 ⫺4.10
MENA Morocco 1980s 0.29 0.21 0.13
MENA Saudi Arabia 1980s ⫺6.41 ⫺5.88 ⫺5.35
MENA Syria 1980s ⫺3.92 ⫺4.42 ⫺4.92
MENA Tunisia 1980s ⫺0.45 ⫺0.54 ⫺0.64
OECD Australia 1980s 0.11 ⫺0.05 ⫺0.21
OECD Austria 1980s 0.81 0.58 0.36
OECD Belgium 1980s 0.70 0.55 0.40
OECD Canada 1980s ⫺0.21 ⫺0.33 ⫺0.46
OECD Denmark 1980s 0.05 ⫺0.03 ⫺0.11
OECD Finland 1980s 0.06 0.03 ⫺0.01
OECD France 1980s 0.99 0.70 0.42
OECD Greece 1980s ⫺1.41 ⫺1.42 ⫺1.42
OECD Iceland 1980s ⫺0.11 ⫺0.17 ⫺0.23
OECD Ireland 1980s 1.34 1.16 0.99
OECD Israel 1980s 1.05 0.98 0.92
OECD Italy 1980s 0.46 0.32 0.19
OECD Japan 1980s 1.23 0.87 0.51
OECD Netherlands 1980s ⫺0.27 ⫺0.26 ⫺0.24
OECD New Zealand 1980s ⫺0.43 ⫺0.50 ⫺0.58
OECD Norway 1980s ⫺1.26 ⫺1.08 ⫺0.90
OECD Spain 1980s 0.67 0.48 0.29
OECD Sweden 1980s 0.84 0.65 0.46
OECD Switzerland 1980s 0.22 0.05 ⫺0.12
OECD United Kingdom 1980s 1.02 0.85 0.68
OECD United States 1980s 1.36 1.15 0.94
S. Asia Bangladesh 1980s ⫺0.19 ⫺0.53 ⫺0.87
S. Asia India 1980s 2.33 2.10 1.87
S. Asia Myanmar 1980s ⫺2.13 ⫺2.24 ⫺2.35
S. Asia Pakistan 1980s 1.21 1.13 1.05
S. Asia Sri Lanka 1980s 0.08 ⫺0.44 ⫺0.96
Africa Botswana 1990s ⫺0.05 ⫺0.39 ⫺0.74
Africa Cameroon 1990s ⫺1.21 ⫺0.95 ⫺0.69

Part II: Labor Markets and Human Capital 199
TFP TFP TFP
Region Country Decade (α= 0.3) (α= 0.4) (α= 0.5)
Africa Ghana 1990s 0.34 0.16 ⫺0.03
Africa Kenya 1990s ⫺1.64 ⫺1.43 ⫺1.22
Africa Lesotho 1990s 0.18 ⫺0.35 ⫺0.88
Africa Malawi 1990s 1.90 2.15 2.39
Africa Mali 1990s 1.66 1.66 1.66
Africa Mauritius 1990s 1.74 1.39 1.03
Africa Mozambique 1990s 2.48 2.24 2.01
Africa Rwanda 1990s ⫺1.75 ⫺1.75 ⫺1.76
Africa Senegal 1990s 0.18 0.10 0.01
Africa Sierra Leone 1990s .. .. ,,
Africa South Africa 1990s ⫺1.29 ⫺1.09 ⫺0.88
Africa Sudan 1990s .. .. ..
Africa Tanzania 1990s 0.41 0.50 0.59
Africa The Gambia 1990s ⫺1.31 ⫺1.38 ⫺1.45
Africa Togo 1990s ⫺0.26 0.04 0.33
Africa Uganda 1990s 3.52 3.43 3.33
Africa Zambia 1990s ⫺1.42 ⫺0.92 ⫺0.43
Africa Zimbabwe 1990s ⫺1.05 ⫺1.15 ⫺1.25
E. Asia China 1990s 5.54 4.64 3.75
E. Asia Indonesia 1990s ⫺0.76 ⫺1.22 ⫺1.68
E. Asia Korea, Rep. of 1990s 0.90 0.23 ⫺0.44
E. Asia Malaysia 1990s 1.35 0.74 0.13
E. Asia Papua New Guinea 1990s 2.04 2.22 2.41
E. Asia Philippines 1990s ⫺0.83 ⫺0.83 ⫺0.83
E. Asia Singapore 1990s 2.31 1.85 1.40
E. Asia Taiwan, China 1990s 2.19 1.43 0.66
E. Asia Thailand 1990s 0.23 ⫺0.40 ⫺1.03
ECA Bulgaria 1990s ⫺1.38 ⫺1.38 ⫺1.38
ECA Hungary 1990s 0.03 ⫺0.11 ⫺0.24
ECA Portugal 1990s 0.18 ⫺0.05 ⫺0.29
ECA Romania 1990s ⫺0.27 0.30 0.87
ECA Turkey 1990s ⫺0.81 ⫺0.95 ⫺1.09
LAC Argentina 1990s 1.98 2.01 2.05
LAC Bolivia 1990s 0.75 0.78 0.82
LAC Brazil 1990s ⫺0.04 0.01 0.06
LAC Chile 1990s 2.24 1.81 1.37
LAC Colombia 1990s ⫺1.06 ⫺1.10 ⫺1.15
LAC Costa Rica 1990s 1.10 0.91 0.72
LAC Dominican Republic 1990s 1.76 1.47 1.18
LAC Ecuador 1990s ⫺1.23 ⫺1.06 ⫺0.90
LAC El Salvador 1990s 0.27 0.25 0.22
LAC Guatemala 1990s 0.32 0.34 0.35
LAC Guyana 1990s .. .. ..
LAC Honduras 1990s ⫺1.37 ⫺1.45 ⫺1.54
LAC Mexico 1990s ⫺0.06 ⫺0.11 ⫺0.17
LAC Nicaragua 1990s ⫺0.49 ⫺0.20 0.09
(Table continues on the following page.)

Breaking the Barriers to Higher Economic Growth200
Annex Table 1. (continued)
TFP TFP TFP
Region Country Decade (α= 0.3) (α= 0.4) (α= 0.5)
LAC Panama 1990s 1.00 0.73 0.45
LAC Paraguay 1990s ⫺1.22 ⫺1.34 ⫺1.45
LAC Peru 1990s ⫺0.31 ⫺0.20 ⫺0.10
LAC Uruguay 1990s 1.27 1.19 1.11
LAC Venezuela 1990s ⫺1.79 ⫺1.44 ⫺1.08
MENA Algeria 1990s ⫺2.38 ⫺1.99 ⫺1.60
MENA Bahrain 1990s 0.23 0.36 0.49
MENA Egypt 1990s 0.58 0.73 0.88
MENA Iran 1990s 0.02 0.16 0.31
MENA Jordan 1990s ⫺0.81 ⫺0.55 ⫺0.29
MENA Kuwait 1990s ⫺2.10 ⫺1.79 ⫺1.49
MENA Morocco 1990s ⫺1.14 ⫺1.15 ⫺1.15
MENA Saudi Arabia 1990s 0.47 0.88 1.28
MENA Syria 1990s 0.68 0.89 1.10
MENA Tunisia 1990s 0.51 0.55 0.59
OECD Australia 1990s 1.00 0.85 0.69
OECD Austria 1990s 0.68 0.47 0.26
OECD Belgium 1990s 0.63 0.45 0.27
OECD Canada 1990s 0.70 0.60 0.50
OECD Denmark 1990s 1.52 1.33 1.13
OECD Finland 1990s 1.05 1.04 1.04
OECD France 1990s ⫺0.19 ⫺0.27 ⫺0.36
OECD Greece 1990s 0.35 0.37 0.39
OECD Iceland 1990s 0.57 0.53 0.49
OECD Ireland 1990s 3.97 3.83 3.68
OECD Israel 1990s 0.46 0.21 ⫺0.05
OECD Italy 1990s 0.12 0.03 ⫺0.05
OECD Japan 1990s ⫺0.53 ⫺0.80 ⫺1.06
OECD Netherlands 1990s 1.16 1.04 0.92
OECD New Zealand 1990s 0.66 0.62 0.57
OECD Norway 1990s 2.32 2.26 2.20
OECD Spain 1990s 0.01 ⫺0.14 ⫺0.29
OECD Sweden 1990s ⫺0.29 ⫺0.20 ⫺0.11
OECD Switzerland 1990s ⫺0.39 ⫺0.52 ⫺0.65
OECD United Kingdom 1990s 0.82 0.65 0.48
OECD United States 1990s 0.95 0.75 0.56
S. Asia Bangladesh 1990s 0.87 0.45 0.02
S. Asia India 1990s 1.42 1.16 0.90
S. Asia Myanmar 1990s 3.40 2.97 2.54
S. Asia Pakistan 1990s 0.92 0.69 0.46
S. Asia Sri Lanka 1990s 1.38 1.13 0.87
Africa 1960s ⫺0.01 ⫺0.49 ⫺0.97
E. Asia 1960s 1.07 1.04 1.01
ECA 1960s 3.42 3.07 2.72
LAC 1960s 1.46 1.20 0.94
MENA 1960s 2.36 1.96 1.55

Part II: Labor Markets and Human Capital 201
TFP TFP TFP
Region Country Decade (α= 0.3) (α= 0.4) (α= 0.5)
OECD 1960s 2.18 1.67 1.15
S. Asia 1960s 0.62 0.27 ⫺0.09
World 1960s 1.28 1.01 0.74
World
(excluding China) 1960s 1.37 0.97 0.56
Africa 1970s ⫺0.07 ⫺0.31 ⫺0.55
E. Asia 1970s 1.61 1.18 0.76
ECA 1970s 0.58 0.08 ⫺0.42
LAC 1970s 1.23 0.91 0.59
MENA 1970s ⫺0.35 ⫺0.96 ⫺1.56
OECD 1970s ⫺0.17 ⫺0.39 ⫺0.62
S. Asia 1970s ⫺0.57 ⫺0.66 ⫺0.76
World 1970s 0.52 0.22 ⫺0.07
World
(excluding China) 1970s 0.01 ⫺0.26 ⫺0.53
Africa 1980s ⫺0.62 ⫺0.70 ⫺0.77
E. Asia 1980s 3.58 3.08 2.59
ECA 1980s 0.95 0.89 0.83
LAC 1980s ⫺2.37 ⫺2.29 ⫺2.22
MENA 1980s ⫺1.39 ⫺1.44 ⫺1.48
OECD 1980s 0.92 0.71 0.50
S. Asia 1980s 1.79 1.56 1.33
World 1980s 1.62 1.34 1.07
World
(excluding China) 1980s 0.58 0.39 0.19
Africa 1990s ⫺0.07 ⫺0.11 ⫺0.15
E. Asia 1990s 4.11 3.32 2.54
ECA 1990s ⫺0.57 ⫺0.55 ⫺0.54
LAC 1990s 0.02 0.03 0.04
MENA 1990s ⫺0.22 ⫺0.03 0.15
OECD 1990s 0.45 0.28 0.11
S. Asia 1990s 1.39 1.11 0.83
World 1990s 1.88 1.51 1.13
World
(excluding China) 1990s 0.55 0.37 0.18
Appendix Notes
1. Nehru and Dhareshwar, 1993.
2. In the case of MENA economies, where there were inconsistencies, the World Bank MENA regional database investment series
was preferred.
3. Barro and Lee, 2000. Educational attainment data (available until 1999) were extended to 2000 assuming constant growth
between 1995-2000.

203
Challenges and Opportunities for
the 21st Century:
Higher Education in the Middle
East and North Africa
Mustapha Nabli
8
Good morning. It is a privilege for me to be here today to address this distin-
guished group, on the opening day of a conference that has tremendous
importance for the Middle East and North Africa region, from my point of
view. Particularly because this conference on tertiary education in the region
comes at a time in which new research is changing the way in which we view
the importance of higher schooling.
Over the last decade, the focus of development practitioners on education-
al priorities has undergone significant change. Until the beginning of the
1990s, many economists, including those within my own institution, main-
tained that developing economies should give highest priority to improving
access to primary and secondary education. They generated the greatest rates
of return, and thus justified government investment and relatively high Bank
intervention in these levels. More than a decade of empirical analyses seemed
to demonstrate that higher education offered lower private and social returns
than basic education. Considering that higher education absorbs considerably
higher investment, a powerful justification was provided for focusing public
educational investment at the basic level. This justification was further rein-
forced by the obvious gains in social equity associated with such a strategy. As
Speech at the Conference on Higher Education in the Middle East and North Africa: Challenges and
Opportunities for the 21st Century; Institut du Monde Arabe; Paris, France; May 23, 2002.

Breaking the Barriers to Higher Economic Growth204
a result, there has been a preference for basic education in the developing
world. In contrast, higher education was relegated to a relatively minor place
on the development agenda.
But over the 1990s, we have seen more and more literature challenging the
old orthodoxy that basic education should be the focus in the education
strategies of developing economies. We now understand, as I am sure my col-
leagues at this conference will better argue, the critical importance of having
a solid base of higher education in every economy—no matter what the level
of income—to support its development strategy. Economic development is
increasingly linked to a nation’s ability to acquire and apply technical and
socioeconomic knowledge. Highly skilled workers, and the capacity for tech-
nological innovation dissemination, are a foundation of economic develop-
ment, not only for economies moving to a higher-technology production
base, but of equal importance for low- and middle-income economies to
obtain increased returns from their large stock of low- or unskilled labor.
Comparative advantages come less and less from abundant natural resources
or cheaper labor, and more and more from technical innovations and the
competitive use of knowledge.1My colleague, Jamil Salmi, will be expanding
upon this theme this afternoon. Prosperity in agriculture is more and more
dependent upon technological innovations and their diffusion. Labor-inten-
sive industries that cannot modernize with new technologies find themselves
increasingly unable to compete.
The strategies of the East Asian economies demonstrate the importance of
higher education with other educational advancements in promoting growth.
The mainstay of their success stories has been the acquisition of foreign tech-
nology and production methods, which were adapted and improved upon by
well-educated nationals. Rapid growth of labor-intensive manufacturing was
made possible, with a highly trained set of individuals to put in place the
structures to achieve greater productivity. Agricultural productivity was
improved through adaptive local research. In countries without a sufficient
stock of higher-educated nationals—Indonesia, Malaysia, Singapore, and
Thailand—there was heavy reliance on multinational corporations that sub-
stituted foreign for missing local skills.
And the empirical evidence has materialized to support this notion. The
pursuit and use of higher education is a critical factor affecting economic
growth. In fact, at least some empirical studies have shown that among low-
and middle-income economies, an increase in the educational stock at higher
education levels promotes as much growth, or more rapid growth, than the
same amount of increase in educational stock at lower levels does.
But, despite these new findings, what remains troubling in my mind is that
the MENA region has been unable to realize the gains from its substantial
investments in education. This applies not only to higher education but to

Part II: Labor Markets and Human Capital 205
lower levels of education, as well. Despite really extraordinary progress over
the last decades in increasing the level of education throughout the region, the
pay-offs have been very disappointing from an international context.
And what I would like to talk about today is how the MENA region can
best improve the social and private returns from its higher education
systems—in an increasingly globalized world. We often spend great amounts
of time discussing how educational systems should be adapted to meet the
needs of a more globalized world; how universities must better equip stu-
dents to respond to new educational demands and opportunities; how
organizations of higher learning need to meet better the growing demand
among learners for improved accessibility and convenience, lower costs,
direct application of content to work settings, and greater understanding of
the dynamic complexity, and often interdisciplinary nature, of knowledge;
how education needs to foster greater flexibility, so that workers can adjust to
new market demand; how education needs to be more responsive to private
sector needs; how skills need to be made more relevant. Indeed, these are
important issues.
But we often approach the issue of educational reform from an internal
point of view, asking how institutes or organizations of learning can adapt to
meet better the demands of development. What is more rarely discussed is
how the ability for educational systems to adapt to new demands and oppor-
tunities is fundamentally affected by policies outside of the educational sector.
There is a large body of literature that has focused on estimating the rela-
tionship between human capital and growth. From these studies, we have
increasingly realized that the relationship depends strongly on the national
context, which determines whether skilled labor is effectively utilized in pro-
ductive activities. Unlike physical capital, human capital responds to incen-
tives that increase its own private return, but sometimes the higher private
rate of return is not derived from growth-enhancing activities, but rather
from rent seeking.2
What I would like to focus my talk on today is the emerging consensus that
the ability for an economy to leverage its stock of human capital is intrinsical-
ly determined by policies that fall outside the traditional area of education
reform. Education can only be internationally competitive if work is. And so,
key to honing the higher education sector to becoming more internationally
competitive are identifying and removing those barriers to the productive use
of education in the economy. Education and its use in the economy is a virtu-
ous circle—if competitive, relevant skills are demanded and rewarded in the
economy, and there will be increased demands on the educational sector to
respond. But sustainable improvements to the higher educational system, by
changes in its supply,cannot be achieved if, in the end, there are few incentives
in place for its productive use.

Breaking the Barriers to Higher Economic Growth206
What are the factors that inhibit the demand for more “productive”skills—
meaning skills that not only garner private returns, but that also are produc-
tively utilized in the economy?
To begin, what is probably the most obvious, and has been most greatly
explored in the development literature, is the degree of public sector employ-
ment. The MENA region maintains one of the highest levels of public sector
employment in the world, and civil service employment as a proportion of
total employment is the highest in the world. In the mid-1990s, civil service
employment, including in health and education, accounted for 17.5 percent
of total employment in the region. In comparison, civil service accounted for
16 percent of total employment in Eastern Europe and the former Soviet
Union, 6.3 percent in Asia, 6.6 percent in Africa, and 8.9 percent in Latin
America and the Caribbean. Still, these figures do not tell the entire story.
The majority of MENA’s civil service employment is not in the health and
education sectors but in government administration. In fact, some 10.5 per-
cent of employment in the MENA region is in government administration.
That compares with 4.2 percent in Eastern Europe and the former USSR, 4
percent in Asia, 4 percent in Africa, and 5.4 percent in Latin America and the
Caribbean.
Until recently, public sector employment was almost a guarantee in the
region for persons with higher or intermediate education. As a result, individ-
uals have often sought higher degrees—with little attention to content or
quality. Public sector employment is characterized by a large number of work-
ers possessing education that bears little relation to the skills needed by either
the private sector or for the efficient functioning of the public sector. Most
public sector jobs are protected, so there is a large element of rent seeking on
the part of those who secure them.
In many countries in the region, new graduates have either obtained or
queued for government jobs, giving rise to a very unusual pattern of unem-
ployment. Unemployment in the region is almost always highest precisely for
those individuals with advanced educations. Obviously, this should raise sig-
nificant alarm for policy makers when addressing higher education reform, if
investments in higher education go unused.
What has been the cost of an overstaffed public sector? A recent study
examined the relationship between public sector employment and economic
growth. Generally, cross-country growth studies are based upon explanations
of economic growth by increases in factors of production and TFP—that por-
tion of economic growth that cannot be explained by accumulation. But not
all factor accumulation goes toward growth. One could argue that human
capital that is diverted to the administrative civil service does not. In this par-
ticular study, the authors deducted the proportion of human capital diverted
to the administrative civil service from the country’s human capital stock, and

Part II: Labor Markets and Human Capital 207
ran traditional growth regressions. Then, from those estimates,they could cal-
culate how much more growth countries would have realized if the human
capital diverted to public administration had gone into the productive econ-
omy.3From this calculation, it was estimated that for the MENA region, the
loss of GDP growth between 1985 and 1995, strictly as a result of public
administration employment, was some 8.4 percent—or close to 1 percentage
point per year.4
But there are other factors relevant to the region that have prevented the
application of education to productive use. One factor that has been the sub-
ject of recent cross-country exploration is the degree of openness. When stud-
ies have examined the relationship between openness and human capital on
total factor productivity, they have found that outward-oriented economies
experience higher total factor productivity, but over and above the positive
effect of openness. Rather, outward orientation is critical for allowing human
capital to have a positive impact on economic growth. The effect of the stock
of human capital on total factor productivity is conditional on the degree of
openness.5
Greater openness fosters competition, encourages modern technology,
increases the demand for highly skilled labor,and promotes learning by doing.
Openness obliges industries to confront their inefficiencies. To compete suc-
cessfully, industries must adapt, thereby creating demand for the new skills
and trades to do so. More specific mechanisms associated with trade liberal-
ization provide incentives for maximizing the private rate of return to educa-
tion through growth-enhancing activities. Too little openness, therefore, does
not allow a country to benefit as greatly from its investments in human capi-
tal. Human capital investment without liberalization of the external sector
may lead to the underutilization of human capital.6
What is the dynamic influence that trade exerts over the use of human cap-
ital? One recent study points to three possible avenues: First, the contact of
trade—the trading technology itself—is skil-intensive. Trade needs packaging
and paperwork that conforms to international standards. It requires knowl-
edge of international markets. It requires knowledge of distribution channels,
legal requirements, government regulations, and import restrictions. The
process of trade requires skills that only educated labor can have.
Second, traded goods are usually of higher skill content than those pro-
duced and consumed at home, especially for developing countries, since they
must compete with foreign suppliers. Once trade has been liberalized, and the
protective barrier of tariffs and direct restrictions has been lifted, home pro-
ducers need to invest in new technology to compete in the global marketplace.
Opening firms to competition provides the incentives for the efficient use of
existing knowledge, the creation of new knowledge, and the flourishing of
entrepreneurship.

Breaking the Barriers to Higher Economic Growth208
Third, the exposure to the foreign markets that trade brings enhances the
knowledge of domestic producers of superior technologies. Such technologies
are usually complementary to skills, because of the abundance of skilled labor
in more advanced countries. Their importation increases the wage premium
paid to skilled labor in developing countries, and so encourages the redeploy-
ment of skilled labor to trade-related activities.7
And so, I would argue today that, along with discussing policy options for
improving the relevance of higher education in MENA, at least some attention
needs to be paid to removing the barriers that insulate the region from glob-
alization, so that the incentive structure for educational attainments to have
their greatest effect on economic and social development is in place.
Trade policy in the region remains one of the most restrictive in the world,
with low level of and speed of integration into the world economy. Tariff rates
remain high and the extent of nontariff barriers large. A number of policy
moves across the region are expected to lead to greater trade openness, but the
region’s progress has been uneven, and in many cases has been superficial at
best. But among the positive moves is the EU association agreements signed
by Tunisia, Morocco,Jordan, Algeria, Egypt, and Lebanon. These policy moves
are expected to stimulate integration, and, one hopes, compel workers to
articulate better their demands of the educational system, based upon expect-
ed returns from productive activities.
In addition to creating economies more open to trade to compel their
potential workforce to seek more relevant skills, more effective use of educa-
tion in the region hinges upon developing a dynamic information infrastruc-
ture to facilitate the effective communication, dissemination, and processing
of information. At first sight, this factor may seem to lie squarely within edu-
cational sector reform. Indeed, much of it is. But, as the focus of my talk is to
detail the factors prohibiting the productive use of, and thereby qualitative
demands for, higher education, the lack of a well-developed information
infrastructure and the other dimensions of the knowledge economy also
impact the way in which education is used.
To begin with, knowledge allows industries to tap into more efficient pro-
duction methods, and in doing so, to better identify skill needs. As a result,
with a keener sense of the precise skills needed to operate effectively, their ulti-
mate workforce has a greater potential for efficiently using educational
attainments.
In addition, the vast availability of knowledge allows workers to discern
better what skills are demanded in the market. This is of immense importance
to sustainable improvements in the higher education sector. Perfect informa-
tion, in which market opportunities and returns for the skilled labor market
are known, provides the ultimate guide for how the higher education system
must respond.

Part II: Labor Markets and Human Capital 209
And finally, knowledge is key to integration. Developed countries are mov-
ing quickly toward integrated “knowledge-based economies,” in which knowl-
edge, information, and communication technologies are becoming keys to
competitiveness and engines for social and economic development. Not par-
ticipating in the knowledge revolution presents the risk to the region of being
further marginalized. As the knowledge gap between developed and develop-
ing economies widens, the potential for successfully competing in the global
economy shrinks. This is a risk that is all too dangerous for a region that has
embraced global integration less than enthusiastically. If this fundamental
tool for being internationally competitive is not widely available, the potential
for reverting to inward-oriented strategies is great.
I would point to a final feature of the MENA economies that undermines
the efficient use of education, and that is the myriad governance issues that
impede efficient business operations, private investment, and entrepreneur-
ship and provide an enabling environment for rent seeking. While the com-
prehensive macroeconomic and structural reform programs espoused by
many of the MENA economies in the early 1990s created an exuberant boost
in their economic outlooks, the MENA region must move beyond “stroke of
the pen”reforms to the more serious, and challenging, issues that obstruct the
development of a strong private sector. Unless the private sector begins to see
itself as an independent source of growth and productivity in the economy,
and society begins to underpin this change economically and politically, the
region will continue to find its educational enhancements underused.
In closing, I want to say that I sincerely welcome this conference on higher
education in MENA, as it gives us the opportunity to explore the multiple
dimensions of education and its applications. But I would add a caveat as we
explore the many layers of educational reform. Improving access to and con-
tent of higher education in MENA is of critical importance in the region’s eco-
nomic and social development. But greater investment in higher education is
not a magic bullet. While expansion in higher education is important in the
context of economic growth, the economic context in which these services are
provided is equally important. Human resource development policy must be
designed in conjunction with an overall development strategy, to allow edu-
cation to find its most effective use in the pursuit of higher economic growth,
the expansion of employment opportunities, and the reduction of poverty.
Thank you.
Notes
1. Jamil Salmi June 2000. “Tertiary Education in the Twenty-First Century: Challenges and
Opportunities.” Human Development Department LCSHD Paper Series No. 62. World Bank.
2. Christopher Pissarides. “Human Capital and Growth: A Synthesis Report.” OECD Development
Center Technical Paper No. 168; 2000.

Breaking the Barriers to Higher Economic Growth210
3. Letting the human capital stock diverted to public administration realize the same rates of return as
the human capital stock not in public administration.
4. Ibid.
5. Stephen M. Miller and Mukti P. Upadhyay. “The Effects of Openness, Trade Orientation, and
Human Capital on Total Factor Productivity.” Journal of Development Economics, Vol. 63. 2000.
6. Ibid.
7. Pissarides, 2000.

211
Labor Market Reforms, Growth,
and Unemployment in
Labor-Exporting Countries in the
Middle East and North Africa
Pierre-Richard Agénor*
Mustapha K. Nabli
Tarik Yousef†
Henning Tarp Jensen††
9
As in other developing regions in the post-World War II era, the demograph-
ic transition in the Middle East and North Africa region has given rise to rapid
labor force growth. But with an average growth rate above 3 percent since the
1960s, no other region comes close to the magnitude and persistence of
MENA’s labor market pressures. And while employment growth was relative-
ly strong in the 1970s, it failed to keep pace with the expansion of the labor
force during the 1980s and 1990s. As a result, MENA recorded some of the
highest unemployment rates among developing regions in the 1990s. Recent
estimates indicate that unemployment rates range from about 2.3 percent in
the United Arab Emirates to close to 29.8 percent in Algeria (see World Bank
2004). For the region as a whole, the unemployment rate is currently estimat-
ed at 15 percent of the labor force.1Based on current trends in job creation,
the prospects for absorbing new entrants into labor markets, as well as those
currently unemployed, are rather bleak. With the labor force growing in the
*World Bank. †Georgetown University. ††University of Copenhagen, Denmark. The authors are grate-
ful to various colleagues for helpful comments on a preliminary version.
Published in Journal of Policy Modeling 29, (2007). Reprinted with permission from Elsevier Limited;
see pages 277–309.

Breaking the Barriers to Higher Economic Growth212
present decade at 3.4 percent per annum, the average unemployment rate for
the region could reach 22 percent by 2010.
Low output growth is often the “proximate”cause for the rise in unemploy-
ment. Following the oil bust in the late 1980s, the region experienced a weak
recovery in the 1990s due to the protracted pace of policy reforms. But there
are also structural reasons for unemployment that were not addressed by pol-
icy makers in the past decade. MENA countries exhibit rigidities in education-
al systems, wage setting, and regulatory regimes due to the dominance of the
public sector in labor markets. Although rates of human capital accumulation
have remained steady, MENA has reaped less than its potential in terms of
economic growth and job creation. Government legislation on hiring and
firing, minimum wages, and collective bargaining agreements, as well as
employment guarantees in the public sector, have interfered with the efficient
functioning of the labor market.2In normal times, the impact of some of
these structural rigidities may well be mitigated by the existence of large infor-
mal sectors. However, in periods of significant structural changes, they could
turn into binding constraints on the expansion of output and employment. In
such conditions, reforming the labor market becomes an important element
of any reform program aimed at stimulating growth and promoting job
creation.
Accordingly, the purpose of this paper is to offer a quantitative analysis of
the impact of labor market reforms on growth, real wages, and unemploy-
ment in labor-exporting MENA countries.3We begin with a brief overview of
the main features of the labor market in five labor-exporting MENA coun-
tries, namely, Algeria, Egypt, Jordan, Morocco, and Tunisia.4The next section
presents a quantitative framework that captures many of these features (such
as a large informal urban sector, active trade unions, public sector employ-
ment, and international labor flows),as well as other important structural fea-
tures of these countries, such as an unfunded pension system.5Following is a
discussion of the calibration procedure and parameter values. Finally, we pres-
ent simulation experiments focusing on four types of individual policy
shocks: a reduction in payroll taxation on unskilled labor, reductions in pub-
lic sector wages and the size of the government workforce, higher employment
subsidies to the private sector, and a reduction in the bargaining power of
trade unions.
We also consider a “composite” reform package, involving a cut in payroll
taxes and public sector employment, as well as a reduction in unions’ bargain-
ing strength. The extent to which high payroll taxes have tended to discourage
the demand for (unskilled) labor has been an important policy issue in MENA
countries in general. Our framework allows us to consider the implications of
both revenue- and budget-neutral changes in these taxes, and the various
channels through which they affect job creation and unemployment. The con-
cluding section summarizes the results and draws together the main policy

Part II: Labor Markets and Human Capital 213
lessons of the analysis. It emphasizes the need for an overall package of
reforms, involving not only labor market policies but also other structural
measures, to foster sustained growth in output and employment in labor-
exporting MENA countries.
Some Basic Facts
The functioning of the labor market in MENA countries in general, and labor-
exporting countries in particular, has been reviewed in a number of recent
contributions.6Here we briefly review some of the salient features of this mar-
ket (as summarized in figure 9.1), in order to motivate the specification of the
model developed in the next section.
Fundamentally, the labor market in labor-exporting MENA countries can
be characterized as consisting of three segments: the rural sector, which con-
tinues to employ a sizable proportion of the labor force in many countries; the
informal urban sector, characterized mostly by self-employment and a limit-
ed proportion of hired labor, a high degree of wage flexibility, low employ-
ment security, and no enforcement of labor regulations; and the formal (pub-
lic and private) urban sector, where workers are hired on the basis of explicit
contracts and the degree of compliance with labor regulation (particularly in
the public sector) is relatively high.
The informal sector accounts for a large fraction of the labor force.
Estimates of informal employment range from a low of 42 percent of nona-
gricultural employment in Syria to a high of 55 percent in Egypt. Although
moderate compared to other developing regions, these estimates are high
given the large share of public sector employment. In most labor-exporting
MENA countries, the public sector is the dominant employer in the formal
sector (see Abrahart, Kaur, and Tzannatos 2002). When measured as a per-
centage of nonagricultural employment, the public sector is the highest
among developing regions. Governments are often considered as “employers
of first resort,” especially for people with middle and higher education levels.
The perpetuation of employment guarantees in government hiring, and mis-
matched wage expectations resulting from generous public sector compensa-
tion and benefits policies, have contributed to the continued preference for
public sector jobs.
Although open unemployment has increased in recent years, underem-
ployment remains pervasive.7Open and disguised unemployment (which
affects disproportionately the young and women) amount to anywhere
between 25 and 60 percent of the labor force in some countries. A large major-
ity of the openly unemployed have secondary or postsecondary degrees, but
open unemployment is also becoming more widespread among unskilled
workers. Part of this unemployment is “queueing” or “wait” unemployment,
resulting from public sector hiring and wage-setting practices, as shown for

Breaking the Barriers to Higher Economic Growth214
Figure 9.1. Labor-Exporting MENA Countries: Economic and Labor Market Indicators
Algeria
Egypt
Jordan
Morocco
Tunisia
average
0 20 40 60 80
Algeria
Egypt
Jordan
Morocco
Tunisia
average
0 5 10 15 20
Algeria
Egypt
Jordan
Morocco
Tunisia
average
0 25 50 75 100 125
Algeria
Egypt
Jordan
Morocco
Tunisia
average
0 10 20 30 40 50 60
Algeria
Egypt
Jordan
Morocco
Tunisia
average
0 10 20 30 40
Algeria
Egypt
Jordan
Morocco
Tunisia
average
0 10 20 30 40 50
Algeria
Egypt
Jordan
Morocco
Tunisia
average
0 1 2 3 4 5 6 7 8 9
Algeria
Egypt
Jordan
Morocco
Tunisia
average
0 5 10 15 20
Labor force with primary education
(% of total)
Workers remittances, net
(in % of GDP)
Size of agricultural sector
(% of GDP)
Employment in informal sector
(% of total employment)
Degree of openness
(total trade in % of GDP)
Employment in public sector
Public sector wage bill
(% of public expenditure)
Agricultural exports
(% of total exports)
2001
2001
2001
2001
2001
1985
1986
1987
1982
1980
1992
1996
1996
1994
1990
1992
1991
1992
1989
1999
1997
2001
1999
1991
1999
1999
1999
1999
1999
1991
2001
2001
2001
2001
2001
2001
2001
2001
2001
Note: Years are given next to bars.

Part II: Labor Markets and Human Capital 215
instance by Assaad (1997) in the case of Egypt. The unemployed are essential-
ly those who would have had a chance at a formal job in the public sector in
the past and continue to have expectations of acquiring such a job. Those with
no education must either accept whatever employment is available to them,
no matter how casual, or create their own jobs in order to survive.
Labor market regulations, including restrictions on hiring and firing, as
well as minimum wage legislation, are widespread in the region. In all labor-
exporting MENA countries (except Jordan), there is a minimum wage regula-
tion, although its impact on wage formation is not always clear.8In Egypt and
Tunisia, compliance with minimum wages is mostly limited to the public sec-
tor. In recent years, high unemployment in Morocco has led authorities to
allow the private sector to hire workers at wages below the minimum rate.
Restrictions on layoffs in the formal sector (and often, generous severance
payments) make firing redundant workers difficult in most labor-exporting
MENA countries. In practice, however, the enforcement of the law is weak;
compliance with existing regulations is limited to the formal sector. Thus,
although labor market regulations may be pervasive on paper, their impact is
mitigated by weak enforcement and the existence of large informal sectors.
Wage determination often departs from market-clearing mechanisms as a
result of legal restrictions, the existence of labor unions, and imperfectly com-
petitive wage-setting behavior by firms. Wages in agriculture and the urban
informal sectors tend to be highly flexible. In contrast, some urban formal sec-
tors show rigid systems that are characterized by segmentation and binding
institutional constraints.9In most countries, civil service pay remains a point
of reference for public enterprises and many large firms in the formal private
sector.This “leadership effect” of public wage settlements is a source of down-
ward rigidity in wage behavior in the private sector. Among nonwage labor
costs, social security contributions (which are typically shared between
employers and employees) are particularly significant. In Algeria, contribu-
tions to the social security system alone constitute more than 36 percent of
total labor costs.
With the exception of Jordan, where collective bargaining is practically
nonexistent, labor unions in Algeria, Egypt, Morocco, and Tunisia play a
significant role in collective bargaining at the national level (see for instance
Assaad and Commander (1994) for Egypt). This occurs despite the fact that
actual unionization rates are relatively low in most of these countries (except
Egypt) and union membership tends to be primarily in the public sector. The
trade union movement is usually highly centralized, except in Morocco; its
influence on wage formation is often through the political process (by lobby-
ing to secure increases in minimum wages, for instance) rather than through
industrial action, such as strikes and other forms of work disruptions.
Through their influence on political parties, unionized workers are also able
to exert considerable pressure to maintain job security.

Breaking the Barriers to Higher Economic Growth216
Finally, international migration flows are an important source of foreign
exchange and income for all of these countries. In the 1970s and 1980s, the
peak years of oil-led growth in the region, the Gulf countries experienced
unprecedented labor force growth, driven primarily by the large number of
immigrants seeking work from the labor-exporting MENA economies, espe-
cially Egypt and Jordan. During the same period, millions of migrants from
Morocco, Tunisia, and, (to a lesser extent) Algeria sought work in Europe. The
1990s witnessed a sharp fall in the outflows of workers from the sending
countries in the region even though remittances remained an important
source of income (see figure 9.1). These flows play a significant role in the
adjustment of the domestic labor market, in ways that we discuss more
specifically below.
A Formal Framework
We now describe a quantitative framework to analyze the impact of labor
market reforms in labor-exporting MENA countries.10 The model captures
many of the structural features of the labor market highlighted in the forego-
ing discussion. In this section, we briefly summarize the main features of the
model (focusing on the production structure, the labor market, and the pen-
sion system), with a complete list of equations provided in Appendix 9.A, and
variable definitions in Appendix 9.B.
Production
The composition of output and the structure of the labor market are summa-
rized in figure 9.2. The basic distinction on the production side is that
between rural and urban sectors. The rural sector (or agriculture) produces
only one good, which is sold both on domestic markets and abroad. Urban
production includes both formal and informal components; in addition, the
formal urban economy is separated between production of a private good and
a public good. Land available for production in agriculture is in fixed supply.
Gross output in the rural sector, as well as in all other sectors, is given by the
sum of value added and intermediate consumption (Equation A1). Value
added is assumed to be produced with a Cobb-Douglas function of a compos-
ite factor, defined as a function that depends on the number of unskilled rural
workers employed in agriculture and the economy-wide stock of public phys-
ical capital (Equation A2).The presence of public physical capital in the pro-
duction function of the agricultural good is based on the view that a greater
availability of public physical capital in the economy (roads, storage facilities,
power grid, and the like) improves the productivity of large-scale producers
and other production units in agriculture, because it facilitates not only trade
and domestic commerce but also the production process itself. For simplicity,

Part II: Labor Markets and Human Capital 217
Figure 9.2 A Stylized View of the Labor Market in Labor-Exporting MENA Countries
Rural sector Urban sector
Formal sector
Private sector good
Public sector
Production
Labor
force
Skilled labor
supply
Skills acquisition
Skilled unemployment
Internal
migration
Unskilled labor demand
formal sector
Skilled labor demand Private
skilled wage
UNSKILLED MINIMUM WAGE
Unskilled labor
supply
informal sector
Informal
sector wage
Unskilled labor demand
informal sector
Labor demand
Agricultural
sector wage
Union's reservation wage
Rural labor supply
Urban
unskilled
labor
supply
PAYROLL TAXES
TRADE
UNIONS
SKILLED
PUBLIC
WAGE
UNSKILLED
PUBLIC
WAGE
Unskilled
unemployment
PUBLIC
UNSKILLED
EMPLOYMENT
Unskilled labor
supply formal
sector
Expected
urban
unskilled
wage
International
migration
FOREIGN
WAGE
Stock of
unskilled
workers
abroad
Foreign
remittances PUBLIC
CAPITAL IN
EDUCATION
TEACHERS IN
THE PUBLIC
SECTOR
PUBLIC
SKILLED
EMPLOYMENT
UNSKILLED MINIMUM WAGE Expected
unskilled private
urban wage
Informal sector
(nontraded)
the area of land allocated to production is normalized to unity. Agricultural
production exhibits decreasing returns to scale in the remaining (composite)
input.
Value added in the informal economy is given as a function of the number
of unskilled workers employed there, with decreasing returns to scale
(Equation A4). Value added in the public sector is generated by combining
skilled and unskilled labor using a CES function (Equation A5). Employment
levels of both categories of workers are treated as exogenous.
Private formal production uses as inputs skilled and unskilled labor, as well
as physical capital. Skilled labor and private physical capital have a higher

Breaking the Barriers to Higher Economic Growth218
degree of complementarity (lower degree of substitution) than physical capi-
tal and unskilled workers. In order to account explicitly for these differences
in the degree of substitutability among inputs, we adopt a nested production
structure. At the lowest level, skilled labor and private capital are combined to
form a composite input with a low elasticity of substitution between them
(Equation A8). At the second level, this composite input is used together with
unskilled labor to form a second composite input (Equation A7). The elastic-
ity of substitution between the first composite input and unskilled workers is
taken to be higher than between skilled employment and private capital. The
final layer combines the second composite input and the stock of government
capital as production inputs (Equation A6).
The Labor Market
Unskilled workers in the economy may be employed either in the rural econ-
omy, UR, or in the urban economy, UU, whereas skilled workers are employed
only in the urban formal sector.
Agriculture and Internal Migration The demand for labor in the agricul-
tural sector, UA
d, can be derived from profit maximization as
(1)
where VAis value added in the agricultural sector, WAdenotes the nominal
wage, and PVAis the value-added price (net of input costs) in the agricultur-
al sector.
Nominal wages in agriculture adjust to clear the labor market. Let UR
s
denote labor supply in the rural sector; the equilibrium condition is thus
given by
(2)
Over time, UR
sgrows at the exogenous population growth rate, gR,net of
worker migration to urban areas, MIG:
(3)
In the spirit of Harris and Todaro (1970), the incentives to migrate are
taken to depend negatively on the ratio of the average expected consumption
wage in rural areas to that prevailing in urban areas. Unskilled migrant work-
ers may be employed either in the private formal sector, in which case they are
paid a minimum wage, WM, or they can enter the informal economy and
receive the market-determined wage in that sector, WI. When rural workers
make the decision to migrate to urban areas, they are uncertain as to which
UU g MIG
RR R
=+
()
−
−,11
UUV
W
PV
R
s
A
d
A
A
A
=⎛
⎝
⎜⎞
⎠
⎟
,.
UV w
A
d
A
XA
A
XA
XA
XA
XA
XA
=−⋅
⎛
⎝
⎜⎞
⎠
⎟
+−
+
11
1
1
1
ρ
η
ρ
ρ
ηβ
α
XXA
wW
PV
A
A
A
,,where =

Part II: Labor Markets and Human Capital 219
type of job they will be able to get, and therefore weigh wages in each sector
by the probability of finding a job in that sector. These probabilities are
approximated by prevailing employment ratios. Finally, migrants consider
what their expected purchasing power in rural and urban areas will be,
depending on whether they stay in the rural sector and consume the “typical”
basket of goods of rural households, or migrate and consume the “typical”
urban basket of goods.
The expected, unskilled urban real wage, EwU, is thus a weighted average of
the minimum wage in the formal sector and the going wage in the informal
sector, deflated by the urban consumption price index, PURB:
(4)
where θUis the probability of finding a job in the urban formal sector, meas-
ured by the proportion of unskilled workers in the private formal sector, rel-
ative to the total number of unskilled urban workers looking for a job in the
urban formal sector, U s
F , net of government employment, UG, in the previous
period:11
(5)
In the rural sector, the employment probability is equal to unity, because
workers can always find a job at the going wage. Assuming a one-period lag,
the expected rural consumption real wage, EwA , is thus
where PRis the rural consumption price index.
The migration function can therefore be specified as
(6)
where 0 < λm< 1 measures the speed of adjustment and σM> 0 the elasticity
of migration flows with respect to expected wages. This specification assumes
that costs associated with migration or other frictions may delay the migra-
tion process, introducing persistence in migration flows.
The Urban Sector. The public sector employs an exogenous number of
unskilled workers, UG,at the nominal wage rate WUG, whereas the demand for
unskilled labor by the formal private sector is determined by firms’ profit
MIG U Ew
Ew
U
Rm
M
U
A
m
=⎛
⎝
⎜⎞
⎠
⎟
⎡
⎣
⎢
⎢
⎤
⎦
⎥
⎥+−
()
−,ln
11
λσλ
RR
R
UMIG
,
,
,
−
−−
1
2
1
Ew W
P
A
A
R
=−
−
,
,
,
1
1
θU
P
F
s
G
U
UU
=−
−
−−
,
,,
.
1
11
Ew
WW
P
U
UM U I
URB
=+−
()
−−
−
θθ
,,
,
,
11
1
1

Breaking the Barriers to Higher Economic Growth220
maximization subject to the given minimum wage, WM. Both wages are
assumed to be fully indexed on the urban formal price index, PF
(7)
where ωMand ωUG measure exogenous real wages.
Labor demand by the formal private sector is determined by firms’ profit
maximization. We assume also that firms pay a payroll tax, at the rate 0
<ptaxU< 1 on unskilled labor. This tax is proportional to the wage bill,
WMUP. Firms also receive a nominal employment subsidy on unskilled labor
of ESU<WMper worker. Unskilled labor demand by the private sector is thus
given by
(8)
where σXP1=1/(1 + ρXP1) measures the elasticity of substitution between
unskilled labor and the composite input.
We assume, as in Agénor (2002), that mobility of the unskilled labor force
between the formal and informal sectors is imperfect, as a result of relocation
and congestion costs. Migration flows are determined by expected income
opportunities, again in line with Harris and Todaro (1970). Specifically, the
supply of unskilled workers in the formal sector (including public sector work-
ers), UF
s,is assumed to change over time as a function of the expected wage dif-
ferential across sectors, measured in real terms. Wage and employment
prospects are formed on the basis of prevailing conditions in the labor market.
Because there is no job turnover in the public sector (as noted earlier), the
expected nominal wage in the formal economy is equal to the minimum wage,
weighted by the probability of being hired in the private sector. Assuming that
hiring in that sector is random, this probability can be approximated by the
ratio between employed workers and those seeking employment during the
previous period, Ud
P,– 1/(Us
F,– 1 – UG, – 1). The expected nominal wage in the
informal economy, WI, is simply the going wage, because there are no barriers
to entry in that sector. Assuming a one-period lag, the supply of unskilled
workers in the formal sector thus evolves over time according to
(9)
where βFis the elasticity of formal sector labor supply growth with respect to
expected wages. The rate of unskilled unemployment in the formal sector,
UNEMPU, is thus given by
U
U
U
UU
W
W
F
s
F
s
P
d
F
s
G
M
I
,
,
,,
,
,
−
−
−−
−
−
=−
⎛
⎝
⎜⎞
⎠
1
1
11
1
1
⎟⎟
⎧
⎨
⎪
⎩
⎪
⎫
⎬
⎪
⎭
⎪>
β
β
F
F
,,0
UT PT
ptax W ES
P
d
UM U
XP
XP
XP
=+−
⎛
⎝
⎜⎞
⎠
⎟
1
11
1
11
()
β
αρ
σXXP1
,
WPWP
UG UG F M M F
==
ωω
,,

Part II: Labor Markets and Human Capital 221
(10)
From (A4), the demand for labor in the informal sector can be derived as
(11)
where VIis value added in the informal sector and wIthe product wage,given by
wI=WI/PVI, with PVIdenoting the price of value added in the informal sector.
The supply of labor in the informal economy, Us
I, is obtained by subtract-
ing the unskilled labor supply in the formal sector, Us
F, from the urban
unskilled labor force, UU:
(12)
The informal labor market clears continuously, so that UI
d= UI
s.From equa-
tions (11) and (12), the equilibrium nominal wage is thus given by
(13)
The urban unskilled labor supply, UU, grows as a result of “natural” urban
population growth and migration of unskilled labor from the rural economy, as
discussed earlier. In addition, a quantity SKL of urban unskilled workers
acquires skills and leaves the unskilled labor force to augment the supply of
skilled workers in the economy. We make the additional assumption that indi-
viduals are born unskilled, and therefore natural urban population growth (not
resulting from migration or skills acquisition factors) is represented by urban
unskilled population growth only,at the exogenous (gross) rate gU. Finally, there
are international migrations, the flow of which is measured by IMIG, and retire-
ment from the urban labor force, measured by δU
NPUU, which is defined below.
Thus, the urban unskilled labor supply evolves according to
(14)
where δU
NP ⱕ1 ⫹gU.
As noted earlier, the employment levels of both skilled and unskilled work-
ers in the public sector are taken as exogenous. Given that some workers retire
in every period, we have
Ug U
GUGNP
U
g
=+ −
()
−
11
δ,,
Ug UMIGSKLIMIG
UUNP
U
U
=+ −
()
+−−
−
11
δ,,
WPV V
U
IXI
II
I
s
−⎛
⎝
⎜⎞
⎠
⎟
β.
UU U
I
s
UF
s
−−.
UVw
I
d
XI I I
=
()
β/,
UNEMP UU
U
U
GP
d
F
s
=1– ()
.
+

Breaking the Barriers to Higher Economic Growth222
where gUG > 0 is the exogenous growth rate of the unskilled labor force in the
public sector.
The nominal wage that skilled workers in the public sector earn, WSG,is
also indexed on the urban formal consumption price index:
(15)
where ωSG is an exogenous real wage.
To determine wages and employment for skilled labor in the private formal
sector, we use the “right to manage” approach, in which firms bargain with
trade unions over the nominal wage, WS, and set unilaterally the level of
employment. In addition, we assume that private urban firms pay a payroll tax
on the skilled labor wage bill, at the rate 0 < ptaxS< 1, and receive a nominal
employment subsidy of ESS<WSper skilled worker. The demand for skilled
labor, Sd
P,is therefore given by
(16)
Following Booth (1995, pp. 124-26) and Layard, Nickell, and Jackman
(1991, pp. 100-3), WSis determined as follows. Assume that all private sector
firms are unionized, or equivalently that all workers belong to a single (repre-
sentative) union. Let ΩSdenote the union’s reservation wage and PROFP
firms’ profits. Under the Nash bargaining approach, the bargained wage must
solve, subject to (16),
(17)
where νis a measure of the trade union’s bargaining power. The bargained
wage must therefore satisfy the first-order condition
because ∂PROFP/∂WS= – Sd
Pby the envelope theorem (each firm will choose
employment ex post such that WSis equal to the marginal value product of
skilled labor). This yields
(18)
vW
WvWS
PROF
S
SS SW
SP
d
P
P
d
S
−=+
Ωε/,
∂
∂=∂
∂+−−=
ln ,
N
W
vS
SW
v
W
S
PROF
S
S
P
d
P
d
SSS
P
d
P
Ω0
max ( ,
WSP
d
SS
v
P
s
NSW PROF=−
⎡
⎣⎤
⎦
Ω
ST PT
ptax W ES
P
d
S
SS S
XP
XP
XP
=+−
⋅
⎛
⎝
⎜⎞
2
22
2
12
κβ
αρ
() ⎠⎠
⎟
σXP 2
.
WP
SG SG F
=ω ,

Part II: Labor Markets and Human Capital 223
where εSd
p/WS= – (∂Sd
p/∂WS)(WS/Sd
p) is the wage elasticity of the demand for
skilled labor. The term on the left-hand side of this expression measures the
proportional marginal benefit to the bargain from the proportional increase
in the skilled wage. The benefit associated with a wage increase incurs only to
the union, so it is weighted by the union’s bargaining power, ν. The first term
on the right-hand side is the union’s proportional marginal cost (the percent-
age reduction in employment due to the proportional increase in the wage),
weighted by the union’s bargaining power. The second term on the right-hand
side represents the firm’s proportional marginal cost. Condition (18) indicates
therefore that the bargained wage is set such that the proportional marginal
benefits to both parties from a unit increase in wages is exactly equal to the
proportional marginal cost to each party.
The union’s reservation wage, ΩS, is assumed to be related positively to
skilled wages in the public sector, WSG, and negatively to the skilled unem-
ployment rate, UNEMPS. Wage setting in the public sector is assumed to play
a signaling role to wage setters in the rest of the economy. When unemploy-
ment is high, the probability of finding a job (at any given wage) is low.
Consequently, the higher the unemployment rate, the greater the incentive for
the union to moderate its wage demands and boost employment. The above
expression can thus be rewritten as
where Ωo,φ1,φ2> 0, and UNEMPSis defined below. Using the implicit func-
tion theorem, it can be established that lower unemployment, higher public
sector wages, or an increase in the bargaining strength of the union, raise the
level of wages in the private sector.
Given that firms are on their labor demand curve, open skilled unemploy-
ment may emerge. The rate of skilled unemployment, denoted UNEMPS,is
given by the ratio of skilled workers who are not employed either by the private
or the public sector, divided by the total (urban) population of skilled workers:
(19)
where ST
Gis the total number of skilled workers in the public sector, engaged
in both the production of public services, SG, and education, SE
G(see below):
(20)
SSS
G
T
GG
E
=+.
UNEMP SS S
S
S
G
T
P
d
=−−,
W
WUNEMPW
WS
vPROF
S
SSSG
SW
SP
d
P
P
d
S
−−− =
−−
Ω012
φφ
ε/00,

Breaking the Barriers to Higher Economic Growth224
ST
Ggrows over time according to
(21)
where gSG > 0 is the exogenous growth rate of the skilled labor force in the
public sector.
We assume that skilled workers who are unable to find a job in the formal
economy opt to remain openly unemployed, instead of entering the informal
sector (in contrast to unskilled workers), perhaps because of adverse signaling
effects, as discussed by Agénor (2003).
The evolution of the skilled labor force depends on the rate at which
unskilled workers acquire skills:
(22)
where 0< δS< 1 is the rate of “depreciation”or “de-skilling” of the skilled labor
force.
Skills Acquisition. The acquisition of skills by unskilled workers takes place
through a free education system operated by the public sector. Specifically, the
flow of unskilled workers who become skilled, SKL, is taken to be a CES func-
tion of the “effective”number of teachers in the public sector, SE
G,and the gov-
ernment stock of capital in education, KE:
(23)
where ϕmeasures the productivity of public workers engaged in providing
education. ϕis assumed to depend on the relative wage of skilled workers in
the public sector, WSG, relative to the expected wage for that same category of
labor in the private sector, which (in the absence of unemployment benefits)
is given by one minus the unemployment rate, 1 – UNEMPS, times the going
wage, WS. Using the effort function derived by Agénor and Aizenman (1999)
yields:
(24)
where 0 <ϕM< 1 denotes the “minimum”level of effort.12
ϕϕ δ
δ
=− −
⎡
⎣
⎢
⎢
⎤
⎦
⎥
⎥
−−
−
1111
1
m
SS
SG
E
UNEMP W
W
E
()
,
,,
,
>> 0,
SKL S K
EG
E
EE
EE
E
=+−
⎡
⎣⎤
⎦
−−
−
βϕ β
ρρ
ρ
() ( ) ,1
1
SSSKL
SNP
S
=− −
()
+
−
11
δδ ,
Sg S
G
T
SG NP
S
G
T
=+ −
()
−
11
δ,,

Part II: Labor Markets and Human Capital 225
International Labor Migration. As noted earlier, international migration is an
important feature of the labor market in labor-exporting MENA countries.
We assume here that migration involves only unskilled workers, and that
potential migrants are in the urban sector (as captured in (14). Moreover,
international migration flows are taken to be determined by two factors: the
expected urban real wage for unskilled labor, Ewu, given by (4), relative to the
expected foreign wage measured in terms of the urban formal price index,
EwFOR, defined as
with WFOR denoting the foreign wage measured in foreign-currency terms,
assumed exogenous, and with ER the exchange rate. Adopting a specification
similar to (6), the migration function is specified as
(25)
where 0 < λim < 1 measures the speed of adjustment, and σIM > 0 the elastic-
ity of migration flows with respect to expected wages. Again, costs associated
with migration (such as relocation costs) are assumed to introduce some
degree of persistence. Remittances associated with international migration
flows of unskilled labor are assumed to benefit unskilled households in the
urban formal and informal sectors.13
The Pay-as-you-go Pension System
We assume that there is a pay-as-you-go pension system, whose current out-
lays to pensioners (retired workers in the urban formal sector, both public and
private), denoted PENSIONS, are financed by payroll taxes on workers in the
private formal sector and transfers from the government, TRSOC:14
(26)
Total pension outlays are given by the product of an average benefit,
BENEF, which is fully indexed (with a one-period lag) on the price index for
the urban formal sector, PF:
(27)
BENEF BENEF PF
=+
−−11
1(ln).
,
Δ
PENSIONS ptax U ptax S TRSOC
UP
d
SP
d
=++.
IMIG U Ew
Ew
Uim
IM
FOR
U
=⎛
⎝
⎜⎞
⎠
⎟
⎡
⎣
⎢
⎢
⎤
⎦
⎥
⎥+−
−,ln
11
λσ λλim
U
U
U
UIMIG
()
−
−−
,
,
,
1
2
1
Ew ER W
P
FOR
FOR
F
=⋅−
−
,
,
,
1
1

Breaking the Barriers to Higher Economic Growth226
The number of pensioners at the current period, NUMPEN, consists of last
period’s “stock” (adjusted for a fixed mortality rate), plus the flow of skilled
and unskilled workers retiring in each period, NEWPEN:
NUMPEN =(1 – δN)NUMPEN–1 + NEWPEN,
where δNis the proportion of pensioners who die in each period. The num-
ber of new pensioners is defined as
This equation indicates that at the beginning of each period, a fixed frac-
tion δU
NP (respectively δS
NP) of employed unskilled (respectively skilled) work-
ers retire from the formal sector labor force.
Thus, total pension outlays are given by:
PENSIONS = BENEF .NUMPEN. (28)
If we assume that the pension fund cannot borrow directly from the pub-
lic, and that its accounts must be balanced, government transfers are deter-
mined from (26), given (28):
(29)
Alternatively, if government transfers are considered fixed, the budget con-
straint can be used to determine the pension benefit, BENEF, after dropping
(27):
Other Model Features
Components of supply and demand are described by equations (A37) to
(A48). Both informal and public sector goods are nontraded. Total supply in
each sector is thus equal to gross production (equations (A39) and (A40).
Agricultural and private formal urban goods, by contrast, compete with
imported goods. The supply of the composite good for each of these sectors
consists of a combination of imports and domestically produced goods (equa-
tions (A38) and (A41).
For the agricultural and informal sectors, aggregate demand consists of
intermediate consumption and demand for final consumption (by both the
BENEF ptax U ptax S TRSOC
NUMPEN
UP
d
SP
d
=++
.
TRSOC PENSIONS ptax U ptax S
UP
d
SP
d
=−−.
NEWPEN U U S S
NP
U
P
d
GNP
S
P
d
G
T
=+
()
++
−− −−
δδ
,, ,,11 11
(()
.

Part II: Labor Markets and Human Capital 227
government and the private sector), whereas aggregate demand for the public
and private goods consists not only of intermediate consumption and final
consumption, but also of investment demand (Equations A42, A43, A44, and
A45). Total demand for intermediate consumption of any good is the sum of
intermediate consumption of this good over all production sectors (Equation
A37). Government expenditure on any good (except informal good) is equal to
a fixed share of total government expenditure (Equation A47). Household con-
sumption of each good is the summation across all categories of households’
consumption of this good (Equation A46). Consumption by individual house-
holds is derived from a Linear Expenditures System (LES).Total private invest-
ment by private urban firms consists of purchases of both public and urban
formal private goods and services (Equation A48).
Regarding external trade, private firms in the urban formal sector allocate
their output to exports or the domestic market according to a production pos-
sibility frontier/transformation function (Equation A9). Allocation of agricul-
tural output to domestic consumption and exports occurs according to a pro-
duction possibility frontier (Equation A3). Profit maximization requires firms
to equate relative prices to the opportunity cost in production (Equation A49).
Imports compete with domestic goods in the agricultural sector as well as in
the private formal sector (Equations A38 and A41). Cost minimization
requires the demand for imported vs. domestic agricultural and private urban
goods to be a function of relative domestic and import prices (Equation A50).
Prices are defined in Equations A51 to A62. The value added price of out-
put is given by the gross price net of indirect taxes, less the cost of intermedi-
ate inputs (Equation A51). The world prices of imported and exported goods
are taken to be exogenously given. The domestic currency price of these goods
is obtained by adjusting the world price by the exchange rate, with import
prices also adjusted by the tariff rate (Equations A52 and A53). Because the
transformation function between exports and domestic sales of agricultural
and urban private goods is linearly homogeneous, the domestic sales prices
are derived from the sum of export and domestic expenditure on agricultur-
al and private goods divided by the quantity produced of these goods
(Equation A54). For the informal and public sectors, the composite price is
equal to the domestic market price, which is in turn equal to the output price
(Equation A56). For the agricultural and private urban production sectors,
the substitution function between imports and domestic goods is also linear-
ly homogeneous, and the composite market price is determined accordingly
by the expenditure identity (Equation A55). The nested production function
of private formal urban goods is also linearly homogeneous; prices of the
composite inputs are derived in similar fashion (Equations A60 and A61). The
price of capital is constructed by using the investment expenditure identity,
which involves public and private-formal urban goods (Equation A62).
Finally, the consumption price indices for the rural, urban, and urban formal

Breaking the Barriers to Higher Economic Growth228
are given as weighted averages of composite good prices, with weights reflect-
ing consumption patterns (Equations A57, A58, and A59).
Profits and income are defined in (A63) to (A70). Firms’ profits in the infor-
mal and agricultural sectors are defined as revenue minus total labor costs
(Equation A63). Profits of private-urban sector firms account for salaries paid
to both skilled and unskilled workers (Equation A64). Firms’ income in the
agricultural and the informal sector is equal to their profits (Equation A65). But
firms’ income in the formal urban economy is equal to their profits minus cor-
porate taxes and interest payments on foreign loans (Equation A66). Household
income is based on the return to labor (salaries), distributed profits, and trans-
fers. Households are defined according to their sector of occupation. There are
four categories of households: rural, urban informal, urban formal, and capital-
ists. The rural household comprises all rural workers; the urban informal
household consists of workers in the urban informal sector; and the urban for-
mal household consists of urban formal sector employees (skilled and
unskilled). Finally, there is a capitalist-rentier household, whose income comes
from firms’ earnings in the formal private sector. Households in the rural and
informal urban economy own the firms in which they are employed—an
assumption that captures the fact that firms in these sectors tend to be small,
family-owned enterprises. Income of rural households is equal to the sum of
value added from production and transfers from the government (Equation
A67). Income of urban informal households also includes a fraction of foreign
remittances from (unskilled) workers employed abroad (Equation A68).
Income of urban formal households depends on government transfers and
salaries, foreign remittances, and pension payments (Equation A69). Firms pro-
vide no source of income, because these groups do not own the production
units in which they are employed. Firms in the private urban sector retain a por-
tion of their after-tax earnings for investment financing purposes and transfer
the remainder to capitalists. Thus, the capitalist-rentier household’s income is
the sum of transfer payments and distributed profits (Equation A70).
Consumption, savings, and investment are described in Equations A71 to
A74. Each category of household saves a constant fraction of its disposable
income, which is equal to total income minus income tax payments (Equation
A71). The portion of disposable income that is not saved is allocated to con-
sumption (Equation A72). The accumulation of capital over time depends on
the flow level of investment and the depreciation rate of capital from the pre-
vious period (Equation A74). The aggregate identity between savings and
investment implies that total investment must be equal to total savings, which
is itself equal to firms’ after-tax retained earnings, total after-tax household
savings, government savings, and foreign borrowing by firms and the govern-
ment (Equation A73). In the simulations reported later,this equation is solved
residually for the level of private investment, which implies therefore that the
model is “savings driven.”

Part II: Labor Markets and Human Capital 229
The government side is described in Equations A75 to A80. Government
expenditures consist of government consumption, which only has demand-
side effects, and public investment, which has both demand-and supply-
side effects. Public investment consists of investment in infrastructure,
education, and health. We define investment in infrastructure as the expen-
diture affecting the accumulation of public infrastructure capital, which
includes (as noted earlier) public assets such as roads, power plants. and
railroads. Investment in education affects the stock of public education
capital, which consists of assets such as school buildings and other infra-
structure determining skills acquisition, but does not represent human
capital. In a similar fashion, investment in health adds to the stock of pub-
lic assets such as hospitals, health clinics, and other government infrastruc-
ture affecting health. All value added in the production of public goods is
distributed as wages. Thus, the current fiscal deficit is equal to tax revenue
minus current household transfers, pension transfers, current expenditure
on goods and services, wage expenditure, and interest payments on foreign
public loans (Equation A75). Net government saving is equal to minus the
overall government budget deficit and is obtained by adding public invest-
ment expenditure to the current fiscal deficit (Equation A76). Total tax rev-
enues consist of revenue generated by import tariffs, sales taxes, income
taxes (on both households and firms in the urban private sector) and pay-
roll taxes net of employment subsidies (Equation A77). Government
investment is the sum of investment in infrastructure, investment in health,
and investment in education, which are all considered exogenous policy
variables (Equation A78). Government investment increases the stock of
public capital in either infrastructure, education or health. Accumulation
of each type of capital is equal to the sum of the capital stock from the pre-
vious period and currentperiod investment minus depreciation of the cap-
ital stock from the previous period (Equation A79). Because we assume
that only the private urban good is used for capital accumulation, we
deflate nominal investment by the demand price for private goods to obtain
real investment. Infrastructure and health capital affect the production
process in the private sector, as they both combine to produce the stock of
government capital (Equation A80).
Finally, the balance of payments is defined in Equations A86 to A88. The
external constraint implies that any current account surplus (or deficit) must
be compensated by a net flow of foreign capital, given by the sum of changes
in foreign loans made to the government and to private firms (Equation A86).
The flow of remittances is equal to the foreign wage, measured in foreign-
currency terms, times the stock of domestic workers abroad (Equation A87).
In turn, the stock of domestic workers abroad is the sum of new immigrants
and the stock of domestic workers abroad from the previous period, minus
attrition (Equation A88).

Breaking the Barriers to Higher Economic Growth230
Calibration and Parameter Values
This section presents a brief overview of the characteristics of the data under-
lying the model’s social accounting matrix (SAM) and discusses the parame-
ter values.
The Social Accounting Matrix
The basic dataset consists of a SAM and a set of initial levels and lagged vari-
ables. The SAM encompasses 37 accounts, including; production and retail
sectors (8 accounts); educational services (2 accounts); labor production fac-
tors and profits (5 accounts); enterprises (1 account), households (4
accounts), government current expenditures, taxes, and pensions (12
accounts); government investment expenditures (3 accounts), private invest-
ment expenditures (1 account), and the rest of the world (1 account).
The SAM dataset was derived to (i) reflect characteristics of a typical labor-
exporting MENA country, and (ii) represent an equilibrium dataset, includ-
ing a balanced government budget, a zero balance of payments (BoP), and
zero net foreign borrowing. Nevertheless, the net factor service account of the
BoP includes private and foreign interest payments reflecting past foreign
borrowing. The characteristics of the SAM data can be summarized as follows.
On the output side, agriculture and the informal sector account for respec-
tively 12 and 35 percent of total output at producer prices. In contrast, private
urban formal production accounts for almost 46 percent of total output. On
the demand side, private current expenditures account for 64 percent of GDP,
whereas government current expenditures and wages account for 11 percent
of GDP. At the same time, total investment expenditures represent 22 percent
of GDP, implying that our “prototype”labor-exporting MENA country is run-
ning a trade surplus equivalent to 2.8 percent of GDP.
Looking at the BoP, total net remittances to households amount to 1.2 per-
cent of GDP. Together with the trade balance surplus, this is financing a deficit
on the net factor services account amounting to 4 percent of GDP. This
includes private foreign interest payments (-1.5 percent of GDP) and govern-
ment foreign interest payments (-2.5 percent of GDP). The trade balance is
dominated by nonagricultural imports and exports, with agricultural exports
accounting for only 8 percent of total export earnings, and nonagricultural
imports accounting for 92 percent of total import expenditures. The level of
trade openness, measured by the ratio of the sum of imports and exports to
GDP, amounts to a moderate 38 percent.
Looking at the government budget, indirect taxes in the form of produc-
tion and retail level taxes (excluding payroll taxes) account for 84 percent of
total government revenues. Household tax revenues, amounting to 13 percent
of total government income, represent the largest revenue item among direct
tax items, whereas enterprise taxes account for only 4 percent of revenues.

Part II: Labor Markets and Human Capital 231
On the expenditure side, domestic transfers and foreign interest payments
account for respectively 24 and 9 percent of the budget, whereas consumption
and savings for investment purposes amount to respectively 40 and 27 percent
of the budget. Overall, the government relies heavily on indirect taxes for rev-
enue collection (as is common in developing countries), while at the same
time allocating a significant fraction of its resources to investment.
Behavioral Parameters
Consider now the behavioral parameters of the model. In agriculture the
Cobb-Douglas (share) parameter in the production of value added equals 0.8,
whereas the CES substitution elasticity between rural labor and public capital
equals 2/3. Public capital is aggregated from infrastructure and health capital
using a substitution elasticity of 1/2. The informal sector only has one factor
of production, unskilled labor, and the Cobb-Douglas (share) parameter also
equals 0.8. In contrast, the private formal sector has a three-level nested pro-
duction structure, with a bottom-level substitution elasticity of 2/3 between
private capital and skilled labor, a middle-level substitution elasticity of 7/6
between the bottom level composite factor and unskilled labor, and a top level
substitution elasticity of 5/6 between the middle-level composite factor and
public capital. Finally, public sector value added is derived using CES aggrega-
tion of skilled and unskilled labor, with a substitution elasticity of unity.
Turning to the factor market and the wage bargaining equation for private
sector skilled wages, elasticities with respect to skilled labor unemployment
and public sector wages are set respectively at -2.0 and 2.0, whereas the wage
elasticity of private skilled labor demand is -1.0. The parameter measuring the
trade union’s bargaining power is set at 0.7. Rural-urban and international
migration elasticities with respect to relative expected wages are set respective-
ly at 0.4 and 0.6, whereas persistence parameters for rural-urban and interna-
tional migration are set respectively at 0.1 and 0.3. The formal-informal sec-
tor migration elasticity with respect to relative expected wages is 0.4. the elas-
ticity of substitution between publicly employed teachers and public educa-
tion capital in the skills-upgrading CES production function is 1/3, whereas
the elasticity of teachers effort with respect to relative wages is 0.8.
The Armington elasticities for rural agricultural and urban formal sector
goods imports are set respectively at 2/3 and 1.5. Similarly, the CET transfor-
mation elasticities for rural agricultural and urban formal sector goods
exports are set respectively at 2/3 and 1.5. Finally, household minimum con-
sumption levels amount to 10 percent of initial consumption levels.
Simulation Experiments
In what follows, we use the framework described above to analyze five types
of labor market reforms: a reduction in payroll taxes on unskilled labor,

Breaking the Barriers to Higher Economic Growth232
assuming both neutral and non-neutral changes on the budget; reductions in
public sector employment and wages; an active labor market policy, consist-
ing of higher employment subsidies to unskilled workers in the formal private
sector; and a reduction in trade unions’ bargaining strength.15 We also consid-
er a composite reform package, which consists of a cut in payroll taxes and
public sector employment, as well as a reduction in unions’bargaining power.
In all of these experiments we use a savings-driven closure rule, and solve
residually for private investment demand, using the aggregate savings-
investment balance (equation (A73). This allows us to study the “crowding in”
and “crowding out” effects of labor market policies through their impact on
the government budget balance.
Reduction in Payroll Taxes on Unskilled Labor
The effects of a permanent, 5 percent reduction in the payroll tax rate on
unskilled labor are illustrated in figure 9.3. In analyzing the impact of this pol-
icy measure on growth and employment, a key aspect involves evaluating its
fiscal implications. For instance, assuming that the policy change must be
neutral with respect to the budget deficit, what are the alternative options for
offsetting the effect of a reduction in payroll taxation? To illustrate this type
of interaction between labor market reforms and fiscal policy, we examine
three alternative “closure” rules on the fiscal side. In the first, there is no off-
setting change in revenue, and the government borrows domestically to bal-
ance its budget—implying therefore full crowding out of private investment,
as implied by the aggregate savings–investment balance (equation (A73). In
the second, the policy is budget-neutral, and the government raises sales taxes
on the private, formal sector good to offset the increase in additional expen-
ditures on transfers to the pension system. In the third, the policy is also
budget-neutral, and the government offsets the reduction in payroll taxes by
an increase in income taxes on capitalists and rentiers.16 In all three cases, we
assume that the accounts of the pension system are balanced through govern-
ment transfers, as indicated in (29). Of course, even if the policy is budget-
neutral in the sense described here, private capital formation may still be
crowded out, as a result of general equilibrium effects leading to, say, smaller
private savings relative to the baseline.
A number of results are common to all three payroll tax experiments.
Reduced labor costs lead to increased employment of unskilled labor, as well as
substitution away from skilled labor (and physical capital) in the private formal
sector. A decrease in skilled labor wages—resulting from the rise in skilled
unemployment, itself due to the reduction in the demand for that category of
labor—partly offsets the initial tax-induced 5 percent increase in the differential
between skilled and unskilled labor costs. The net increase in the labor cost dif-
ferential amounts to around 3.8 percent in the longer term, suggesting a likely
trend away from skilled labor toward unskilled labor even in the long run.

Part II: Labor Markets and Human Capital 233
Figure 9.3 Simuation Results: Five Percentage Points Reduction in Unskilled Labor Payroll Tax Rate
Percentage deviations from baseline, unless otherwise indicated
1
2
3
4
5
1
⫺2.5
⫺2.0
⫺1.5
⫺1.0
⫺0.5
⫺3.0
⫺1.5
⫺1.0
⫺0.5
0
0.5
⫺2.0
⫺1.8
⫺1.6
⫺1.4
⫺1.2
⫺1.0
⫺0.8
⫺1.9
12345678910
12345678910
12345678910
123456789101234 5678910
1234 5678910
1234 5678910
1234 5678910
⫺0.1
0
0.1
0.2
0.3
0.4
0.5
⫺0.4
⫺
0.35
⫺0.3
⫺
0.25
⫺0.2
⫺
0.15
⫺0.1
⫺
0.05
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
Nominal wages in the informal sector
Unskilled unemployment rate
(urban formal sector)a
Value added in urban formal sectorbValue added in urban informal sector
b
Real private investment
Nominal skilled wages
in the private formal sector
Skilled employment
in the private formal sector
Private formal unskilled employment
(% of Informal sector employment)
a
non-revenue neutral revenue-neutral sales tax revenue -neutral income tax
Note: a. Absolute deviations from baselines; b. Real terms.

Breaking the Barriers to Higher Economic Growth234
The increase in unskilled employment is drawn partly from the pool of
unemployed workers in the formal sector and partly from increasing migra-
tion from the informal sector. Increasing migration to the formal sector
occurs against the backdrop of strongly increasing informal sector wages. The
minimum wage increases over time as a result of full indexation on urban
consumption prices, and higher demand in the formal sector for unskilled
labor leads to a rise in the expected formal sector wage. Over time, the ensu-
ing increase in the expected formal-informal sector wage differential leads to
higher migration flows to the formal economy. This pushes the informal sec-
tor wage upwards. Employment in the informal sector also increases in the
short run but tapers off in the medium to long term. The initial combination
of reduced overseas and formal sector migration flows—a year after the shock
and beyond—increases informal sector labor supply by 0.3-0.4 percent.
However, although reduced migration overseas, due to increased domestic
urban wages, continues to add workers to the informal sector labor force, the
subsequent reversal of formal sector migration, coupled with migration
outflows to the rural sector, leads to a gradual reversal of the initial increase in
labor supply. The reduced supply of labor is matched by a reduction in
demand due to the high informal sector wages. The mirror image of reduced
informal sector employment is a gradual increase in rural employment.
Accordingly, the cumulative effect of medium-term outward migration to
rural areas, due to increasing agricultural wages, leads to a relatively strong
expansion of labor supply in the rural sector.
At the aggregate level, the reduction in payroll taxes leads to a strong initial
increase followed by a gradual decline in nominal GDP. In contrast, real GDP
increases over time, indicating that the declining growth path for nominal
GDP is a purely nominal phenomenon. The adjustment process involves a real
exchange rate depreciation, which gradually raises exports and lowers imports
in the long term. The increasing trade surplus is used to finance the net factor
service income deficit, which rises due to declining migration overseas.
Whereas the growth paths of GDP and trade aggregates are relatively similar
across payroll tax experiments, the growth paths of private consumption and
investment are relatively sensitive—as could be expected—to whether or not
there are offsetting changes in taxes. The growth paths of private consumption
and investment are diverging strongly in the experiment, with a non-neutral
budget closure. The ensuing budget deficit is financed through domestic bor-
rowing, leading to crowding out of private investment and higher disposable
household income, which translates into increased private consumption.
The strong crowding-out effect disappears when sales or income taxes are
raised to pay for increased transfers to the pension system. However, differ-
ences persist. Sales taxes on formal sector goods raise the price of investment
goods and intermediate inputs, which depresses (everything else equal) prof-
its, savings, and private capital accumulation. In contrast, an increase in the

Part II: Labor Markets and Human Capital 235
tax rate on capitalists’ income reduces household disposable income and pri-
vate consumption, but allows for increased investment in the long term. Thus,
the scenario where reduced payroll tax revenues and increased transfers to the
pension system are offset by increased household income taxes, results in the
highest long-term GDP growth rates.17 Growth in the formal sector resulting
from increased private capital accumulation is responsible for higher long-
term aggregate growth in the latter scenario.
The impact on the government budget is of course related to the budget
closure rule. Government transfers to the pension system will, in each case,
increase by around 0.2 percent of GDP in the long run, due to the decline in
own-financing, itself resulting from the reduction in payroll taxation.
Without the introduction of alternative financing sources, this leads to a
domestic borrowing requirement of 0.3 percent of GDP in the long term. In
contrast, domestic borrowing is completely avoided in the long term if
budget-neutral specifications with variable sales and/or income taxes are
applied. Looking at the pension system, total pension payments increase the
most when higher government transfers are financed by increased sales taxes
on formal sector goods. The number of workers who qualify to enter the pen-
sion system rises (due to the expansion of formal sector employment) by
around 0.5 percent in the long run, regardless of the budget closure. However,
whereas the average benefit rate decreases by 0.4 percent— when increased
government transfers are financed by higher income taxes or not financed by
tax increases—it only falls by 0.1 percent when higher formal sector sales taxes
lead to increases in prices of formal sector goods.
Cut in Public Sector Wages
We now examine the effects of wage reductions for both unskilled public
workers and skilled public employees (excluding teachers). Results of a per-
manent 5 percent reduction in the wage rate for each labor category are sum-
marized in figures 9.4 and 9.5.
The reduction in public sector wages leads, in both cases, to a reduction in
the public sector borrowing requirement and to crowding-in of private invest-
ment. A reduction in unskilled wages in the public sector, by contrast, has lit-
tle impact on growth in the long term. The main channel through which pub-
lic sector wage reductions are transmitted is a reduction in aggregate demand,
induced by lower government consumption expenditures, and an increase in
total domestic savings (due to a reduction in the budget deficit only, because
private savings fall concomitantly with income), and thus private investment.
With employment in the public sector fixed, the expansion in investment
demand leads to increased private sector employment of unskilled labor and
a reduction in the unemployment rate. In turn, the reduction in unemploy-
ment (which raises the probability of finding a job in the private formal econ-
omy) leads to strong migration into the formal sector. The increased employ-

Breaking the Barriers to Higher Economic Growth236
Figure 9.4 Simuation Results: Five Percentage Points Reduction in Public Sector Unskilled Labor Wage
Percentage deviations from baseline, unless otherwise indicated
Note: a. Absolute deviations from baselines; b. Real terms.
⫺0.3
⫺0.2
⫺0.1
0
0.1
0.2
⫺0.4
⫺0.2
⫺0.1
0
0.1
0.2
0.3
⫺0.3 0.5
0.55
0.6
0.65
0.7
0.75
0
0.05
0.1
0.15
0.2
⫺0.08
⫺0.06
⫺0.04
⫺0.02
0
0.02
0.005
0.01
0.015
0.02
0.025
0.03
0.035
⫺0.02
0
0.02
0.04
0.06
0.08
0.02
0.04
0.06
0.08
0.1
0.12
0.14
Nominal wages in the informal sector
Unskilled unemployment rate
(urban formal sector)a
Value added in urban formal sectorbValue added in urban informal sectorb
Real private investment
Nominal skilled wages
in the private formal sector
Skilled employment
in the private formal sector
Private formal unskilled employment
(% of Informal sector employment)a
1234567891012345678910
1234567891012345678910
1234567891012345678910
1234567891012345678910

Part II: Labor Markets and Human Capital 237
Figure 9.5 Simuation Results: Five Percentage Points Reduction in Public Sector Skilled Labor Wage
Percentage deviations from baseline, unless otherwise indicated
Note: a. Absolute deviations from baselines; b. Real terms.
0
0.1
0.2
0.3
0.4
0.5
0.6
⫺0.2
⫺0.1
0
0.1
⫺0.3 1.3
1.35
1.4
1.45
1.5
⫺0.95
⫺0.9
⫺0.85
⫺0.8
⫺1
⫺0.05
0
0.05
0.1
0.15
0.75
0.8
0.85
0.9
0.95
1
⫺0.08
⫺0.06
⫺0.04
⫺0.02
0
0.02
0.04
0.2
0.25
0.3
0.35
0.4
0.45
0.5
0.55
Nominal wages in the informal sector
Unskilled unemployment rate
(urban formal sector)a
Value added in urban formal sectorbValue added in urban informal sectorb
Real private investment
Nominal skilled wages
in the private formal sector
Skilled employment
in the private formal sector
Private formal unskilled employment
(% of Informal sector employment)a
1234567891012345678910
1234567891012345678910
1234567891012345678910
1234567891012345678910

Breaking the Barriers to Higher Economic Growth238
ment and formal sector migration levels are, however, immediately reversed.
The combination of declining demand for formal sector goods and increasing
accumulation of capital lowers the demand for unskilled labor over time.
Moreover, increasing informal sector wages and formal sector unemployment
leads to outward migration from the formal sector. In the long term, employ-
ment levels are unchanged in every sector, and the higher level of production
(in the formal sector) is entirely driven by increased private capital accumula-
tion. Migration flows reverse themselves and unemployment rates remain
unchanged at baserun levels. Looking at the pension system, overall pensions,
including transfers from the government, remain unchanged. Accordingly,
both the number of pensioners and the level of pension benefits are
unchanged in the long term.
In contrast, a cut in skilled wages in the public sector has a relatively strong
long-term impact on growth because of the leadership effect on private skilled
wages. Reduced public sector wages spills over into lower private skilled wages
and higher employment of that category of labor in the private sector.
Combined with the cumulative effect of increased private investment, this
leads to higher growth rates in the formal sector. At the same time, growth in
the rural and urban informal sectors is driven by declining migration into the
urban (formal) sector. The outward migration is driven by substitution of
skilled for unskilled labor in private formal sector production, which leads to
a declining formal-informal sector wage differential. In addition, an increase
in the expected urban wage, due to increasing informal sector wages and a
gradually declining unskilled unemployment rate, leads to a decline in inter-
national migration flows. The combination of declining international and for-
mal sector migration leads to increasing labor supply in the urban informal
sector, and, because of increasing rural wages, to further out-migration and
higher labor supply in the rural sector.
Interestingly, a cut in public sector skilled wages leads to a long-term
reduction in both skilled and unskilled formal sector unemployment rates.
The reduction in unemployment among skilled workers results from
increased formal sector employment. In contrast, the reduction in open
unemployment among unskilled workers follows mainly from outward
migration to the informal sector.
Looking at the current account, reduced international migration (resulting
from the increase in the expected urban wage) leads to an increase in the net
factor service account deficit. This is counterbalanced by an improvement in
the trade balance, which comes about as a result of a small depreciation of the
real exchange rate.
The pension system sees some minor changes in the long term, including a
small increase in the number of retirees and a small decrease in the average
pension benefit rate. The number of pensioners increases due to increased

Part II: Labor Markets and Human Capital 239
formal sector employment levels, while pension benefits decline because the
strong supply response in the formal sector reduces formal sector prices. The
overall effect is to leave overall pension payments, including government
transfers to the pension system, unchanged in the long run. The main impact
on the government budget is therefore to reduce the budget deficit through
reduced consumption.
Public Sector Layoffs
The experiments of this section include public sector layoffs of both unskilled
workers and skilled public employees (excluding teachers). Results of a per-
manent 5 percent reduction in the number of workers in each labor category
are summarized in figures 9.6 and 9.7.
Reducing the size of public sector employment has the twin effects of
increasing private capital accumulation and raising levels of employment in
the private sector. Crowding-in of private investment is achieved because non-
budget neutral layoffs turn into a smaller domestic government borrowing
requirement. The increased private capital accumulation has a positive
supply-side effect on formal sector output. Nevertheless, the net increase in
relative demand for formal sector goods (resulting from higher private invest-
ment and lower private consumption), means that private formal sector
employment levels also increases for both categories of workers.
Aggregate growth effects are absent in the case of unskilled labor layoffs.
Real GDP declines in the short term, and returns to baserun levels in the longer
run. Nevertheless, there is positive growth over time in every production sec-
tor other than public services, including the rural and urban informal sectors.
Accordingly, while the aggregate growth impact is neutral in the long term, due
to lower value added in the public sector, the growth path has a distinct upward
trend toward the end of the 10-year horizon. Positive formal sector growth is
particularly evident throughout the simulation period. Short-term growth is
due to increased employment of unskilled labor in the private sector, whereas
long-term growth is mainly driven by higher private capital accumulation. In
fact, employment of unskilled labor in the private sector falls over time and
returns close to baserun levels in the long term. At the same time, growth in the
rural agricultural and urban informal sectors is driven primarily by increased
labor supplies, due to the cumulative impact of outward migration from the
urban (formal) sector. Outward migration is particularly strong because
declining employment prospects lower the expected urban wage.
The unskilled unemployment rate increases sharply in the short term,
because the rise in private sector employment is insufficient to absorb laid-off
workers in the public sector. In the longer run, unskilled unemployment
returns toward its baserun level, due to the cumulative effect of migration
outflows into the urban informal and rural sectors. Accordingly, formal sector

Breaking the Barriers to Higher Economic Growth240
Figure 9.6 Simuation Results: Five Percentage Points Reduction in Public Sector Unskilled Labor Employment
Percentage deviations from baseline, unless otherwise indicated
Note: a. Absolute deviations from baselines; b. Real terms.
⫺0.6
⫺0.5
⫺0.4
⫺0.3
⫺0.2
⫺0.1
0
0.1
0.2
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
1.38
1.39
1.40
1.41
1.42
1.43
1.44
1.45
1.46
1.47
0.3
0.31
0.32
0.33
0.34
0.35
0.36
0
0.05
0.1
0.15
0.04
0.05
0.06
0.07
0.08
0.09
⫺0.05
0
0.05
0.1
0.15
0.1
0.15
0.2
0.25
0.3
0.35
0.4
Nominal wages in the informal sector
Unskilled unemployment rate
(urban formal sector)a
Value added in urban formal sectorbValue added in urban informal sectorb
Real private investment
Nominal skilled wages
in the private formal sector
Skilled employment
in the private formal sector
Private formal unskilled employment
(% of Informal sector employment)a
1234567891012345678910
1234567891012345678910
1234567891012345678910
123456 7891012345678910

Part II: Labor Markets and Human Capital 241
Figure 9.7 Simuation Results: Five Percentage Points Reduction in Public Sector Skilled Labor Employment
Percentage deviations from baseline, unless otherwise indicated
Note: a. Absolute deviations from baselines; b. Real terms.
0.3
0.4
0.5
0.6
0.7
0.8
0.25
⫺0.5
⫺0.4
⫺0.3
⫺0.2
⫺0.1
0
0.1
0.2
0.3
0.4
1.8
1.9
2.0
2.1
2.2
2.3
⫺2.2
⫺2.1
⫺2.0
⫺1.9
⫺1.8
⫺1.7
⫺1.6
⫺1.5
⫺1.4
⫺1.3
⫺0.05
0
0.05
0.1
0.15
0.2
0.25
1.3
1.4
1.5
1.6
1.7
⫺0.09
⫺0.08
⫺0.07
⫺0.06
⫺0.05
⫺0.04
⫺0.03
0.3
0.4
0.5
0.6
0.7
0.8
Nominal wages in the informal sector
Unskilled unemployment rate
(urban formal sector)
a
Value added in urban formal sectorbValue added in urban informal sectorb
Real private investment
Nominal skilled wages
in the private formal sector
Skilled employment
in the private formal sector
Private formal unskilled employment
(% of Informal sector employment)a
12345678910 123456 78910
12345678910 12345678910
12345678910 123456 78 910
12345678910 12345678910

Breaking the Barriers to Higher Economic Growth242
employment of unskilled labor is reduced to baserun levels, whereas the sup-
ply of unskilled labor in the formal sector declines by more than 1 percent in
the long term. International migration flows increase in the short term but
move in the opposite direction in the longer term, due to higher domestic
wages and lower unemployment levels.
In contrast to layoffs of unskilled government workers, layoffs of skilled
public employees have markedly positive growth effects. Real GDP increases
in both the short and the long term. Again, output growth occurs in all non-
public sectors, but in contrast to the previous experiment involving unskilled
labor layoffs, more balanced and stronger growth is recorded among the rural
and urban informal and formal sectors. Strong formal sector growth is driven
by increased employment of (skilled) labor in the short term. Thus, the short-
term impact of skilled labor layoffs follows the pattern of unskilled layoffs.
However, the long-term growth effect is driven by a combination of increased
employment levels and private capital accumulation. Thus,the initial increase
in the level of (skilled) labor employment persists in the long term, in contrast
with the unskilled labor layoffs.
Growth in the rural and urban informal sectors is again driven by declin-
ing urban migration flows, resulting from layoffs of unskilled labor in the pri-
vate sector. Sharply declining skilled labor wages (resulting from reduced
union wage demands, due to increased unemployment among skilled work-
ers) leads to increased substitution of skilled for unskilled labor in the private
formal sector. The accompanying narrowing in wage differentials leads to
declining urban (formal) migration and increasing rural and urban informal
labor supplies. This leads to a decline in the open unskilled unemployment
rate, and an accompanying increase in the expected urban formal sector wage,
which partly reverses the strong initial reduction in migration from the for-
mal sector. But high and increasing informal sector wages mean that formal
sector migration remains lower in the long term.
In line with the previous experiment involving layoff of unskilled workers
in the public sector, layoff of skilled workers leads to higher unemployment
for that category of labor while reducing unemployment among unskilled
workers. Moreover, the reduction in unskilled unemployment is magnified in
the longer run by the fall in the supply of unskilled labor in the formal sector
that is induced by migration flows. This long-term reduction is, however,
markedly smaller than in the case of unskilled layoffs. The increase in infor-
mal and formal urban wages, combined with the declining open unemploy-
ment rate, means that international migration flows decline in the long run as
well. The cumulative effect on the domestic workforce abroad again leads to
an increasing deficit in the net factor services account over time. This is made
up for by a moderate improvement in the trade balance, which comes about
through a small real exchange rate depreciation.

Part II: Labor Markets and Human Capital 243
The main impact of public sector layoffs on the government budget is to
reduce current consumption and domestic government borrowing require-
ments. In contrast, transfers to the pension system remain relatively
unchanged: Overall, pension payments decline by around 0.2 percent in the
long run. In the case of unskilled layoffs, the decline in pension payments is
due solely to a decline in the number of retirees. Layoff of unskilled workers
results in the largest reduction in employment, and therefore results in the
largest reduction in pensioners. In contrast, layoff of skilled workers leads to
lower pension payments, partly because of declining numbers of retirees, and
partly because of a declining average benefit rate. A stronger supply effect of
skilled labor layoffs leads to declining formal sector prices and, accordingly, to
an increasing pension benefit rate.
Subsidies to Private Employment
We now turn to an analysis of the impact of subsidies to employment of
unskilled labor in the formal private sector under various government budg-
et closures, including: (i) a non-neutral budget closure; (ii) a budget-neutral
increase in sales taxes on private formal sector goods; and (iii) a budget-
neutral increase in income taxes on capitalists. In each case, the increase in
employment subsidy amounts to 5 percent of the base year private formal
unskilled wage level. The simulation results are summarized in figure 9.8.
The employment subsidy for unskilled labor increases the differential
between skilled and unskilled labor costs, in a manner similar to the 5 percent
reduction in payroll taxes considered earlier. The only difference between the
two sets of results comes from the fact that the payroll tax rate applies to the
unskilled wage rate, while the employment subsidy does not. The unskilled
wage rate is fixed in real terms, but variations in consumer prices in the for-
mal sector lead to some variation in its nominal value. However, these wage
changes have minimal impact on the results, which are therefore almost iden-
tical to the results of the payroll tax experiments.
In particular, the reduction in unskilled labor costs leads to a strong
increase in formal sector employment and a decline in open unskilled unem-
ployment. The increase in unskilled employment is the main engine for for-
mal sector growth. Initial migration flows also fuel increases in labor supply
in the informal sector. However, subsequent outward migration from urban
to rural areas leads to an increase in rural labor supply, and the initial increase
in informal sector labor supply therefore tapers off. In the long term, aggre-
gate growth is driven equally by rural and urban formal sector growth when a
non-neutral government budget closure is used. An increase in the public sec-
tor borrowing requirement leads to crowding out of private investment,
reduced capital accumulation, and depressed formal sector growth. In con-
trast, urban formal sector growth is markedly stronger than rural agricultur-

Breaking the Barriers to Higher Economic Growth244
Figure 9.8 Simuation Results: Five Percentage Points Increase in Unskilled Labor Employment Subsidy
Percentage deviations from baseline, unless otherwise indicated
Note: a. Absolute deviations from baselines; b. Real terms.
1
2
3
4
5
⫺3.0
⫺2.5
⫺2.0
⫺1.5
⫺1.0
⫺0.5
-2.0
⫺1.5
⫺1.0
⫺0.5
0
0.5
⫺1.5
⫺1.0
⫺1.9
⫺0.1
0
0.1
0.2
0.3
0.4
0.5
⫺0.4
⫺0.35
⫺0.3
⫺0.25
⫺0.2
⫺0.15
⫺0.1
⫺0.05
0.3
0.4
0.5
0.6
0.7
0.8
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1234 5678910 1 23 4 56 7 8 910
1234 5678910 1 23 4 56 7 8 910
1234 5678910 1 23 4 56 7 8 910
1234 5678910 1 23 4 56 7 8 910
Nominal wages in the informal sector
Unskilled unemployment rate
(urban formal sector)a
Value added in urban formal sectorbValue added in urban informal sectorb
Real private investment
Nominal skilled wages
in the private formal sector
Skilled employment
in the private formal sector
Private formal unskilled employment
(% of Informal sector employment)
a
non-revenue neutral revenue neutral sales tax revenue neutral income tax

Part II: Labor Markets and Human Capital 245
al growth when employment subsidies are financed by higher income tax rev-
enues. The domestic government borrowing requirement is eliminated and
crowding out of investment is reversed, leading to slightly increased capital
accumulation in the long term. Unemployment among skilled workers
increases slightly, regardless of the government budget closure. In contrast,
strongly increasing labor demand in the private sector reduces the unskilled
unemployment rate, in spite of a sharp long-term increase in labor supply.
The deficit of the net factor services account increases, due to the cumulative
effect of reduced international migration, but a compensating improvement
in the trade balance means that the current account balance is left essentially
unchanged. Finally, pensions payments increase marginally, due to a relative-
ly strong increase in the number of retirees (itself resulting from a rise in for-
mal sector employment) and a smaller decline in the pension benefit rate (due
to declining formal sector prices).18
Reduction in Unions’ Bargaining Strength
We now turn our attention to a reduction in labor union bargaining strength.
To do so we reduce the bargaining strength parameter, the coefficient νin
Equation 17, from its initial level of 0.7 to a value of 0.6. The results are shown
in figure 9.9.
The direct impact of the reduction in union bargaining power is to lower
skilled wages and increase employment for skilled labor in the private sector.
Increased production leads to lower formal sector goods prices and increased
real demand for all goods, including agricultural and informal sector goods.
The expansion of real private investment demand is particularly large.
Increasing real demand, combined with a lower wage bill, raises firms’ profits
in the private formal sector. In turn, this boosts private savings and invest-
ment. The combination of higher skilled employment, capital accumulation
in the private sector, and increased unskilled employment in urban informal
and rural sectors leads to a rise in output, which persists in the long term. This
growth scenario is therefore broad-based, in the sense that it stems from both
rural and urban formal sector growth, and to a lesser extent from urban infor-
mal sector growth.
The sharp drop in skilled wages leads to strong substitution away from
unskilled labor in the urban formal sector. Although the large initial decline
in private formal unskilled employment is partly reversed over time, it
remains below its baseline value in the long term, as a result of the permanent
nature of the reduction in the wage differential. Growth in the formal sector
is thus driven by increased employment of skilled labor and private capital
accumulation. The strong initial substitution away from unskilled labor also
leads to high unemployment in that category and a drop in expected urban
wages. This leads to a marked initial decline in formal sector migration flows,

Breaking the Barriers to Higher Economic Growth246
Figure 9.9 Simuation Results: Reduction in Labor Union’s Bargaining Strength
Percentage deviations from baseline, unless otherwise indicated
Note: a. Absolute deviations from baselines; b. Real terms.
0
0.5
1.0
1.5
2.0
2.5
3.0
⫺1.5
⫺1.0
⫺0.5
0
0.5
1.0
1.5
2.0
0.5
1.0
1.5
2.0
⫺5.5
⫺5.0
⫺4.5
⫺4.0
⫺3.5
⫺0.1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
3
3.1
3.2
3.3
3.4
3.5
⫺0.5
⫺0.4
⫺0.3
⫺0.2
⫺0.1
0.4
0.5
0.6
0.7
0.8
0.9
1
0.35
Nominal wages in the informal sector
Unskilled unemployment rate
(urban formal sector)
a
Value added in urban formal sectorbValue added in urban informal sectorb
Real private investment
Nominal skilled wages
in the private formal sector
Skilled employment
in the private formal sector
Private formal unskilled employment
(% of Informal sector employment)a
12345678910 123456 78910
12345678910 12345678910
12345678910 123456 78910
12345678910 12345678910

Part II: Labor Markets and Human Capital 247
and an equally large initial increase in labor supply in the informal sector.
However, the strong migration response also reduces unemployment to below
baserun levels. This increases the expected urban wage and, subsequently,
leads to a reversal in the direction of formal sector migration flows. In the
longer run, migration flows tend to increase labor supply in the urban infor-
mal sector at a constant rate, and labor supply in the rural sector at an increas-
ing rate. This leads to relatively fast long-term growth in the rural sector, and
slower, but robust, growth in the urban informal sector.
After an initial adjustment period, formal sector unemployment rates drop
well below baserun levels and stay there in the long term. The reduction in
skilled unemployment follows from increased employment of skilled labor in
the private sector. In contrast, the long-term reduction in unskilled unem-
ployment follows from a reduction in the formal sector supply of unskilled
labor, itself resulting from reduced migration flows.
The international wage differential drops in the long term, leading to a decline
in overseas migration flows. This contrasts with the long-term decline in migra-
tion between rural and urban areas, where a high degree of persistence implies
weak migration flows, in spite of a long-term increase in the wage differential.
The decline in international migration leads to an increase in the net fac-
tor services income deficit. This reduction is counterbalanced in the long term
by an improvement in the trade balance. The government budget is character-
ized by long-term declines in overall income- and expenditure-to-GDP ratios,
and by a long-term decline in the domestic borrowing requirement, which
allows for crowding-in of private investment. Government transfers to the
pension system do not change significantly. Looking at the overall pension
system, pension payments increase only slightly in the long run. An increase
in the number of retirees is evened out by a drop in the pension benefit rate.
The increase in the number of retirees follows from increased private sector
(skilled) employment, whereas the reduction in the pension benefit rate fol-
lows from declining formal sector prices associated with the strong formal
sector supply response.
A Composite Reform Program
Finally, we consider a composite, “realistic” package of policies, which com-
bines reductions in payroll taxes on unskilled labor with a reduction in
unions’ bargaining strength, and a cut in unskilled employment in the public
sector. We assume that the payroll tax on unskilled labor is reduced by 5 per-
cent, the number of unskilled workers in the public sector is reduced by 5 per-
cent, and that the bargaining strength of trade unions is reduced from an ini-
tial level of 0.7 to a “neutral” value of 0.6. We consider the three alternative
budget closure rules specified above, but to save space we do not report the
results graphically.

Breaking the Barriers to Higher Economic Growth248
Given that the transmission channels of each of the individual components
of the composite program have been described extensively in previous sec-
tions, we restrict our attention here to the impact and long-run effects of the
program on growth and unemployment. Simulation results with a non-neutral
public deficit closure do not show evidence of large nonlinear effects.
Specifically, the impact and long-term effects on the growth rate of aggregate
real value added amount to 0.2 and 0.9 percent, respectively. This is approxi-
mately equal to the sum of the growth rates derived from the individual simu-
lations described earlier. Similarly, the impact and long-term effects on private
formal sector employment are very close to linear, amounting to (minus
sign)0.1 and 4.0 percent respectively for unskilled labor, and 3.2 and 3.1 per-
cent, respectively, for skilled labor. Unskilled employment increases slightly by
0.2 percent when income taxes are used to keep the public deficit unchanged.
Overall, although the simultaneous implementation of the individual compo-
nents of the policy reform package above does bring some benefits (suggesting
therefore that complementarity between labor market policies is desirable
from an economic point of view, independently of other, political economy
considerations) the impact of a “realistic” package of labor market reforms on
growth and employment does not appear to be large. This has important
implications for the design of adjustment programs in a region where the chal-
lenge is not only to reduce an existing high level of unemployment, but also to
create sufficient jobs to absorb new entrants in the labor force.
Summary and Policy Lessons
The purpose of this paper has been to analyze the impact of labor market
reforms on growth, real and relative wages, and the composition of employ-
ment and unemployment in labor-exporting MENA countries. We first pro-
vided a brief overview of the main features of the labor market in some of
these countries. We then presented a simulation model, based on the IMMPA
framework developed by Agénor (2003), Agénor, Izquierdo and Fofack
(2003), and Agénor, Fernandes, and Haddad (2003), which captures many of
these features (such as a large informal urban sector, a significant role of pub-
lic sector employment and “leadership effects” of public sector wages, power-
ful trade unions, and international migration of labor) as well as other impor-
tant structural characteristics of these countries (such as a pay-as-you-go pen-
sion system). After discussing the calibration procedure and our choices of
parameter values, we presented and discussed a series of simulation experi-
ments. We focused on a reduction in payroll taxation on unskilled labor,
reductions in public sector wages and workforce, an increase in employment
subsidies to the private sector, changes in the bargaining strength of trade
unions, and a composite reform package involving several of these policies. In

Part II: Labor Markets and Human Capital 249
the case of payroll taxation, employment subsidies and the composite pack-
age, we considered both neutral and non-neutral changes. Specifically, we
considered three alternative fiscal “closure” rules. In the first, we assumed no
offsetting change in revenue, and the government borrows domestically to
balance its budget—implying full crowding out of private investment. In the
second and third, the policy is budget-neutral, and the government raises,
respectively, sales taxes on the private, formal sector good, and income taxes
on capitalist-rentier households, to cover increased expenditures (resulting
from an increased deficit in the pension system and increased labor subsidy
payments). Therefore, there is no crowding out of private capital formation in
the second and third experiments.
There are a number of policy lessons that emerge from our results. For
instance, we found that, regardless of how a cut in payroll taxes on unskilled
labor is financed (either by borrowing from the private sector or by imple-
menting revenue-neutral changes in sales or income taxation) reduced labor
costs lead to increased employment of unskilled labor, as well as substitution
away from skilled labor (and physical capital) in the private formal sector. A
decrease in skilled labor wages—resulting from a rise in skilled
unemployment—partly offsets the tax-induced increase in the differential
between skilled and unskilled labor costs. The net increase in the labor cost
differential is still very significant in the longer term, implying a strong sub-
stitution away from skilled labor toward unskilled labor, even in the long run.
At the same time, our results showed that the overall rate of output growth
varies significantly across experiments—essentially because the behavior of
private investment depends very much (as could be expected) on whether or
not there are offsetting changes in taxes. When the increase in the budget
deficit resulting from a cut in the payroll tax is financed through domestic
borrowing, private investment is crowded out; the lower rate of capital accu-
mulation has an adverse effect on growth, and thus on the demand for both
categories of labor. Thus, the indirect or “level” effect on the demand for
unskilled labor may mitigate significantly the substitution effect triggered by
the change in relative labor costs. The results associated with a reduction in
trade unions’ bargaining power in wage negotiations also indicate large long-
term gains in overall employment associated with general equilibrium effects.
The main policy lessons of our simulation experiments can be summarized
as follows. The first is that, in assessing the impact of labor market reforms on
growth and unemployment, it is important to account not only for the direct
(partial equilibrium) effects of these policies, but also for their indirect (gen-
eral equilibrium) effects on income and aggregate demand (resulting from
changes in the government budget). For instance, it has been argued that a
policy of subsidizing employment in the private sector would help to reduce
unemployment in MENA countries. A simple, partial equilibrium analysis of

Breaking the Barriers to Higher Economic Growth250
this policy is indeed unambiguous: by lowering the relative cost of unskilled
labor, a subsidy leads to an increase in the demand for that category of labor,
which may be particularly significant if wages are fixed (as a result, for exam-
ple, of a binding minimum wage). As long as the increase in labor demand
does not prompt greater participation in the labor force (that is, if unskilled
labor supply is fairly inelastic), unskilled unemployment is thus likely to fall.
However, a partial equilibrium view can be misleading. The increase in
subsidies must be financed, and this can occur in a variety of ways. Our exper-
iments indicate that general equilibrium effects can be significant, even with a
fairly neutral investment specification. Accordingly, if spending is kept con-
stant, and the government chooses to let its fiscal deficit increase and borrow
from the rest of the economy, this can have a large crowding-out effect on pri-
vate investment, if private savings do not adjust quickly; the fall in investment
may, over time, restrain the expansion of demand for all categories of labor—
including unskilled labor.Thus, the longer-run effect of the policy on employ-
ment may be either nil or negative. Similarly, an increase in, say, taxes on
capitalist-rentier households may restrain private capital formation and have
an adverse effect on employment in the medium and the long run. More
specifically, our general equilibrium analysis indicates that the overall impact
of a change in payroll taxes on the demand for unskilled labor may be com-
pounded or mitigated, depending on how the government chooses to adjust
its tax and spending instruments to maintain a balanced budget. In the pres-
ence of large crowding-out effects on private investment (and possibly sav-
ings), the direct gains (in terms of higher employment) associated with a
reduction in payroll taxes may be highly mitigated. Similarly, whether
employment subsidies end up reducing open unemployment in the formal
sector may depend on the extent to which a higher perceived probability of
finding a job in that sector affects workers’ decisions to remain in the infor-
mal sector.
The second policy lesson that our simulation results indicate is that a
“piecemeal” approach to labor market reforms is unlikely to bring substantial
benefits in terms of growth and employment. By contrast, a comprehensive
approach will bring broad-based growth and reductions in both skilled and
unskilled unemployment. The idea that labor market reform programs must
be sufficiently broad (in the sense of covering a wide range of complementa-
ry policies) and deep (of substantial magnitude) to have much of an effect is
emphasized by Coe and Snower (1997) and Orszag and Snower (1998). At the
same time, however, although a “realistic” package of policies (which com-
bines reductions in payroll taxes on unskilled labor, a reduction in unions’
bargaining strength, and a cut in unskilled employment in the public sector)
may have a significant impact on the composition of employment in labor-
exporting MENA countries, fostering a sustained increase in growth rates and

Part II: Labor Markets and Human Capital 251
job creation in these countries may require a more comprehensive program of
structural reforms—involving, in particular, financial sector reforms, privati-
zation, and measures aimed at increasing private sector participation.
Notes
1. Youth unemployment ranges from 37 percent of total unemployment in Morocco to 73 percent in
Syria, with a simple average of 53 percent for all countries for which data are available. Except in Jordan
and Lebanon, first-time job seekers make up more than 50 percent of the unemployed (ILO 2003).
2. See for instance Pissarides (1993) and Pritchett (1999). In countries where wage levels fell in real
terms during the 1990s, educated and experienced workers were affected the most,leading to a degra-
dation of skills. Despite this, recent estimates suggest that returns on education are generally higher in
the public sector than in the private sector at nearly all education levels but the university level (Assaad
2002 and World Bank 2004).
3. Our focus on labor-exporting MENA countries only is due to the fact that the characteristics of the
labor market in these countries (as well as other structural economic features) differ significantly from
those of the labor-importing countries of the region.
4. Yemen is also an important labor-exporting country in MENA, but due to the lack of reliable infor-
mation on the labor market in that country we chose to exclude it from our review.
5. In principle, a rigorous analysis of pension systems and pension reform would require the use of an
intergenerational framework, such as the OLG-CGE model discussed by Farmer and Wendner (1999),
or more recently by Beetsma, Bettendorf, and Broer (2003); Cavalletti and Lubke (2002); and Fehr,
Hans, and Erling Ateigum (2002). Our analysis should be viewed as an approximation only.
6. See Shaban et al. (2001) and World Bank (2004). A more detailed discussion of the features of the
labor market in MENA countries is provided by Hollister and Goldstein (1994) and Salehi-Esfahani
(2001). See also Agénor (1996, 2004) for a more general discussion of the features of labor markets in
developing countries.
7. Published measures of unemployment mostly include unemployed workers looking for jobs in the
formal sector, but not underemployed workers in the informal and rural sectors, that is, disguised
unemployment. For the difficulty of measuring unemployment in MENA countries, see for instance
Rama (1998) for Tunisia.
8. Said (2001) found no close correlation between minimum wages and unskilled wages in Morocco;
and during the 1980s, real wages for these workers fell faster than the real legal minimum wage.
However, other evidence, reviewed by Agénor and El Aynaoui (2003), suggests the minimum wage is
binding in urban areas.
9. Labor market segmentation refers to a situation where observationally identical workers (that is,
workers with similar qualifications) receive different wages depending on their sector of employ-
ment. It is a pervasive feature of labor-exporting MENA countries. Segmentation may be induced by
various factors: government intervention in the form of minimum wages; trade unions, which may
prevent wages from being equalized across sectors by imposing a premium for their members; and
efficiency wages, resulting from nutritional factors, turnover costs or productivity considerations (see
Agénor (1996, 2004). The first two sources of segmentation are incorporated in the model presented
in the next section.
10. The model is based on the Integrated Macroeconomic Model for Poverty Analysis (IMMPA)
framework developed by Agénor (2003),Agénor, Izquierdo and Fofack (2003), and Agénor, Fernandes,
and Haddad (2003), modified to account for international labor migration (as in Agénor and El
Aynaoui (2003) and a social security system.
11. Note that the specification of θU assumes complete job turnover outside the government sector
every period.

Breaking the Barriers to Higher Economic Growth252
12. Note that we do not explain endogenously the allocation of unskilled workers’ time between pro-
duction and learning-an important trade-off from the individual’s point of view. Allocating more time
to learning reduces the workers’ current labor income, but enhances their human capital, thereby
increasing their earnings in the future. To the extent that public capital in education enters as an input
to the human capital production function, as in (23), it would also affect private decisions to accumu-
late human capital. See, for instance, Glomm and Ravikumar (1998) for a formal model of the labor-
learning choice, which emphasizes, however, flow spending on education.
13. See Glystos (2002, 2003) for a discussion of the macroeconomic effects of foreign remittances in
several MENA countries, including Egypt, Jordan, Morocco, and Tunisia.
14. Note that, in our model, only employers pay payroll taxes that serve to finance the pension system.
In practice, workers also make significant contributions. However, we abstract from these contribu-
tions, given that the focus of our simulation experiments is on changes in payroll taxes paid by firms.
15. Results of all these experiments are summarized in Figures 9.3 to 9.9.A set of Excel sheets, contain-
ing more detailed results, are available upon request.
16. Of course, other offsetting changes, such as increases in direct tax rates on enterprises, or an
increase in indirect taxes on urban goods, could also be considered.
17. Note that the model does not capture the disincentive effects of higher direct tax rates on labor
supply (or participation rates). To the extent that these effects are large, the impact of higher out-
put growth rates on unemployment would be ambiguous, because a reduction in labor supply
(which would tend to lower unemployment) could be offset by large substitution effects toward
physical capital.
18. A limitation of our analysis is that we do not account for the fact that employment subsidies may
have unintended consequences, such as subsidized workers replacing unsubsidized ones, or employers
firing subsidized workers once the subsidy period ends.
References
Abrahart, Alan, Iqbal Kaur, and Zafiris Tzannatos. 2002. “Government Employment and Active Labor
Market Policies in MENA in a Comparative International Context.” In Employment Creation and
Social Protection in the Middle East and North Africa, ed. by Heba Handoussa and Zafiris Tzannatos.
Cairo: American University in Cairo Press.
Agénor, Pierre-Richard. 1996. “The Labor Market and Economic Adjustment.” IMF Staff Papers 43
(June): 261-335.
———. 2002. “Fiscal Adjustment and Labor Market Dynamics.” Unpublished. World Bank,
Washington, DC (October). Forthcoming, Journal of Development Economics.
———. 2003. “Mini-IMMPA: A Framework for Assessing the Unemployment and Poverty Effects of
Fiscal and Labor Market Reforms.” Policy Research Working Paper No. 3067. World Bank,
Washington, DC (May).
———. 2004. The Economics of Adjustment and Growth. 2nd ed. Boston, MA: Harvard University
Press.
Agénor, Pierre-Richard, Alejandro Izquierdo, and Hippolyte Fofack. 2003. “IMMPA: A Quantitative
Macroeconomic Framework for the Analysis of Poverty Reduction Strategies.” Policy Research
Working Paper No. 3092. World Bank, Washington, DC (June).
Agénor, Pierre-Richard, and Karim El Aynaoui. 2003. “Labor Market Policies and Unemployment in
Morocco: A Quantitative Analysis.” Policy Research Working Paper No. 3091. World Bank,
Washington, DC (June).
Agénor, Pierre-Richard, Reynaldo Fernandes, and Eduardo Haddad. 2003. “Analyzing the Impact of
Adjustment Policies on the Poor: An IMMPA Framework for Brazil.” Unpublished. World Bank,
Washington, DC (June).

Part II: Labor Markets and Human Capital 253
Assaad, Ragui. 1997.“The E_ects of Public Sector Hiring and Compensation Policies on the Egyptian
Labor Market.” World Bank Economic Review 11 (March): 85-118.
———. 2002. “The Transformation of the Egyptian Labor Market: 1988-98.” In The Egyptian Labor
Market in an Era of Reform, ed. by Ragui Assaad. Cairo: American University in Cairo Press.
Assaad, Ragui, and Simon Commander 1994.“Egypt.”In Labour Markets in an Era of Adjustment,ed.
by Susan Horton, Ravi Kanbur, and Dipak Mazumdar. Washington DC: World Bank.
Beetsma, Roel, Leon Bettendorf, and Peter Broer. 2003. “The Budgeting and Economic Consequences
of Ageing in the Netherlands.” Economic Modelling 20 (September): 987-1013.
Booth, Alison L. 1995.The Economics of the Trade Union. Cambridge, UK: Cambridge University Press.
Cavalletti, Barbara, and Eckhard Lubke. 2002. “Ageing Population and Pension Reform in Italy.” In
Policy Evaluation with Computable General Equilibrium, ed. by Amedeo Fossati and Wolfgang
Wiegard. London: Routledge.
Coe, David T., and Dennis J. Snower. 1997. “Policy Complementarities: The Case for Fundamental
Labor Market Reform.” IMF Staff Papers 44 (March): 1-35.
Farmer, K., and R. Wendner. 1999. “The Use of Multi Sector OLG-CGE Models for Policy Analysis.”
Research Memorandum No. 99-03. Department of Economics, University of Graz (March).
Fehr, Hans, and Erling Ateigum. 2002. “Pension Funding Reforms in a Small Open Welfare State.” In
Policy Evaluation with Computable General Equilibrium, ed. by Amedeo Fossati and Wolfgang
Wiegard. London: Routledge.
Glomm, Gerhard, and B.Ravikumar. 1998.“Flat-Rate Taxes, Government Spending on Education, and
Growth.” Review of Economic Dynamics 1 (January): 306-25.
Glystos, Nicholas P. 2002. “The Role of Migrant Remittances in Development: Evidence from
Mediterranean Countries.” International Migration 40 (March): 5-26.
———. 2003. “A Macroeconometric Model of the Effects of Migrant Remittances in Mediterranean
Countries.” In Human Capital: Population Economics in the Middle East, ed. by Ismail Siraguldin.
London: I. B. Tauris Publishers.
Harris, John, and Michael P. Todaro. 1970. “Migration, Unemployment and Development: A Two-
Sector Analysis.” American Economic Review 60 (March): 126-43.
Hollister, Robinson G., and Markus P. Goldstein. 1994. Reforming Labor Markets in the Near East:
Implications for Structural Adjustment and Market Economies. San Francisco, CA: ICS Press.
International Labor Organization. 2003. Global Employment Trends. Geneva: International Labor
Office.
Layard, Richard, Stephen Nickell, and Richard Jackman. 1991. Unemployment: Macroeconomic
Performance and the Labor Market. Oxford: Oxford University Press.
Norback, Pehr-Johan. 2001. “Cumulative Effects of Labor Market Distortions in a Developing
Country.” Journal of Development Economics 65 (June): 135-52.
Orszag, Mike, and Dennis Snower. 1998. “Anatomy of Policy Complementarities.” Swedish Economic
Policy Review 5 (Autumn): 303-51.
Pissarides, Christopher A. 1993. “Labor Markets in the Middle East and North Africa.” Discussion
Paper No. 5, MENA Region. World Bank, Washington, DC (March).
Pritchett, Lant. 1999.“Has Education Had a Growth Payo_ in the MENA Region?” Discussion Paper
No. 18, MENA Region. World Bank, Washington, DC. (June).
Rama, Martin. 1998.“How Bad Is Unemployment in Tunisia? Assessing Labor Market Efficiency in a
Developing Country.” World Bank Research Observer (February): 59-77.
Said, Mona. 2001. “Public Sector Employment and Labor Markets in Arab Countries: Recent
Developments and Policy Implications.” In Labor and Human Capital in the Middle East,ed.by
Djavad Salehi-Ishfahani. Reading, UK: Ithaca Press.

Breaking the Barriers to Higher Economic Growth254
Salehi-Ishfahani, Djavad. 2001. Labor and Human Capital in the Middle East.Reading, UK: Ithaca Press.
Shaban, Radwan A., Ragui Assaad, and Sulayman Al-Qudsi. 2001. “Employment Experience in the
Middle East and North Africa.” In Labor and Human Capital in the Middle East, ed. by Djavad
Salehi-Esfahani. Reading, UK: Ithaca Press.
Strand, Jon. 2003. “The Decline or Expansion of Unions: A Bargaining Model with Heterogeneous
Labor.” European Journal of Political Economy 19 (June): 317-40.
United Nations. 2003. Arab Human Development Report. Geneva: International Labor Office, UN.
World Bank. 2004. “MENA’s Employment Challenge in the 21st Century: From Labor Force Growth
to Job Creation.” Unpublished. World Bank, Washington, DC (January).

Part II: Labor Markets and Human Capital 255
Appendix A. List of Equations*
Production
(A1)
(A2)
(A3)
(A4)
(A5)
(A6)
(A7)
(A8)
(A9)
Employment
(A10)
(A11)
XVX a
jjj ij
i
=+∑
VU K
AXAXAA XAG
XA XA
XA
=+−
{}
⎡
⎣
⎢
⎢
⎢
⎤
⎦
−−
−
αβ β
ρρ
ρ
()1
1
⎥⎥
⎥
⎥
−1ηXA
XE D
ATATAA TAA
TA TA TA
=+−
{}
αβ β
ρρ
ρ
()1
1
VU
IXII
XI
=α β
VU S
GXGXGG XGG
XG XG XG
=+−
()
{}
−−
−
αβ β
ρρ
ρ
1
1
VT K
PXPXP XPG
XP XP XP
=+−
()
{}
−−
−
αβ β
ρρ
ρ
1
1
1
TT U
XP XP XP P
XP XP XP
1112 1
1
11
1
1=+−
()
{}
−−
−
αβ β
ρρ
ρ
TS K
XP XP P XP P
XP XP XP
222 2
1
22
2
1=+−
()
{}
−−
−
αβ β
ρρ
ρ
XE D
PTPTPP TPP
TP TP TP
=+−
()
{}
αβ β
ρρ
ρ
1
1
UU g MIG
RR R
=+
()
−
−,11
UV PV
W
A
d
AXA
A
A
XA
XA
XA
XA
XA
=−
()
⎛
⎝
⎜⎞
⎠
⎟
+−
111
ρ
η
ρ
ηβ
α
⎛⎛
⎝
⎜
⎜
⎞
⎠
⎟
⎟
+
1
1ρXA
*Unless otherwise indicated, the indexes iand j,with i, j = A, I, P, G refer to production sectors and h
= A, I, F, KR to households.

Breaking the Barriers to Higher Economic Growth256
(A12)
(A13)
(A14)
(A15)
(A16)
(A17)
(A18)
(A19)
(A20)
(A21)
(A22)
(A23)
(A24)
UUV
W
PV
R
s
A
d
A
A
A
=⎛
⎝
⎜⎞
⎠
⎟
,
UT PT
ptax W ES
P
d
UM U
XP
XP
XP
=+−
⎛
⎝
⎜⎞
⎠
⎟
1
11
1
11
()
β
αρ
σXXP1
U
U
U
UU
W
W
F
s
F
s
P
d
F
s
G
M
I
,
,
,,
,
,
−
−
−−
−
−
=−
⎛
⎝
⎜⎞
⎠
1
1
11
1
1
⎟⎟
⎧
⎨
⎪
⎩
⎪
⎫
⎬
⎪
⎭
⎪
βF
UU g MIGSKLIMIG
UU UNP
U
=+−
()
+−−
−,11δ
Ug U
GUGNP
U
G
=+ −
()
−
11
δ,
UUU
I
s
UF
s
=−
ST PT
ptax W ES
P
d
S
SS S
XP
XP
XP
=+−
⎛
⎝
⎜⎞
⎠
2
22
2
12
κβ
αρ
() ⎟⎟
σXP 2
UNEMP SS
S
S
G
T
P
d
=− +
1()
UNEMP UU
U
U
GP
d
F
s
=− +
1()
.
WwP
MMF
=
WPV V
U
IXI
II
I
s
=⎛
⎝
⎜⎞
⎠
⎟
β
WwP
UG UG F
=
W
WUNEMPW
WS
PROF
S
SSSG
SW
SP
d
P
P
d
S
−−− =
−
Ω012 0
φφεν
/

Part II: Labor Markets and Human Capital 257
(A25)
(A26)
(A27)
(A28)
(A29)
(A30)
(A31)
(A32)
(A33)
(A34)
(A35)
(A36)
WwP
SG SG F
=
MIG U Ew
Ew
U
Rm
M
U
A
m
=⎛
⎝
⎜⎞
⎠
⎟
⎡
⎣
⎢
⎢
⎤
⎦
⎥
⎥+−
()
−,ln
11
λσλ
RR
R
UMIG
,
,
−
−−
1
2
1
Ew WW
P
U
UM U I
URB
=+−
−−
−
θθ
,,
,
()
11
1
1
θU
P
F
s
G
U
UU
=−
−
−−
,
,,
1
11
Ew W
P
A
A
R
=−
−
,
,
1
1
IMIG U Ew
Ew
U
U
UimIM
F
U
m
U
=⎛
⎝
⎜⎞
⎠
⎟+−
()
−
−
,ln
,
1
1
1
λσ λ
UU
IMIG
,−−
2
1
Ew ER W
P
FOR
FOR
URB
=⋅−
−
,
,
1
1
SKL S K
EG
E
EE
EEE
=
()
+−
()
⎡
⎣
⎢⎤
⎦
⎥
−−−
βϕ β
ρρρ
1
1
ϕϕ
δ
=− −
()
⎡
⎣
⎢
⎢
⎤
⎦
⎥
⎥
−−
−
1111
1
m
SS
SG
UNEMP W
W
E
,,
,
SSS
G
T
GG
E
=+
Sg S
G
T
SG NP
S
G
T
=+ −
()
−
11
δ,
SSSKL
SNP
S
=− −
()
+
−
11
δδ

Breaking the Barriers to Higher Economic Growth258
Supply and Demand
(A37)
(A38)
(A39)
(A40)
(A41)
(A42)
(A43)
(A44)
(A45)
(A46)
(A47)
(A48)
Trade
(A49)
(A50)
INT a X
jjii
i
=∑
QM D
A
s
QA QA A QA A
QA QA QA
=+−
()
{}
−−
−
αβ β
ρρ
ρ
1
1
QX
I
s
I
=
QX
G
s
G
=
QM D
P
s
QP QP P QP P
QP QP QP
=+−
()
{}
−−
−
αβ β
ρρ
ρ
1
1
QCGINT
A
d
AA A
=++
QCINT
I
d
II
=+
QCGZINT
G
d
GG P
G
G
=+++
QCG ZZ INT
P
d
PP P
P
GP
=++ +
()
+
CC xcc CON PQ x
PQ
iihih
hih h i iih
i
hh
==+ −
()
∑∑ ΣΣ
Ggg
NG
PQ gg j A G P
jj
j
j
===
∑
,,,,1for
Zzz
PK Z
PQ zz j G P
P
j
j
P
i
j
=⋅==
∑
,,,1for
ED
PE
PD iAP
ii
i
i
Ti
Ti
Ti
=⋅
−
⎛
⎝
⎜⎞
⎠
⎟=
1β
β
σ
,,for
MD
PD
PM iAP
ii
i
i
Qi
Qi
Ti
=⋅
−
⎛
⎝
⎜⎞
⎠
⎟=
β
β
σ
1,,for

Part II: Labor Markets and Human Capital 259
Prices
(A51)
(A52)
(A53)
(A54)
(A55)
(A56)
(A57)
(A58)
(A59)
(A60)
(A61)
(A62)
Income
(A63)
(A64)
(A65)
PV V PX indtax a PQ X
ii i i jij
j
i
=−−
⎧
⎨
⎪
⎩
⎪
⎫
⎬
⎪
⎭
⎪
−∑
11()
PE wpe ER i A P
ii
==,,for
PM wpm tm ER i A P
iii
=+ =(), ,1for
PX PD D PE E
XiAP
i
ii ii
i
+=,,for
PQ PD D PM M
QiAP
i
ii i i
i
=+=,,for
PQ PX PD i I G
iii
== =,,for
PwrPQ wr
Rii i
ii
==
∑∑ ,with 1
PwfPQ wf
Fii i
ii
==
∑∑ ,with 1
PwuPQ wu
URB i i i
ii
=−
∑∑ ,with 1
PT TPT ptax W ES U
T
UM U P
1
22
1
1
=++ −
⎡
⎣⎤
⎦
()
PT PROF ptax W ES S
T
PSSSP
2
2
1
=++ −
⎡
⎣⎤
⎦
()
PK
PQ Z
Z
PQ Z PQ Z
Z
ii
iGP
G
PP
P
==
+
∑
PROF PV V W U i A I
iiiii
=− =,,for
PROF PV V ptax W ES U
ptax
PPP UMUP
S
=−+ −
⎡
⎣⎤
⎦
−+
()
()
1
1WWESS
SSP
−
⎡
⎣⎤
⎦
YF PROF i A I
ii
==,,for

Breaking the Barriers to Higher Economic Growth260
(A66)
(A67)
(A68)
(A69)
(A70)
Consumption, Savings, and Investment
(A71)
(A72)
(A73)
(A74)
Government
(A75)
(A76)
(A77)
(A78)
(A79)
(A80)
YF entax PROF IF ER FL
PPP
=−
()
−⋅ ⋅ −
11,
YH TRH PV V
AA AA
=+γ
YH TRH PV V ER REMIT
II II F
=++−⋅γτ()1
YH PENSIONS TRH W U W U
WS W S
FFMPUGG
SP SGG
=+++
()
++
(
γ
))
+⋅τFER REMIT
YH TRH re YF
KR KR P
=+−
()
γ1
SAV sr inctax YH
hh h h
=−
()
1
CON inctax YH SAV
hhhh
=−
()
−1
PK Z PQ Z re YF SAV CDEF ER FL FL
PPG P h P G
h
⋅+ =⋅ + − + +
()
ΔΔ
∑∑
KKZ
PPPP
=−
()
+
−−
111
δ,,
−= −− −
−− ⋅−
CDEF TXREV TRH TRSOC W S
NG IF ER FL
SG G
E
GG,11
−=−−ODEF CDEF PQ Z
PG
TXREV wpm tm M ER ptax W U ptax W S
iii UMP
d
SSP
d
iA
=++
=,PP
iii P RR
i
indtax PX X entax PROF inctax YH
i
∑
∑
++⋅+
+nnctax YH inctax YH ES U ES S
AA FF UP
d
SP
d
−−
PQ Z I I I
PG INF E H
=++
KK
I
PQ iINFHE
iii
i
P
=−
()
+=
−−
−
11
1
1
δ,
,
,
,,,where
KK K
GGGINF GH
GG
G
=+−
()
{}
−−
−
αβ β
ρρ
ρ
1
1

Part II: Labor Markets and Human Capital 261
Pension System
(A81)
(A82)
(A83)
(A84)
(A85)
Balance of Payments
(A86)
(A87)
(A88)
TRSOC PENSIONS ptax U ptax S
UP
d
SP
d
=−−
BENEF BENEF PF
=+
()
−−11
1Δln ,
NUMPEN NUMPEN NEWPEN
N
=−
()
+
−
11
δ
NEWPEN U U S S
NP
U
P
d
GNP
S
P
d
G
T
=+
()
++
−− −−
δδ
,, ,,11 11
(()
PENSIONS BENEF NUMPEN=⋅
01
=−
()
+−⋅
−
−
=
∑wpe E wpm M REMIT IF FL
IF
ii i i P
iAP
G
,
,
FFL FL FL
GGP,−++
1ΔΔ
REMIT W FORL
FOR
=−1
FORL FORL IMIG
IMIG
=−
()
+
−
11
δ

Breaking the Barriers to Higher Economic Growth262
Appendix B. Variable Names and Definitions*
Endogenous Variables
BENEF Average pension benefit
CiConsumption of good iby the urban and rural private sector
Cih Consumption of good iby household h
CONhTotal nominal consumption by household h
CDEF Current public budget deficit
DiDomestic demand for good i= A, P
EiExport of traded goods for i= A, P
ESSNominal employment subsidy on skilled labor in the private
sector
ESUNominal employment subsidy on unskilled labor
in the private sector
EwUExpected urban unskilled wage
EwAExpected agricultural wage
FORL Stock of domestic workers abroad
GiGovernment spending on good i= A, G, P
IMIG International migration
INTiIntermediate good demand for good i
KEPublic capital in education
KGTotal public capital
KHPublic capital in health
KINF Public capital in infrastructure
KPPrivate capital
MiImports of good i= A,P
MIG Migration to urban area
NEWPEN Flow of skilled and unskilled workers retiring in each period
NUMPEN Number of pensioners
ODEF Overall budget deficit
PFFormal urban price index
PRRural price index
PSPrice index for skilled labor
PURB Urban price index
PDiDomestic price of domestic sales of good i= A,P
PEiPrice of exported good i= A,P
PENSIONS Total amount of pension paid to pensioners
PK Price of capital
PMiPrice of imported good i= A,P
PQiComposite good price of good i
PROFiProfits by firms in sector i= A,I,P
*Unless otherwise indicated, the index i= A, I, P, G refers to production sectors and h= A, I, F, KR to
households.

Part II: Labor Markets and Human Capital 263
PT1Price of composite input T1
PT2Price of composite input T2
PViValue added price of good i
PXiSales price of good i
Qs
i,Qd
iComposite supply and demand of good i
REMIT Foreign currency value of the flow of remittances from abroad
S Skilled workers
Sd
PDemand for skilled workers in private urban formal sector
SAVhSaving by household h
srhSaving rate for household h
SKL New skilled workers
SPSkilled labor employed in private urban formal
T1Composite input from T2and unskilled labor
T2Composite input from capital and skilled labor
TRH Transfers to households
TRSOC Net government pension transfers
TXREV Tax revenues
UiUnskilled labor employed in sector i= A,I,P
URUnskilled workers in rural sector
Us
RUnskilled labor supply in the rural sector
UUUnskilled workers in urban sector
Ud
iDemand for labor in sector i= A,I,P
US
FUnskilled labor supply in the urban formal sector
US
IUnskilled labor supply in the informal sector
UNEMPSSkilled unemployment rate
UNEMPUUnskilled unemployment rate in the formal sector
ViValue added in sector i
WiNominal wage for labor employed in sector i= A, I
wiReal wage rate for unskilled labor employed in sector i= A, I
WMMinimum wage (unskilled labor in urban formal private sector)
wMReal minimum wage (unskilled labor in urban formal private
sector)
WSNominal wage rate for skilled worker in the private urban for-
mal sector
wSReal wage rate for skilled worker in the private urban formal
sector
WSG Nominal wage rate for skilled labor in the government sector
wSG Skilled wage in the public sector, real terms
WUG Nominal wage rate for unskilled labor in the government
sector
wUG Unskilled wage in the public sector, real terms

Breaking the Barriers to Higher Economic Growth264
xih Subsistence level of consumption of good iby household h
XiProduction of good i
YFiIncome by firms in sector i= A, I, P
YHhHousehold income for household h
ZTotal investment demand
ZiInvestment demand for good i= P, G
Zi
pInvestment demand for good i= P, G by formal private sector
Exogenous Variables
Name Definition in text
entax Corporate income tax
ER Nominal exchange rate
EwFOR Expected real foreign wage (in terms of domestic prices)
FLiForeign loans to sector i=G, P
GCGovernment consumption
gRPopulation growth in rural economy
gSG Growth rate of the skilled labor force in the public sector
gUG Growth rate of the unskilled labor force in the public sector
gUPopulation growth in urban economy
IEInvestment in education
IF Foreign interest rate
IFGInterest rate on government foreign loans
IHInvestment in health
IINF Investment in infrastructure
inctaxhIncome tax rate for h
indtaxiRate of indirect taxation of output in sector i
NG Total government current expenditure on goods and services
ptaxSPayroll tax for skilled labor in private urban sector
ptaxUPayroll tax for unskilled labor in private urban sector
SGSkilled workers in public sector
SE
GSkilled labor in the public sector engaged in education
ST
GTotal number of skilled workers in the public sector
tmiImport tariff for good i= A, P
UGUnskilled workers in public sector
WFOR Nominal foreign wage
wpeiWorld price of export for i= A, P
wpmiWorld price of import for i= A, P

Part II: Labor Markets and Human Capital 265
Parameters
Name Definition in text
aij Input-output coefficient
αGShift parameter in the public capital equation
αQi Shift parameter in the total supply function of good i= A, P
αTi Shift parameter in transformation function between exported
and domestic production of good i= A, P
αXi Shift parameter in production of good i= A, P
αXP1 Shift parameter in composite input of unskilled and
skilled/capital composite input
αXP2 Shift parameter in composite input of skilled workers and pri-
vate capital
βEParameter determining the weight of skilled labor in produc-
tion of education
βFSpeed of adjustment for the supply of unskilled labor in the
formal private sector
βGShare parameter in the public capital equation
βQA Shift parameter in agricultural composite good
βQP Shift parameter in urban composite good
βTi Shift parameter between exported and domestic production of
good i= A, P
βXi Shift parameter in production of good i= A, I, P
βXP1 Share parameter between unskilled and skilled/capital com-
posite input
βXP2 Share parameter between skilled workers and private capital
ccih Shares of household hin consumption of good i
δEDepreciation rate of education capital
δHDepreciation rate of health capital
δINF Depreciation rate of infrastructure
δIMIG Rate of “attrition” of the stock of migrants
δS
NP Rate of skilled retirement in the urban formal sector
δU
NP Rate of unskilled retirement in the urban formal sector
δESensitivity of skilled effort level to relative wages
δPDepreciation rate of private capital
δSRate of “depreciation” or “de-skilling” of the skilled labor
εSP
d/WSWage elasticity of the demand for skilled labor
ϕProductivity of public workers engaged in providing education
ϕmMinimum level of effort
ηXA Coefficient of returns to scale in the agricultural value added
function
γh Share of transfers allocated to household h

Breaking the Barriers to Higher Economic Growth266
ggiShare of government expenditure on good i=A, G, P
κSShift parameter for skilled private sector employment
λim Speed of adjustment in the international migration equation
λmPartial adjustment rate on migration
ϕjParameters determining the nominal wage rate for the skilled
labor for j=1, 2
re Percentage of profits retained
ρESubstitution parameter between skilled labor in production of
education and educational capital stock
ρGSubstitution parameter in the public capital equation
ρQi Substitution parameter in total supply of good i=A, P
ρTi Substitution parameter between exported and domestic pro-
duction of good i=A, P
ρXi Substitution parameter in production of good i=A, P
ρXP1 Substitution parameter between unskilled and skilled/capital
composite input
ρXP2 Substitution parameter between skilled workers and private
capital
σIM Elasticity of international migration flows with respect to
expected wages
σMElasticity of migration flows with respect to expected wages
σQA Elasticity of agricultural composite good
σQP Elasticity of private urban composite good
σSElasticity of saving rate to deposit rate
σTi Elasticity of transformation between exported and domestic
production of good i=A, P
σXP1 Elasticity of substitution between unskilled workers and
composite input of skilled workers and private capital
σXP2 Elasticity of substitution between skilled workers and private
capital
τFFraction of remittances that are allocated to households in the
formal sector
θUShare of urban unskilled workers employed in formal sector
υMeasure of the trade union’s bargaining power
wfiInitial share of good iin consumption of formal sector goods
wriRelative weight of good iin rural consumption
wuiInitial share of good iin urban unskilled workers’
consumption
zziShare of investment expenditure on good i= G, P

267
Economic Reforms and
People Mobility for a More
Effective EU-MED Partnership
Ishac Diwan*
Mustapha Nabli
Adama Coulibaly†
Sara Johansson de Silva†
10
Introduction
The new framework for partnership between the European Union (EU) and
the Southern Mediterranean countries was launched at the Barcelona
Conference in November 1995; the Extraordinary Conference held in
Barcelona in November 2005 marked its 10th anniversary. The purpose of the
initiative was to reinvigorate the partnership between the EU and the
Southern Mediterranean (MED) countries, and work toward integration and
convergence. The core mechanism for achieving these results has been the
gradual establishment of a free trade area for industrial goods.
Other examples of regional partnerships, both involving the EU (the
Central European countries that became members of the EU in 2004) and
outside (the United States-Canada-Mexico NAFTA agreement) are evidence
of the potential benefits from integration between richer and poorer areas.
For the EU-MED Partnership, however, there has been little progress so far
compared with what the agreements set out to achieve. Growth in the MED
*World Bank. †World Bank.
Published as “Rendre le Partenariat EU-MED plus effectif: Réformes et Mobilité de la Main d'œuvre,”
in L'Union Européenne et le Moyen-Orient: Etat des Lieux. Reprinted by permission from Bruylant,
Presses Universitaires de Saint-Joseph.

Breaking the Barriers to Higher Economic Growth268
countries is relatively modest, exports remain undiversified with limited non-
mineral exports growth, and foreign companies still hesitate to invest in the
region. At the same time,lack of growth and lack of job creation are giving rise
to a serious economic, political and social crisis.
This paper argues that to reverse these negative trends, the EU-MED agen-
da needs to be enriched in two significant directions. First, there should be a
clearer, much broadened agenda for supporting growth-enhancing structural
reforms that complement trade liberalization. The second and parallel track
focuses on the potential for including labor mobility between the EU and the
MED countries on the agenda. Managed migration flows, by construction of
limited magnitude, would explicitly address the critical political and social
implications of the employment situation in the MED countries.
The EU-MED Agreements: Background
The EU-MED Partnership initiative following the 1995 Barcelona Conference
led to new association agreements between the EU and each of the Arab
Southern Mediterranean countries. Eight have been already signed: Tunisia
(1995), Israel (1995), PLO for the benefit of the Palestinian Authority (1997),
Morocco (1996), Jordan (1997), Egypt (2001), Algeria (1992), and Lebanon
(2002). Negotiations are continuing with Syria (table 10.1).
The agreements have several intertwined purposes: to provide a framework
for political dialogue between the EU and the MED region; to establish the
conditions for gradual liberalization of trade in goods, services, and capital to
promote trade and the expansion of economic and social relations between
the EU and Mediterranean region; and to encourage regional integration
within the Southern Mediterranean zone. The commitment to trade liberal-
ization is intended to improve competitiveness by allowing for import com-
petition and more effective resource allocation. It is also expected to attract
foreign direct investment (FDI) into the MED region: firstly, since the main
regulatory obstacles to FDI are removed, and secondly, since companies locat-
ing in the MED region benefit from geographic proximity to the EU and the
lower labor costs in the MED region. The acceleration of trade and investment
should result in economic integration and bring further benefits, including a
much-needed revival of economic growth in the MED region.
Although the agreements’ objectives are broad based, the central mecha-
nism of the EU agreements is rather narrowly focused on establishing free
trade areas, covering industrial products between the EU and each partner
country within a 12-year period. Since manufacturing products from the
MED countries already have access to the EU market, the agreement is in
essence a means for gradually allowing EU manufacturing products into the
MED markets. Further liberalization of agricultural goods and services is
expected to proceed on a different time frame, following additional negotia-
tions. A parallel financial instrument (MEDA) is used to support and facilitate

Part II: Labor Markets and Human Capital 269
the adjustment process. This financial assistance is intended to complement
the free trade agenda by supporting economic reforms and compensating for
the socioeconomic costs of the transition to a market economy.1
The Achievements in Terms of Economic Integration and Convergence
Have Been Limited
Regional integration, in particular between a developed and an underdevel-
oped area, can bring several political and economic gains (World Bank 2000).
On the economic side, market enlargement could increase the incentives for
domestic and foreign private investment through scale effects, and at the same
time force firms to undertake efficiency improvements in response to
increased competition. On the political side, regional commitments can pro-
vide a mechanism for increasing the credibility of economic and political
reforms. It is also clear, however,that these benefits are far from automatic and
that whether they materialize, and to what extent, will depend on the charac-
teristics of both the partners and the agreement itself.
In this light, what has been achieved in the more than 10 years that have
passed since the Barcelona Conference? It may seem early to assess achieve-
ments, since only two of the eight agreements became effective before the year
2000 (Tunisia, West Bank and Gaza), meaning that in most countries, reforms
have just begun. However, we will argue that some assessment is already pos-
sible, for at least three reasons. First, a significant impact of the agreements
should occur from the announcement effect, even before the agreements are
fully implemented. Second, for some countries, some dismantling of the tar-
iffs started even before the agreements became effective (Tunisia, Morocco),
and financial assistance was also activated. Third, the experience of other
countries, as highlighted below, shows that the effects of similar agreements
Table 10.1. Progress on the EU-Mediterranean Association Agreements a
Conclusion of Signature of Effectiveness End of
negotiations agreement date transition periodb
Tunisia Jun 1995 Jul 1995 Mar 1998 2008
Israel Sep 1995 Nov 1995 Jun 2000 2012
Morocco Nov 1995 Feb 1996 Mar 2000 2012
PLO/PA Dec 1996 Feb 1997 Jul 1997c 2009
Jordan Apr 1997 Nov 1997 May 2002 2014
Egypt Jan 2001 Jun 2001 June 2004 2016
Algeria Dec 2001 Apr 2002 Sep 2005 2017
Lebanon Jan 2002 June 2002 Mar 2003c 2015
Syria In progress .. .. ..
a. As of October 2006. b. The date on which all tariffs on imported goods from the EU—as included in the
agreements—will have to be dismantled. c. Interim agreement.

Breaking the Barriers to Higher Economic Growth270
can be significant in a relatively short time, in fact similar to the time that has
passed since the beginning of the Barcelona process.
Unfortunately, the evidence so far points to limited success in achieving the
objectives envisaged by the MED agreements. In the discussion following, we
focus on the experience of four countries that have signed the agreements
with the EU: namely,Tunisia,Morocco,Jordan, and Egypt. As we will see, even
in countries where the positive effects should be most strongly felt, given the
agreements that are in place, there has been little progress toward economic
integration and convergence.
Contrary to their purpose, the agreements have not accelerated the process
of economic integration. Generally, the export performance of the four MED
countries has not improved since the mid-1990s: nonmineral exports as share
of GDP have remained more or less stagnant and at low levels by international
standards. Nor has there been a discernible change in trade flows: the four coun-
tries together provide less than 1 percent of total EU nonmineral imports, with
virtually no increase in market share over the last decade. Foreign direct invest-
ment has fluctuated significantly between the years, largely because they have
been concentrated on large and bulky infrastructure deals. With the exception
of Jordan, however, inflows have not increased significantly (table 10.2).
Table 10.2. Lack of Economic Integration and Growth in the MED Countries, 1990–2000
Egypt Jordan Morocco Tunisia
Nonfuel merchandise
exports, % of GDP
1990–94 4.4 24.3 15.4 24.0
1995–99 2.9 24.2 19.9 26.6
2000–04 3.7 28.1 20.3 29.3
Nonfuel merchandise
exports to EU (14),
% of total EU (14)
merchandise importsa
1990–94 0.13 0.08 0.30 0.24
1995–99 0.11 0.09 0.36 0.27
2000–04 0.13 0.11 0.33 0.28
FDI, % of GDP
1990–94 1.4 0.2 1.4 2.3
1995–99 1.1 2.3 0.1 1.9
2000–04 0.9 4.2 1.6 2.8
GDP per working
age population,
av. annual growth (%)
1990–95 0.7 ⫺1.2 ⫺1.8 1.0
1995–00 2.8 ⫺0.4 1.3 3.0
2000–04 0.8 2.3 2.4 2.3
Source: World Bank (2006).
a. EU (14) excludes countries that became members in 2004. In addition, Luxembourg is excluded for lack of data.

Part II: Labor Markets and Human Capital 271
The lack of economic integration is mirrored in the comparatively weak
growth rates in the MED countries. Generally, growth rates of GDP per work-
ing age population have been lower than in the EU. As a result, there has not
yet been any discernible economic convergence between the two regions.
Yet, the potential rapid gains from integration between a large developed
region and a less developed area are well illustrated elsewhere. Successful
experiences with regional partnerships include the NAFTA trade liberaliza-
tion agreement among the United States, Canada, and Mexico, which took
effect in 1994. Another example are the (considerably broader) preaccession
agreements between the EU and applicant countries in Central and Eastern
Europe (CEE), who became members in 2004.
The lack of economic integration between EU and the MED countries
stands in stark contrast to the cases of both Mexico-United States and the EU-
accession countries. In the following discussion, we present information for
Mexico and three of the CEE countries: namely the Czech Republic, Hungary,
and Poland. Since the early 1990s, both Mexico and the three CEE countries
have doubled their market share (that is, share of total imports) in the United
States and EU respectively (figure 6.1). This was not a result of pure trade
diversion, however: in Mexico, total nonfuel merchandise exports increased
from less than 10 percent of GDP in 1990 to 25 percent of GDP by 2004, and
over the same period, the three CEE countries almost doubled their share of
nonfuel exports in GDP to nearly 50 percent. Foreign direct investment saw
even more spectacular development (figure 6.2). Total FDI flows into Mexico
increased from 1 percent of GDP in the beginning of the 1990s to almost 3
percent of GDP by the end of the decade. In the group of CEE countries, for-
eign direct investment increased more significantly, from around 1 percent of
GDP to 5 percent of GDP on average, and reaching as high as 10 percent in
the Czech Republic.As seen, the MED4 countries saw no comparable increase
in either market share or investment.
A similar picture of the potential benefits from regional integration
emerges from the accession process of Spain and Portugal, which became
members of the European Union in 1985. Between 1980 and 1988, foreign
direct investment more than tripled in Spain and Portugal and exports
increased rapidly over the same period. As economic growth rates increased,
Portugal and Spain began catching up with the rest of Europe (figure 10.3).
Greece, which joined the EU in 1980, did not benefit as clearly from regional
integration, however. The differences in outcomes are linked to the depth of
the reforms which Spain and Portugal undertook in response to EU integra-
tion, but which Greece avoided.
In sum, it is clear that regional cooperation can deepen economic integra-
tion and lead to economic convergence, but this does not appear to be hap-
pening for the MED countries. The question is then how to explain the poor
performance of the EU-MED integration process.

Breaking the Barriers to Higher Economic Growth272
The Needed Agenda for Complementary Reforms Has Failed to Materialize
The lack of progress in economic integration and convergence, compared to
the underlying objectives and initial expectations, clearly begs the question as
to whether the EU-MED agenda in its current form is sufficient to deliver the
desired results. The far more superior performance—in terms of concrete
results and beneficial impact—of the NAFTA and EU preaccession agree-
ments in the 1990s, suggests that a comparative exercise could be useful to
evaluate the content and limitations of the EU-MED agenda as it stands. In
particular, the NAFTA and preaccession agreements appear to offer a more
Figure 10.1. Nonfuel Merchandise Exports, Percent of
Total EU (14) Merchandise Imports (MED4, CEE3) or U.S.
Merchandise Imports (Mexico)
Figure 10.2. FDI, Percent of GDP
1990–94
1995–99
2000–04
MED4
percent
CEE3
a
Mexico
0
2
6
10
14
4
8
12 1990–94
1995–99
2000–04
MED4
percent
CEE3 Mexico
0
0.5
1.5
4.0
5.0
1.0
2.0
3.0
2.5
3.5
4.5
Source: World Bank 2006.
a. No data available for 1990–94.
Source: World Bank 2006.
a. No data available for 1990–94.
Figure 10.3. GDP per Potential Worker as Percent of Italy’s: Spain, Portugal and Greece,
1980–2004
percent
40
45
50
55
60
65
70
75
80
1980 1985 1990 1995 2000
Spain
Portugal
Greece
Source: World Bank 2006.

Part II: Labor Markets and Human Capital 273
complete form of integration and harmonization, whereas that complemen-
tary agenda of wider institutional reform and convergence is missing in the
case of the EU-MED partnership.
The slow transition toward a well-functioning liberalized market has been
a key factor in containing growth in the MED countries, in limiting the sup-
ply response of domestic firms and in reducing the region’s attractiveness as a
host for foreign direct investment, especially for exports-oriented companies.
Although the four countries have launched reforms in several areas, consider-
able acceleration and widening of the reform agenda are required to catch up
with other regions. Such growth-promoting structural reforms relate to the
general business and investment climate, including reduction of the role of
the state in economic activities. Broad liberalization of the domestic market
also needs to be accompanied by improved competition policies, strengthened
property rights, and other complementary reforms.
Although private investment and growth depend on reforms, the opposite
also holds: without growth there will be no strong political constituency sup-
porting a reform program. Economic policy reforms aimed at liberalizing the
economy and opening up the sphere for private sector activities, will succeed
only if indeed there is an adequate supply response from the private sector to
replace the public sector as the main source of growth and employment. Thus,
because of self-fulfilling expectations, only a credible reform process will gen-
erate the response necessary for its own success. Credibility will hinge on
whether the private sector (domestic and foreign investors) believes not only
that the reforms undertaken are sufficient to improve the economic environ-
ment, but also that they will not be reversed.2
Whereas the four MED countries succeeded in maintaining macroeco-
nomic stability throughout the 1990s, they lag behind their comparators in
both the scope and speed of structural reforms. Figure 10.4 shows two com-
posite indices of progress on reforms for the MED countries and their
comparators—here, Mexico and two CEE countries, Poland and Hungary
(data are lacking for the Czech Republic for 1985–2001). One index relates to
economic stability and comprises fiscal and current account balances,
exchange rate black market premium, and inflation rates. The other refers to
structural reforms and comprises tariff rates, tax rates, PPP distortions, and
privatization revenues. 3
As can be seen, the MED countries have been more successful than the
comparator countries from the point of view of macroeconomic stability:
they were more stable at the outset and have remained comparatively stable
over time. However, with respect to structural reforms, the picture is very dif-
ferent: a difficult mix of an unfavorable starting point, in terms of the level of
distortions present in the economy, and slow progress on removing those dis-
tortions, left the MED countries far behind Mexico and the CEE accession
countries in the 1990s. The structural reform index for the MED countries not

Breaking the Barriers to Higher Economic Growth274
only starts at a lower level than the comparator countries, it also improves at
a slower speed, further widening the gap.
For example, although average unweighted tariff rates for the MED coun-
tries were reduced by almost a third over the 1990s, they remained more than
twice the level of those of Hungary, Poland, or Mexico. Mexico’s tariff rates
were more than halved between 1985 and 1998, and whereas tariff rates were
not much more reduced in the CEE countries than in MED countries over the
1990s, they were already at a lower level from the outset.4Privatization pro-
ceeds, however, increased rapidly in the 1990s in the comparator countries,
but remained relatively small in the MED countries.
What explains the lag in implementing economic reforms? The most com-
pelling arguments are related to the motivations to undertake reforms within
the specific social and political context facing the MED countries.
First, international experience shows that reforms are easier to launch in
times of deep economic crisis when there are no alternatives left, for example,
a situation with unsustainable economic imbalances, and when other
resources have been depleted or become inaccessible. As we have seen, there
has been no major stabilization crisis in the MED countries since 1985.
Moreover, although aid inflows have fallen over time, they remain much high-
er than in other regions (figure 10.5). Combined with sizable fuel exports
receipts, especially for Egypt, the continued availability of foreign resources
may have reduced the incentive to foster economic reforms.
Further (and related), the very structure of the EU-MED agreements—a
gradual and back-loaded trade liberalization scheme together with front-
Figure 10.4. Progress on Economic Stability Reforms (Left) and Structural Adjustment (Right), MED and Comparator
Countriesa, b
30
40
50
60
70
80
90
1985 1990 1995 1998 2001
MED4
MEX
CEE2
60
62
64
66
68
70
72
74
76
78
80
1985 1990 1995 1998 2001
MED4
MEX
CEE2
Source: World Bank 2003 employment study references.
a. The Economic Stability (ES) and Structural Reform (SR) indices are composite indices. The ES index is a weighted average of indices for fiscal and
current account balances, exchange rate black market premium, and inflation rates. The SR index is a weighted average of indices for tariff rates,
tax rates, PPP distortions and cumulative privatization revenues. The separate indices are calculated so that 100 = best value for all countries and
all years 1985–1998. The larger the deviation from 100, the worse the reform performance of the specific country. To arrive at an average for each
set of countries, the individual composite indices have been averaged (unweighted). b. CEE2 = average for Poland and Hungar y.

Part II: Labor Markets and Human Capital 275
loaded financial assistance to support reforms—does not provide incentives
to accelerate economic restructuring. The early financial assistance intended
to ease the transition and implementation costs of reforms may instead delay
them by providing alternative means for financing an inefficient economic
structure. In addition, the reforms imposed by the agreements—dismantling
of trade barriers for industrial goods—have a long implementation period on
the part of the MED countries. Although other complementary reforms are
included on the agenda, either for future negotiations (services and agricul-
tural goods), or as areas for harmonization and cooperation (for example,
product standards, competition policy, international property rights), there is
no mechanism within the agreement that would ensure timely implementa-
tion and guarantee continued adherence.
Finally, the present political and social environment in MENA is proving
a particularly difficult setting for launching reforms. The combination of low
growth, high and increasing unemployment rates, and sometimes worsened
poverty levels has resulted in a tense social climate where political survival
may hinge on the state retaining its role as a guarantor of employment, and
in which domestic and foreign private investors are deterred from investing
because of social, political and economic uncertainty. The result is low
growth and job destruction, which further increases the instability of these
countries.
The high uncertainty surrounding the political situation as well as the
reform agenda in the MED countries is a deterrent to private investment,
whether foreign or domestic. There is strong evidence from MENA countries
and elsewhere that uncertainty has an important negative impact on invest-
ment decisions (Sondergaard 2001). The irreversible nature of most invest-
ments will make investors hesitant to act in a high-risk environment—there is
Figure 10.5. Average Annual Inflows of Aid, US$ per Capita
0
10
20
30
40
50
60
70
80
MED4 CEE3 Latin America
and the Caribbean
Sub-Saharan
Africa
1990–94
1995–99
2000–04
Source: World Bank 2006.

Breaking the Barriers to Higher Economic Growth276
almost always the option to wait and see what happens today, and adjust the
investment strategy so as to make an optimal investment choice tomorrow.
The contrast to the accession partnerships in this respect is telling (table
10.3). The preaccession strategy for the CEE countries was based on strict and
detailed accession criteria and contained precise commitments on the part of
the candidate country relating to a broad agenda, including democracy,
macroeconomic stabilization, industrial restructuring, and the adoption of
the acquis communautaire. The criteria included free movement of goods,
services, capital, and labor (albeit with a transition period), and implied a
complete overhaul of institutions, law, and regulations. Moreover, although
the reform agenda was being complemented with financing along the way,
these reforms were not parallel but preconditions for EU accession. The CEE
countries therefore essentially faced a back-loaded process, in that the main
benefit it provided came after reforms had been undertaken.
The comparison with NAFTA is, at first glance, less clearcut. Like the EU-
MED agreements, NAFTA is limited to enforcing trade liberalization—
although it included service liberalization early—and possible complementa-
ry reforms are on the agenda as areas for cooperation without any binding
commitments regarding content, implementation, or timing. However, the
Mexican accession to NAFTA was in large part the logical conclusion to, rather
than the beginning of, a process of liberalization undertaken unilaterally by
the Mexican government since the mid-1980s (Galal and Hoekman 1997).
Thus, the purpose of the agreement was more to anchor the substantial
achievements already made in liberalizing the Mexican economy than to actu-
ally initiate liberalization reforms, as has been the case with the EU-MED
agreements. As mentioned earlier, Mexican tariffs have not fallen significant-
Table 10.3. The Complementary Reform Agenda Is Missing in the EU-MED Agreements
Reforms included
in the agreements EURO-MED U.S.–Mexico NAFTA EU accession
Free Movement
Industrial goods Yes Yes Yes
Agricultural goods To be negotiated Yes Yes
Services To be negotiated Yes Yes
Capital No No Yes
Labor No No Yes, with transition
period
Complementary Reforms
Competition Policy
IPRs
Privatization
Company Law
Financial Sector Reform
Included as areas
for cooperation and
harmonization, but
no mechanisms for
implementation
Included as areas
for cooperation and
harmonization, but
no mechanisms for
implementation
Included as precon-
ditions for
accession

Part II: Labor Markets and Human Capital 277
ly during the 1990s, but in 1990 were already at half the level of (the average
of) those of the MED countries in 1999. The difference between NAFTA and
EU-MED distinguishes the role of a free trade agreement as an anchor to
enhance the credibility of reforms already undertaken, from that of a vehicle
for initiating reforms.
The Looming Social Crisis and the Labor Market in the MED Countries
At the same time, the comparatively weak economic performance in the MED
countries poses a serious threat to political and social stability in the region.
The combination of economic stagnation or recession, massive migration to
urban areas, and a demographic transition that translates into a rapidly grow-
ing workforce, is increasing pressure on the labor market and, along with it,
political and social unrest. Evidence from the region and elsewhere shows that
from the perspective of political stability, the lack of job opportunities for the
educated and politically vocal middle class is particularly serious.
30. Labor markets in the MED countries are characterized by low or no job
creation and low turnover, and the trends are very worrisome. First, average
overall unemployment rates are already reaching high levels and invariably
increased over the 1990s (table 10.4). More important, the rate of unemploy-
ment is much higher among the young, the educated, and the new entrants to
the labor market, and again, the tendency is worsening. In Egypt, for example,
new labor market entrants made up around 70 percent of all unemployed in
1998, compared to 58 percent in 1988. Of these unemployed, first-time job-
seekers, more than 90 percent had completed at least secondary education.
More than 90 percent of all unemployed were aged between 15 and 30 years,
and 80 percent of all unemployed had secondary education or more (Assaad
2000).5The picture is similar for the other MED countries.
These characteristics are all the more serious, given that the inflow to the
formal labor market is bound to increase over the coming years. As in most
developing countries, the urban population is still growing fast, rapidly
increasing the supply of job seekers in cities: in Jordan and Morocco, urban
population growth is almost 3 percent per year; in Morocco and Tunisia, it is
many times higher than rural population growth (figure 10.6). Moreover, the
combination of a demographic transition increasing the available workforce
and improved access to education will further increase the share of educated
new entrants in the labor market. In Egypt, the number of people in the labor
force with at least secondary education increased, on average, by some
440,000 per year between 1988 and 1998. Measured as a percentage of total
labor force, this is equivalent to an annual increase of 3 percent. At these
arrival rates, the net addition to the Egyptian labor force in the coming years
would consist entirely of mid- or highly skilled labor. Two-thirds of the
Egyptian labor force would have at least secondary education by 2010, with

Breaking the Barriers to Higher Economic Growth278
correspondingly high expectations of a payoff to their human capital invest-
ments (figure 10.7).
Traditionally, the public sector has constituted a major source of employ-
ment for the educated middle class in the MENA region. Up until the mid-
1980s, the parallel expansion of the parastatal sector and of publicly provided
services absorbed most of the skilled labor. As the economy slowed down—
and with it, private sector job creation—the public sector continued to absorb
the residual workforce, in effect acting as a form of unemployment insurance.
By now it is clear, however, that the public sector can no longer sustainably
absorb new cohorts of well-educated workers as it has reached its upper lim-
its in absorbing fiscal resources (figures 10.8 and 10.9). Most countries are
Table 10.4. Unemployment in MED Countries by Age and Level of Education
Egypt Jordan MoroccoaTunisia
(1998) (2005) (2004) (2001)
Overall unemployment rate, 1988–89 5% 10% 16% 15%
Overall unemployment rate, latest available year 8% 15% 18% 15%
Youth unemployment rate 15% 39% 33% 35%
(ages 15–19) (ages 15–19) (ages 15–24) (ages 15–19)
19% 29% — 25%
(ages 20–29) (ages 20–24) (ages 20–29)
Young unemployed by age group, 84% 81% 60% —
% of total unemployed ages 15–64 (% unemployed (% unemployed (% unemployed
ages 15–29) ages 15–24) ages 15–34)
Unemployed with secondary
education or more, % of total 80% 35% — 42%
unemployed
Unemployment rate, economically active — 18% 26%b—
population with tertiary education
Source: Egypt: Assaad (2000); Morocco: Direction de la Statistique; Jordan: Jordan Dept. of Statistics; Tunisia: IMF (2001), World Bank (2004).
Note: Latest available year 1998–2005. a. Urban areas. b. With diploma.
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
Egypt Jordan Morocco Tunisia
urban
rural
Figure 10.6. Urban Population Growth High (Average Growth per Year, 2000–04, Urban and
Rural Population)
Source: World Bank 2006.

Part II: Labor Markets and Human Capital 279
12 13 13
5
9
17
0
5
10
15
20
25
30
35
1988 1998 2010
million people
with secondary education or
more
with less than secondary
Figure 10.7. Increasing Share of the Labor Force Expects Payoff for Education (Egypt)
Source: Authors’estimates based on Assaad 2000.
Figure 10.8. Public Sector Absorbs Large Share of Employment and Resources
0
5
10
15
20
25
30
35
40
45
Egypt Jordan Morocco Tunisia
public employment, % of total
gov wage bill, % of GDP
Source: World Bank 2006, World Bank 2003. Latest available 1997–2004.
Figure 10.9. Weight of Wage Bill High in MED Countries Compared to Other Regions
gov wage bill, % of GDP
0
2
4
6
8
10
12
14
MED4 Sub-
Saharan
Africa
Latin
America
and the
Caribbean
East
Asia
OECD Europe
and
Central
Asia
South
Asia
Source: World Bank 2006, World Bank 2003. Latest available 1997–2004.

Breaking the Barriers to Higher Economic Growth280
faced with a pressing need for public workforce downsizing, and at any rate,
expansion of the public sector is likely impossible.
New jobs for skilled (and unskilled) workers will consequently have to be
created in the private sector. Yet the domestic private sector has so far failed to
provide a source of employment growth—in large part because of persistent
structural problems in the economies, including the strong presence of the
state.
The cost of long-term youth unemployment and the resulting disillusion-
ment of a whole generation could be very high. The region is already seeing a
large exodus of highly skilled people. More important, social and economic
malaise could in itself block the reform process necessary to reverse the nega-
tive trends. In this way, the political and social costs of high unemployment
rates and worsening poverty trends limit the possibilities for public sector
reform and wide-scale industrial restructuring. This status quo bias is exacer-
bated by the phenomenon of self-fulfilling expectations, which is creating a
vicious negative growth circle. The high political uncertainty and instability
deter private investment and increase capital flight, which further depresses
growth rates, and thus further increases political instability and uncertainty.
New Directions for the Barcelona Framework
It is clear that the current framework for cooperation and integration between
EU and the MED countries so far has not proved sufficient either to achieve
the goals set out in the Barcelona agreements, or to mitigate the social and
political tensions in the MED countries. Instead of contributing to growth
and savings, the increasing workforce has become a barrier. Political and
social phenomena are reducing the incentives for reform, making the coun-
tries unattractive for investment and unfit for productivity improvements.
Given this dire background, this paper proposes that the Barcelona frame-
work be enriched along several dimensions. First, the constituency that should
find it in its interest to support the EU partnership needs to be broader and
stronger. In the present framework, one can hardly find groups within MED
countries that can clearly identify benefits from the agreements for them-
selves. On the contrary, many see costs: the business sector, which has to adapt
to European competition; the workers, who may bear some of the costs of
adjustment to trade liberalization; and the young (and less young) popula-
tion, who find it increasingly difficult to enter Europe because of more strin-
gent entry requirements. Political support for the Barcelona agreements is
weak.
Second, more credible mechanisms for anchoring the implementation of a
broad and intense agenda for reforms complementary to trade liberalization
should be developed, to make sure that the expected benefits from the agree-
ments do actually materialize sooner rather than later. It is clear that financial

Part II: Labor Markets and Human Capital 281
support cannot be the only answer, even though it may be part of a larger
package. On the contrary, as argued previously, merely providing financial
flows could have a perverse effect on incentives to undertake reform.
In this respect, widening and accelerating the agenda for liberalization on
agricultural trade is important for the agricultural constituency, and may play
a role in helping reduce rural poverty. Yet this measure on its own is unlikely
to be enough to reverse the trend of overall rural migration to the cities, for
several reasons. First, productivity improvements in the agricultural sector in
response to trade liberalization are more likely to actually reduce demand for
labor. Second, the severe water shortage characteristic of most of the MED
countries will limit the supply response from the rural sector. Third, compe-
tition from Poland and Hungary is also likely to limit demand from the EU
for agricultural products from the MED countries.
Similarly, broadening the agenda to services liberalization may also
increase the benefits of the agreements, in particular as its implementation
would highlight the need for harmonization and reforms in a number of areas
outside tariff reduction. However, risks remain that the complementary
reform agenda does not materialize.
A more credible reform process would hence require a strong early com-
mitment on the part of MED governments to carry out and maintain critical
reforms in complementary areas such as public sector reform and financial
sector modernization. These reforms should be accompanied by official sup-
port from the EU, a signal that would then further support the credibility of
the reform program.
Third, labor mobility at present is not included in the partnership frame-
work, nor is it on the agenda for negotiations. However, the strong links
between labor market conditions in MENA on the one hand, and political
resistance to growth-enhancing reforms on the other, suggest that migration
could well be an important ingredient in improving the framework. Indeed, a
well-designed significant program of managed migration from MED countries
to the EU can be a major vehicle for achieving the objectives of the partnership.
First, it would create an immediate constituency that benefits from the part-
nership. Second, it would help relieve socioeconomic and political pressures
from unemployment in MED countries, and facilitate undertaking the com-
plementary reform agenda. This would smooth the impact of the transition
costs of trade liberalization, as well as the impact of the bulge in the demand
for jobs over the coming 15–20 years that results from the demographic tran-
sition. Both factors would help support and enhance the reform agenda.
Intensified trade relations, as in the EU partnership, have been put forward
as a substitute for interregional migration, and indeed as a means of contain-
ing mass migration from poorer to richer areas. First, from a purely theoreti-
cal perspective, factor endowment trade theory would suggest that trade in
goods is indirectly trade in factors, including labor. Second, increased trade is

Breaking the Barriers to Higher Economic Growth282
expected to lead to higher growth, especially in poorer countries, which in
turn leads to economic convergence, thereby reducing the incentives for
migration in the first place.
Yet research on trade and migration shows that trade liberalization and
migration controls are not necessarily substituting policies, at least not over
the short term (Faini et al. 1999). In fact, historical evidence suggests the
opposite: the globalization drives at the end of the 19th century and in the
20th century postwar period were accompanied by a surge in migration flows,
as both labor and goods benefited from a decline in communication and
transportation costs. High-income countries that liberalize trade with
middle-income countries are not likely to have to fear massive migration
flows. However, trade liberalization between high-income countries and sig-
nificantly poorer countries may at least initially result in increased pressures
for migration, if production is polarized and if migration previously has been
repressed by financial constraints.6The NAFTA agreement,for example,appears not to have curtailed migration over the short term.Migration and trade policies will therefore need to accompany one another to increase the benefits for all parties over the medium term.
In the following discussion, we undertake to advance the debate by devel-
oping the arguments for increased movement of people between the MED
countries and Europe.
People Mobility between EU and MED Countries as Part of the Solution
What are the possibilities for channeling migration in ways that maximize the
potential current and future gains for the sending country and use the
reduced labor market pressures to make a big push for reforms? Migration,
and perhaps especially from Muslim countries, has a bad name in Europe. A
variety of reasons explain the poor reputation: high overall unemployment
rates in the EU; previous large waves of migrants, especially from the Maghreb
(Morocco) region (to France, Italy, Spain) and Turkey (mostly to Germany),
among whom unemployment rates are often higher than among nationals;
high fertility rates among immigrant communities; continued flows of illegal
immigration, among others. The strong tightening of migration policies in
the 1980s and 1990s is evidence of the changing perception toward foreign
labor. The debate over the perceived cost of opening the borders to workers
from the EU accession countries in Central and Eastern Europe, and the
imposition of a seven-year transition period for free movement of labor to
this end, has highlighted the lack of a common agreed framework for migra-
tion within the EU. It has also revealed the tension surrounding the manage-
ment of migration flows and the concerns of EU members—in particular in
the Mediterranean area, which at present receives an important share of
inflows from Maghreb. This paper, however, advances the benefits from a
scheme with managed migration for mid-skilled labor: that is, with secondary
education from the MENA region, which would differ from past waves of
immigration.

Part II: Labor Markets and Human Capital 283
Critics would argue that from the perspective of EU countries, opening the
borders for workers outside the EU will put upward pressure on the already
high unemployment rates in the EU, and especially in the countries which, for
cultural and historical reasons, are most likely to be recipients of large-scale
immigration. Indeed, unemployment rates in EU (15) averaged 8 percent
between 2003 and 2005 and reached 10 percent in France, Germany, and
Spain (OECD 2006). From the perspective of the MED countries, on the other
hand, migration of skilled labor would imply losing its best workforce to for-
eign countries, at a time when capacity and skilled labor are considered key in
improving international competitiveness and in the leveling of the private
sector.
Yet several facts speak in favor of a deeper integration of labor markets
between EU and the MED countries. First, whereas the present trade-focused
format of the EU-MED agreements seems insufficient to lead to economic
convergence,international and historical experience shows that migration can
be a major force in bringing it about. A study by Taylor and Williamson
(1994) suggests that migration between the old and the new world at the end
of the 19th century was the main factor behind convergence in real wages,
GDP-worker, and GDP-capita between Europe on the one hand and the
United States and Australia on the other. In short, migration served to take off
pressure on the labor markets in Europe and instead provide a much needed
workforce in the new countries, with an impact on both GDP and the com-
pensation of workers (figure 10.10).
More recently, migration within the Middle East and North Africa has also
served as a means of convergence. Migration has provided a vehicle for dis-
tributing oil revenues within the region: millions of Egyptians, Palestinians,
GDP per worker
1.5%
1.3%
1.7%
0.9%
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
United States Italy
with migration
without migration
Real wages
1.0%
1.6%
1.3%
1.0%
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
United States Italy
percent
percent
with migration
without migration
Figure 10.10. Average Annual Growth Rates of United States and Italy between 1870–1910, with Migration (Actual)
and Without Migration (Counterfactual)
Source: Authors’estimates based on Taylor and Williamson 1994.

Breaking the Barriers to Higher Economic Growth284
Yemenis, Syrians, Lebanese, and Jordanians have worked in the oil-producing
Gulf countries for extended periods, sending home money to their families.
Later on, many have returned home as investors or skilled and experienced
labor. As shown in figure 10.11, GDP growth in these labor-exporting coun-
tries moved together with oil revenues and GDP growth in the Gulf countries
during the 1970s, 1980s, and 1990s (except for the beginning of the 1980s),
suggesting important spillover effects. With the Gulf war at the beginning of
the 1990s, however, labor demand fell radically in the oil-rich economies,
leading to a large involuntary return of workers. The shrinking labor market
in the Gulf has been one key factor in pushing unemployment rates up in
Jordan and Egypt, for example.
Further, some access to the EU labor market for the MENA countries, even
if limited, could potentially bring important economic, political, and social
gains for the sending countries. The risk for “brain drain” is a real and serious
one, but the migration option must be contrasted with the current situation,
where skills are not put to use. Long-term unemployment persistence, espe-
cially for new and well-educated entrants to the labor market, is at present
both increasing frustration and eroding skills that are not maintained through
on-the-job training. The resulting political and social tension is in turn ham-
pering the investment climate and economic growth.All the while, labor skills
are becoming obsolete.
Moreover, the lack of opportunities is already pushing highly skilled labor
to leave the region, but to other regions such as the United States, Canada, or
Australia. By way of illustration, out of the more than 50,000 Egyptians
migrants that resided in the United States in 1990 (as recorded by the census),
Figure 10.11. Oil Revenues Trickle Down in the MENA Region
⫺6
⫺4
⫺2
0
2
4
6
8
10
12
1975–79 1980–84 1985–89 1990–94 1995–99
average annual oil exports growth (1995 US$), Gulf countries
average annual GDP growth, Gulf Countries
average annual GDP growth, Maghreb Countries
Source: World Bank 2006, United Nations 2001.

Part II: Labor Markets and Human Capital 285
98 percent had completed secondary education and almost 80 percent had
tertiary education (Carrington and Detragiache 1998).
Although geographical proximity to the EU area will induce emigration, it
will also facilitate return migration. This is not necessarily the case for
migrants who have taken the step to move to continents much further away
from the MENA region. The more rooted immigrants become in the host
country, the less likely they are to return home; and as immigrants lose their
personal and professional ties to their home countries, the benefits associated
with sending workers abroad will also weaken.
An appropriately managed migration to the EU area, in contrast, would
bring important benefits in the form of foreign exchange savings (workers’
remittances) and simultaneously provide better management and capitaliza-
tion of human capital assets. Such a scheme, although focused and limited in
size, would at the same time relieve some critical pressure on labor markets in
the MENA countries. With several migrants returning home after a certain
time spent in the EU, the region would also gain from upgrading skills and
business connections.
Workers’ remittances already make up an important source of balance of
payments support in the Middle East and North Africa Region, and have, as
shown in table 10.5, by far exceeded direct investment inflows in the four
MED countries.
Migration can also provide the means for skilled labor to capitalize on
investment in human capital already made, in a place where returns to educa-
tion are higher, and to continue upgrading skills overseas. When conditions at
home have improved, they can return home with accumulated savings as well
as new skills. There is strong evidence that an important percentage of
migrants do not lose their ties with their home countries but choose to return
home after a certain time abroad. It has been estimated that of all U.S. immi-
grants, one-third return to live in their home country at some point during
their lives, and 20 percent return within the first 10 years after arriving in the
United States (Stalker 2000).
Returning migrants often bring back significant skills, experience, and con-
nections within business networks in Europe or elsewhere, as well as savings
Table 10.5. Workers’ Remittances and Foreign Direct Investment, Percent of GDP (Average)
1990–94 1995–99 2000–04
Workers’ Workers’ Workers’
remittances FDI remittances FDI remittances FDI
Egypt 10.8 1.4 4.4 1.1 3.3 0.9
Jordan 15.5 0.1 20.2 2.3 19.4 4.2
Morocco 7.1 1.4 5.7 0.1 8.2 1.8
Tunisia 3.8 2.4 3.7 1.9 4.8 2.8
Source: World Bank 2006.

Breaking the Barriers to Higher Economic Growth286
for investment. There are also indications that savings in the form of workers’
remittances are more directly translated into growth than other forms of sav-
ings, since they tend to be used for productive investment to a higher degree.
For example, a study of the use of international remittances by households in
rural Egypt showed first, that a substantial share of migrant household
incomes is spent on investment, and second, that the propensity to invest is
larger than for nonmigrant households (Adams Jr. 1991).
Clearly, however, the propensity to return home to put savings and experi-
ence to use will depend on whether the work and investment climate in the
home country improves or not. International experience shows that return
migration is ruled by “push” factors—worsening labor market climate in host
countries—but also to a large extent by “pull” factors, that is, increasing
attractiveness of the home country (Dustmann 1996; Gmelch 1980).
The southern European countries’ experiences of migration to northern
Europe (primarily Germany, France, and Switzerland) indicate an inverse u-
pattern in the propensity to migrate (figure 10.12). As work opportunities
increased in the buoyant northern European economies in the 1960s, migra-
tion from Greece, Portugal, Spain, and Turkey rapidly took off, and fell
together with the declining opportunities after the first oil shock. However, as
the economies in Northern Europe improved in the 1980s, net migration
from Southern Europe did not pick up, although there were still considerable
gaps in real wages and incomes between the two regions. Although stricter
immigration policies did play a role, it does not fully explain the trend toward
return migration, since migrants from other countries continued to arrive in
the northern European countries.
Empirical work suggests that the propensity to migrate in the southern
European economies was determined by income disparities between the
home and the host country and by the individual level of wealth of the poten-
tial migrant—that is, his resources for migration, but also by the strength of
cultural and social ties to the home country (Faini and Venturini 1994).
Hence, in a poor country migration is likely to initially increase with growth,
as potential migrants acquire the financial means and the level of education
necessary to go abroad. But as conditions at home improve, the incentive to
migrate is reduced and some of the migrants return home to their preferred
country of residence, becoming an important source of dynamism and a
bridge between the two markets.
Similarly,it could be argued that the quality of investment made by return-
ing migrants—in choice of sector, productive use, growth potential, and
spillover effects—is likely to depend on opportunities for investment in the
home country (Chandavarkar 1980). The quality of the investment climate at
home will thus be important in determining not only the propensity to
return, but also the productive use of remittances.

Part II: Labor Markets and Human Capital 287
Finally,the willingness of political decision makers to undertake important
but politically difficult economic reforms is likely to increase with migration.
Outflows of workers would alleviate labor market pressures during structural
reforms that may have a potentially negative bearing on the labor market, at
least in the short run—including civil service reform, changes in the educa-
tion system, and so forth. This can create a positive feedback loop, in which
migration allows for a more decisive take on economic reforms without
immediate large social costs, which in turn improves the investment climate
and growth prospects, and thus attracts migrants back after a certain time.
Can the EU Cope?
However strong the arguments from the sending countries, a migration
scheme that is costly to the recipient countries, especially from the prospect of
labor market impacts, will not be politically feasible. The expected impact on
the EU labor market of taking in more skilled migrants from North Africa is
therefore a key issue in assessing the viability and applicability of a migration
scheme. It is evident from the current debate on migration in Europe that the
perception of migration as socially and economically costly for the host coun-
tries is widespread. As discussed previously, there are several counterargu-
ments to increasing migration: high unemployment rates suggest that the EU
labor markets can no longer absorb their own workforce, let alone take in for-
eign labor. This further increases already high levels of immigration that
would lead to even higher unemployment, or a drop in real wages, or both; the
EU needs to concentrate on accommodating workers from the accession
countries in the CEE, rather than on nonmember countries (Maghreb) that
Figure 10.12. Migration Rates in Southern Europe, 1960–1988
(Migrants per 1,000 Inhabitants)
0
5
10
15
20
25
1960 1962 1964 1966 1968 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988
Greece
Portugal
Spain
Turk ey
Source: Faini and Venturini 1994.

Breaking the Barriers to Higher Economic Growth288
have already been allowed their share of immigration for decades. As we will
see, however, these arguments do not necessarily hold up to close scrutiny.
First of all, although the EU has received a large share of immigrants, the
share of foreign-born labor force remains lower than in the United States and
Canada. Yet unemployment rates in the United States and Canada have
remained below those of the EU area. This could be due to a variety of factors,
including differences in labor markets, migration types, and selection criteria,
but the fact remains that the United States and Canada have managed migra-
tion flows and labor market pressures better.
Second, it is true that earlier waves of migration from the MENA region,
and in particular from the Maghreb countries, have resulted in an important
presence in several EU countries, especially Belgium, France, Italy, and Spain.
During the industrial boom of the 1960s, several EU countries actively
recruited workers from the Maghreb to occupy low-skill jobs in the industri-
al sector.As the recession hit Europe in the mid-1970s,migration policies took
a sharp turn and access to the EU market for foreigners tightened consider-
ably. However, as economic conditions subsequently worsened in the MENA
region as well, in particular after the collapse in oil prices in the mid-1980s,
there was still an important supply of migrants. With stricter migration poli-
cies, illegal immigration increased, in particular to the southern European
countries, where the Mediterranean coast poses an obstacle to tight border
entry control. With the exception of a sharp increase in Moroccan immigra-
tion to Italy and Spain (partly a reflection of regularization of existing illegal
immigrants), there has been little or no increase of the share of North African
immigrants since the mid-1980s (table 10.6). Immigrants from other coun-
tries have instead increased their presence in Northern Europe.
Table 10.6. Stock of Foreign Population by Nationality (Percent of Total Foreign), Selected EU Countries, Earliest and
Latest Available Yeara
As % of total Belgium France Germany Italy Netherlands Spain
foreign population 1986 2001 1982 1999 1986 2002 1986 2001 1985 1998 1985 1998
Morocco 15 11 12 15 .. .. 1 12 10 10 3 21
Algeria 1 1 22 15 .. .. .. .. .. .. .. ..
Tunisia .. .. 5 5 .. .. 1 3 1 0 .. ..
Egypt .. .. .. .. .. .. 2 2 .. .. .. ..
North Africa 16 12 39 35 0 0 4 17 11 10 3 21
Turkey 9 5 3 6 32 26 .. .. 12 11 .. ..
Poland .. 1 2 0 3 4 2 2 .. .. .. ..
Other 75 82 56 59 65 70 94 81 77 79 97 79
Total foreign population 853 846 3,714 3,263 4,513 7,336 450 1,363 1,217 1,675 539 1,109
Source: OECD (2003).
1. Definitions of foreign population differ across countries. In the table above, the numbers refer to foreign population, except
for the Netherlands, where it refers to the foreign-born population. For Italy and Spain, figures include results of regularization
programs (Italy: in 1987–88, 1990, 1995; Spain: in 1991).

Part II: Labor Markets and Human Capital 289
Third, demographic developments in the EU would in fact imply that
increased workforce migration would be highly desirable. Under the prevail-
ing pay-as-you-go social security system, a rapidly aging population is threat-
ening to weigh heavily on future workers and taxpayers who will need to pro-
vide for retirees with increasing life expectancy. Short of an immediate and
drastic change in fertility rates in the EU area, migration provides the only
means of reinforcing the shrinking workforce. At present, the EU receives
around 0.6 million migrants (on a net basis) every year. Demographic projec-
tions by the UN suggest that to keep the working-age population constant,
another million people would need to enter the EU each year (UN 2000). To
keep the ratio of old population to working population constant, that is, to
ensure that each retiree can rely on the same number of workers, the region
would need as much as 10 million additional immigrants per year (figure
10.13).
From the perspective of replacement migration: can and will the additional
workforce needed be more than adequately provided by the accession coun-
tries in Central and Eastern Europe? The discussion regarding EU enlargement
and labor mobility has revealed large differences in perception regarding the
potential pool of immigrants from Central and Eastern Europe: whereas EU
members fear mass immigration, the CEE countries appear to expect emigra-
tion to be more limited. Several characteristics that distinguish the CEE coun-
tries from MED countries speak in favor of the latter scenario. First, the CEE
countries have already started their integration and convergence process, and
the lion’s share of migration might already have taken place at the beginning of
the 1990s after the fall of the Iron Curtain. Net inflows from Poland to
Germany, for example, declined from an average of 58,000 people per year in
Figure 10.13. European Union: Current Net Migration (1995–1999) and Projected Average
Annual Net Migration Needs, 2000–30
0.7
10.2
1.6
0.6
0
2
4
6
8
10
12
actual net
migration
1995–99
to keep the
population
constant
to keep the
working age
population
constant
to keep the
ratio of old
population to
working population
constant
millions
Source: Authors’estimates based on United Nations 2000.

Breaking the Barriers to Higher Economic Growth290
1989–1991, to 15,000 in1999–2001 (OECD, 2003). Second, the accession coun-
tries are also at a much higher per capita income than the MED countries and,
following the pattern of the southern European countries accession, improved
living conditions at home are likely to dampen the tendency to migrate.As dis-
cussed earlier, evidence suggests that trade liberalization of high-income coun-
tries with middle-income countries tends to foster convergence and discourage
migration. Third, their demographic structure is mirroring that of the EU, with
shrinking working populations, reducing the number of workers who would
or could migrate (table 10.7). Indeed, as shown above, the accession of Greece,
Portugal, and Spain in the 1980s did not lead to mass immigration, in spite of
widely held fears, but had almost an opposite effect. The accession process pro-
vided better conditions in the home countries, which led people to return
home to take advantage of improved prospects.
A Managed EU-MED Migration Scheme—Mid-Skilled Labor in
Search of Experience
The prospects of relatively limited migration flows from CEE accession coun-
tries would open the way for migration flows from the MENA region to the
EU. However, these flows should be different from previous waves of legal
migration of unskilled labor, or the illegal migration flows currently taking
place. In line with changing characteristics and upgrading of the education
level of the MENA labor force, the new form of migration flows would con-
sist of middle-skilled labor going to work in the formal sector, with full eco-
nomic and social rights.
Focusing on managed migration to the formal sector means that only labor
with some skill proficiency will be targeted. In EU, comparatively high mini-
mum wages topped with high social security contributions (often amounting
to 100 percent of wages) result in a high minimum cost per worker. The high
cost of labor implies in turn that only workers with higher productivity, that
is, with some skill or education level, can get jobs in the formal sector.
In these circumstances the relative importance of MENA labor inflows
will be of comparatively small and manageable magnitude. Concentrating on
the MED countries—the most populous countries in the MENA region,
except Iran—the potential pool of migrants would be limited to the labor
Table 10.7. CEE Accession Countries: Labor Force Growth
Labor force Growth rate Share of EU (15)
2004 (million) p.a., 1999–2004 labor force, 2004
Czech Rep. 5.4 ⫺0.1% 2.8%
Hungary 4.4 0.4% 2.3%
Poland 19.8 ⫺0.2% 9.5%
Source: World Bank 2006.

Part II: Labor Markets and Human Capital 291
force with mid- to higher skill level of education. Although the share is
increasing rapidly, it remains small in absolute numbers compared to the EU
labor force (table 10.8).
The goals of an EU-MED migration scheme would be threefold: first, to
help tighten labor markets in MENA, especially for mid-skilled workers with
at least secondary education; second, help ensure that returns to education are
high; third, help produce experienced investors who, when returning to their
home country, would have built networks within the EU for future use in their
business activities.
The design of such a scheme should take into account previous experience
from migration schemes, within and outside the EU, to minimize illegal
immigration and maximize the benefits for both sending and receiving coun-
tries. It would need to ensure that workers from the MED countries are chan-
neled to sectors where they are most needed, and where their ability to gain
useful work experience and contacts is high. It will also be important to put in
place mechanisms that facilitate and encourage voluntary return migration:
for example, by increasing business opportunities in the sending countries
over the long term. The migration scheme would thus be an integral part of
the EU-MED agenda, complementing other areas for cooperation, including
growth-enhancing economic reforms.
A survey of international experience of migration schemes, whether for
temporary and permanent, high-skilled and low-skilled workers, presents
mixed results (Coulibaly 2001).A key lesson from the multitude ofschemes that have been implemented—and in some cases dropped— is that as
long as there is a demand for foreign labor,increased sending pressures,decreasing transportation costs,and an existing web linking immigrants and prospective migrants,immigration will take place irrespective ofthe policy in place.Even ifthe “front door”(legal economic migration) is closed,there will be entry through the “side door”(legal immigration for social and humanitarian reasons),or the “back door”(illegal immigration).
A managed migration scheme that responds to EU needs for mid-skilled and
skilled workers, although ensuring positive returns to migration and education
for MED economies, would therefore need to contain the following provisions:
• Focus on (i) mid-skilled labor (ii) on training experience in the EU, and on
(iii) encouraging return migration.
Table 10.8. Pool of Potential Migrants from MED Countries
A. Economically B. Labor force with C. Unemployed with
active at least secondary at least secondary
population education education
(millions) (millions) (millions)
Jordan 1.8 0.6 0.1
Tunisia 3.3 1.2 0.2
Egypt 22.3 9.3 1.4
Morocco 11.0 1.8 0.3
TOTAL 38.4 12.0 2.0
% of EU(15) labor force 21.1 6.6 1.1
Source: Labor force and unemployment numbers—Egypt, Jordan: 2004 data from WDI 2006; Morocco: 2003 data
from Direction de la Statistique; Tunisia: 2001 data from World Bank 2004, EU: 2004 data from WDI 2006. Shares for
LF and U by education (applied to LF numbers in column A): Egypt: 1998 data from Assaad 2000; Jordan: 2005 data
from Department of Statistics; Morocco: 1999 data from Boudarbat 2005; Tunisia: 2001 data from World Bank 2004.

Breaking the Barriers to Higher Economic Growth292
• In the absence of a common EU migration policy, allow for bilateral
arrangements, but with the support of a common EU Fund.
• Create an EU fund for cofinancing with the private sector. This fund would
be used to support the financing of both training programs, and return
migration-integration of graduating trainees or returnees. A three-year
trainee program could be offered to selected graduates geared toward the
industry and services sectors. A shorter training opportunity could be
offered to interested workers as part of company transfers under an EU-
MED Guest Worker Program. Trainees would be partly funded by the EU,
and partly by the private sector in the respective countries.At the end of the
training program, MED country nationals who receive a job offer should
have the possibility to stay.
• Provide assistance on both sides of the Mediterranean to help manage these
programs and assist trainees as well as returnees in their effort to adapt and
reintegrate. This would include technical and financial assistance needed
by returnees.
• Establish clear and accessible information channels for prospective EU
employers and MED employees. This could include a virtual market that
could allow for job and skills matching and training offers made by local or
foreign employers.
• MED governments must engage their responsibility and pledge their direct
participation in efforts to manage and indeed encourage return migration.
In addition to active participation in the scheme management together
with the EU (especially jobs and skills matching, and so on), these efforts
need to focus on setting the ground rules for a friendly business environ-
ment with safety nets for returnees, and facilitating the readmission of their
illegal immigrants.
• This framework could be augmented with easier entry for business visits
and educational purposes, which would both encourage human capital
development and strengthen the personal and professional links between
the regions.
The scheme would in itself be time-limited, but with stronger emphasis
both on encouraging voluntary return migration, and on providing the
opportunity for qualified MED country nationals to remain working in EU
after the trainee period. For example, the guest worker programs in Germany
and Switzerland, as well as the foreign trainee program in Japan, largely failed
as temporary schemes, as there were no mechanisms in place either to inte-
grate these workers in their host country, or to encourage (or coerce them) to
return. With a rapidly aging population, the EU may be obliged to adopt a
more pragmatic approach to migration. Here, the EU countries may have
something to learn from the experiences of other large labor-importing
regions, including the United States, Canada, and Australia, where economic

Part II: Labor Markets and Human Capital 293
migration is still possible, and where programs appear better targeted to the
needs of both sending and receiving countries.
Conclusions
The EU-MED partnership needs to be revitalized and broadened if it is to ful-
fill its purpose of encouraging economic growth and the process of integra-
tion between the two regions. In this paper, we have suggested that the broad-
ening take place along two parallel and complementary tracks. First, reform
efforts need to be intensified, to lay the groundwork for increased private
investment, economic growth, and job creation in the MED countries.
Second, allowing for limited economic migration flows between the regions
would ease off the demographic pressure on labor markets in the MED coun-
tries, and thus indirectly ease some of the political and social tension that are
blocking reforms.
There is a need to find mechanisms that enhance the credibility of the
reform programs in the MED countries. The EU-MED agenda needs to go
beyond trade. On the part of the MED countries, this will require a strong
early commitment by undertaking reforms in critical areas, including public
sector reform. EU support should accompany rather than lead these reforms.
A credible program that results in both EU official support and private FDI
would further support and facilitate the reform process, creating the positive
credibility feedback loop that is missing at present.
A migration program that would open up the EU labor market for mid-
skilled workers in MED countries could improve labor market conditions and
increase public support for both the EU-MED agreement as such, and for
reform efforts more generally. The risk of brain drain, although important,
must be seen in the light of what is happening at present, when new labor
market entrants see their skills eroding and where an exodus of skilled labor
is taking place, but with less positive spillovers than under a managed migra-
tion scheme. When economic reforms at home deliver better conditions in
due time, several of these migrants are likely to return home, bringing back
savings, experience, and contacts and the potential to contribute further to
economic growth. This would increase the potential gains for both recipient
and sending countries, and strengthen the link between the two regions.
Notes
1. Under MEDA I (1995-1999), a total of €3 billion was committed with a payment ratio of about 30
percent. Since 2000,MEDA II has been implemented. Between 2000 and 2004, €3 billion was commit-
ted, with a 77 percent payment ratio.
2. See, for example, Rodrik 1989 or 1991.
3. Note that these indices refer to reform outcomes rather than reform efforts per se.

Breaking the Barriers to Higher Economic Growth294
4. The EU-Med agreements are in fact by construction, leading to a transitional increase in effective
trade protection in the MED countries as the initial stages of tariff dismantling focus on intermediate
and capital goods.
5. The analysis in Assaad (2000) is based on the Egypt Labor Market Survey 1998 and the Labor Force
Survey of 1988. To ensure comparability between the two surveys, the author uses an extended defini-
tion of the labor force, including all employment in subsistence agriculture, and defines unemployed
as those seeking work. These definitions give rise to a lower unemployment rate than would otherwise
be the case.
6. Several factors will motivate international (economic) migration, including wage differentials,pref-
erences for the home country, and the cost of migration. In a very poor country, migration to a rich-
er area may be attractive from the point of view of wage differentials. but may be contained simply by
lack of resources. In higher middle-income countries, although prospective migrants have the
resources to move abroad, the wage differentials may not be a sufficient incentive, especially not if the
prospects for growth in the home country are good (Schiff 1996). The highest migration pressures
could therefore come from lower middle-income countries where the financial constraint is less bind-
ing, but where the wage differential still provides an incentive.
Bibliography
Adams Jr., R. 1991. “The Economic Uses and Impact of International Remittances in Rural Egypt.”
Economic Development and Cultural Change 39: 695–722.
Assaad, R. 2000. “The Transformation of the Egyptian Labor Market: 1988–1998.” Middle East and
North Africa Region. World Bank, Washington, DC.
Boudarbat, Brahim. 2005. “Job Search Strategies and Unemployment of University Graduates in
Morocco.” EBRD. May 5.
Carrington, W., and E. Detragiache. 1998. “How Extensive is the Brain Drain?”Finance & Development
(IMF) 36:46–49.
Chandavarkar, A. 1980. “Use of Migrants’ Remittances in Labor-Exporting Countries.” Finance &
Development, IMF, 17:2: 36–39.
Coulibaly, A. 2001. “A Survey of the International Experience of Migration Schemes.” Mimeo. Middle
East and North Africa Region. World Bank, Washington, DC.
Dustmann, C. 1996. “Return Migration: the European Experience.” Economic Policy: A European
Forum 22: 215–50.
Faini, J., and A. Venturini. 1994. “Migration and growth: The experience of Southern Europe.” CEPR
Discussion Paper No. 964. CEPR, London.
Faini, R., J. de Melo,and K. Zimmermann. 1999. Migration: The Controversies and the Evidence. CEPR,
London.
Gmelch, G. 1980.“Return Migration.” Annual Review of Anthropology (9): 135–159.
Galal, A., and B. Hoekman. 1997. Regional Partners in Global Markets: Limits and Possibilities of the
Euro-MED Agreements. Ed. CEPR, London.
IMF. 2001. Tunisia: Article IV.
Rodrik, D. 1989. “Promises, Promises: Credible Policy Reform via Signaling.” The Economic Journal
(99), 1989.
________ . 1991. “Policy Uncertainty and Private Investment in Developing Countries.” Journal of
Development Economics 36 (2)..
Schiff, M. 1996. “North South Migration and Trade: A Survey.” Working paper 1696, Research
Department. World Bank, Washington, DC.
Sondergaard, L. 2001. “Private Investment in Egypt, Morocco and Tunisia.” Mimeo, Middle East and
North Africa Region. World Bank, Washington, DC.

Part II: Labor Markets and Human Capital 295
Stalker, P. 2000. Workers without frontiers: the impact of globalization on international migration.ILO,
Geneva.
Taylor,A., and J. Williamson. 1994.“Convergence in the Age of Mass Migration.” NBER working paper
4711. NBER, New York.
World Bank. 2000. Trade Blocs. Washington, DC.
World Bank Employment Study. 2003. Unlocking the Employment Potential in the Middle East and
North Africa: Toward a New Social Contract. MENA Development Report. World Bank,
Washington, DC.
World Bank. 2004. Tunisia Employment.
United Nations. 2000. Replacement Migration.New York.
Data sources
World Development Indicators Database. 2006. World Bank, Washington, DC.
United Nations ComTrade Database. 2001. United Nations, New York.
Labor Statistics Database. 2001. International Labour Organization, Geneva.
Main Economic Indicators. 2006. Organisation for Economic Co-operation and Development.
Trends in International Migration. 2003. Organisation for Economic Co-operation and Development,
Paris.

Part III
Trade, Competitiveness,
and Investment

299
Cruise Control, Shock Absorbers,
and Traffic Lights
The Macroeconomic Road to
Arab Competitiveness
Mustapha K. Nabli
11
Good afternoon. I am sincerely happy to be here today to talk to this diverse
and distinguished group. The topic of competitiveness, and a conference
devoted exclusively to the subject, I think, is well overdue for the region. I also
think it is a good omen that the meeting has been designated the “First Arab
World Competitiveness Meeting,” so I’m encouraged that we can anticipate
sequels to this forum in the future.
This lunchtime dialogue has been set aside to discuss some of the macro-
economic drivers of competitiveness. I found myself uncomfortable with the
term drivers, because I think it creates an inaccurate picture of the process by
which an economy becomes internationally competitive.
The road to competitiveness is a long journey, and as such requires respon-
sible macroeconomic policy and a supportive business environment to facili-
tate long-term planning, promote investment and knowledge transfer, and
support the efficient allocation of resources to ensure that the drivers reach
the intended destination.
And the drivers on this road trip to competitiveness are the firms—
enterprises that find ways to create increasing value added, through their abil-
ity to invent products, adopt new technologies, and respond to changes in
market conditions at home and abroad.
Speech for the first Arab World Competitiveness Meeting at the World Economic Forum; Geneva,
Switzerland; September 9, 2002.

Breaking the Barriers to Higher Economic Growth300
So let me build on this analogy of a road trip and focus on what I think are
three of the key forces in the macroeconomic arena that determine whether
the drivers on the road to competitiveness are best equipped to succeed on
this challenging expedition of long drives, unexpected terrain, and the satis-
faction of an ever-approaching destination.
In my mind, one of the central requisites for competitiveness is appropri-
ate exchange rate management.In a sense, it is the choice between working the
brake and gas pedals yourself to determine your speed—or relying on cruise
control to do it for you.
And what are the pitfalls that can develop from inappropriate exchange
rate management? The most important, in my mind, is an overvaluation of
the currency. From almost all assessments of economic policy and perform-
ance throughout the world, it has been shown that best performers are coun-
tries that have maintained an “appropriate” exchange rate regime. In particu-
lar, countries that have been successful in promoting manufactured exports
have avoided exchange rate overvaluation.
Profitability of production hinges on prices—prices of inputs that go into
the production process, and the price that can be obtained in the market for
output. Overvaluation is damaging to competitiveness because it artificially
alters the price ratio between tradables and nontradables. So the producers of
tradable goods find they are less able to compete with either imported goods
or with other countries’ exports.
Economies that, in reality, have cost advantages in labor and domestically
produced inputs begin altering their production processes and substituting
for capital equipment and imported inputs. And the greater the overvaluation
that takes place, the more difficult it becomes for the existing drivers of com-
petitiveness to maintain their competitive edge, and the more it discourages
new drivers from getting on the road.
What has been the experience of the Middle East and North Africa region
(MENA) with exchange rate management? In a recent paper I co-wrote, we
examined this issue by looking at the degree of exchange rate misalignment in
the region. What we found is discouraging. In the past three decades, MENA
countries experienced substantial overvaluation of their exchange rate—
around 29 percent a year from the mid-1970s to the mid-1980s, and 22 per-
cent a year from the mid-1980s to 1999. Currency overvaluation has been
prevalent in MENA for a long time, unlike in most other regions, and espe-
cially relative to trends in the past few years.
What accounts for the significant overvaluation that has prevailed in
MENA? In my opinion, it is almost wholly the result of relying on fixed
exchange rate regimes that no longer are appropriate for responsible econom-
ic management.
Prior to the late 1980s and early 1990s, most economies in MENA opted for
a fixed exchange rate regime as the most effective strategy for combating high
inflation.

Part III: Trade, Competitiveness, and Investment 301
The adoption of fixed exchange rates, either de facto or de jure, was suc-
cessful in contributing to macroeconomic stability. However, once the imme-
diate threats of inflation running out of control had been averted, only a
handful of countries shifted to more flexible exchange rate arrangements.
From the battery of empirical investigations looking at the subject, a central
piece of evidence has emerged: Exchange rates are overwhelmingly more like-
ly to become overvalued under fixed systems than under more flexible ones.
In the analogy of the road trip, the choice of the exchange rate regime is
akin to relying on cruise control to determine your speed of travel. While tak-
ing pressure off the driver of the vehicle, the cruise control option does not
take into account weather conditions, the terrain of the road, or other vehicles
in your path. If your foot is not free to access the gas and break pedals, you
can’t change gears, limiting your maneuverability options.
And so it has been in the MENA economies, having to adjust the exchange
rate periodically as it has gotten off course. But these fixes are only temporary
and do not wholly address the problem. Moreover, these ad hoc adjustments
to the exchange rate discourage the entry of companies into the export mar-
ket, since exchange rate volatility sends confusing signals to economic agents
and raises the uncertainty of long-run investment in producing tradable
goods.
My point: Keep your foot on the gas pedal. Choose an exchange rate system
that will suffer from fewer limitations and distortions, so that economic
agents—these drivers of competitiveness—know where their true cost advan-
tages lie, and are able to make decisions based on those advantages.
The second issue on which I would like to focus is what I would compare
to driving through rough and rocky terrain on the road to competitiveness:
MENA countries have historically suffered from excessive volatility in output
and external balances.
In the 1970s and 1980s, output was more volatile in MENA than in Latin
America or in East Asia and the Pacific. And although volatility has fallen in
the 1990s, it still exceeds, by far, the world average.
There are several reasons behind the high volatility in MENA. First, coun-
tries in the region remain relatively undiversified and dependent on a few
export commodities that often experience strong fluctuations in relative
prices. Oil prices in particular have tended to cause sharp fluctuations in fis-
cal and external accounts, either through direct dependence on oil revenues or
through strong linkages between oil producers and other countries in the
region through trade, financial and labor flows, including workers’
remittances.
Moreover, several countries in the region are subject to frequent droughts,
which in turn produce sharp cuts in rural incomes and agricultural produc-
tion. Volatility is higher in agriculture than in any other sector,and agricultur-
al volatility is the main force behind output fluctuations in many non-oil-
exporting countries. Morocco, for example, experienced as many as five

Breaking the Barriers to Higher Economic Growth302
droughts in the 1990s, and booms and busts in total output were correlated
directed with these droughts.
In addition, most countries in MENA are adversely affected by the instabil-
ity of regional security, either directly, by association, or through the various
economic linkages within the region. For those areas directly affected by con-
flict, the economic costs of an interrupted and damaged economic structure
have been very high. But almost all countries have been affected by associa-
tion, through lower tourism revenues, and more generally through lower con-
fidence in their economies, bringing down private domestic investment and
foreign direct investment, and so forth.
What is the cost of volatility to competitiveness? Perhaps the biggest cost is
lower investment from domestic and external sources. An economic slow-
down and drop in private consumption trigger a response by businesses to cut
their expenses and investments to adapt to declining demand. Weaker busi-
ness activity further dampens consumer confidence, exacerbating the reces-
sion. At the same time, the effect on investment is not altogether compensat-
ed for by a commensurate economic boom the following period.
In addition to capacity constraints, volatility dampens investment because
economic agents cannot be certain that a recovery is sustainable. So business-
es tend to adapt their investment decisions more to the downswing in the busi-
ness cycle than the upswing. This dampening effect on investment, particular-
ly foreign investment, makes competitiveness for existing companies in the
region substantially more difficult, with fewer opportunities for information
flows between economic agents, and thus fewer opportunities to tap into the
technological, organizational, and managerial capabilities of other companies.
I compare the volatility in output growth and external balances to driving
on a segment of the road that is full of potholes and fallen rocks. Volatility cre-
ates a lack of confidence in the road trip altogether. And while ensuring that
the vehicle is equipped with the appropriate shock absorbers is a multistep
process, including diversifying economic structures and stabilizing the securi-
ty environment in the region, there are other steps that can be taken on the
macroeconomic policy front to mitigate the effects of volatility.
In particular, monetary and fiscal (and even exchange rate) policies are
often a transmission channel for external shocks. Weak policies amplify the
impact of volatility as it is transmitted to the economy. In some MENA coun-
tries, procyclical fiscal policies have historically acted as an important ampli-
fier of terms of trade shocks. The absence of mechanisms for intertemporal
government smoothing of revenue shocks has in turn rendered government
expenditures very volatile in the past, and although public expenditures
volatility was reduced in the 1990s compared to earlier decades, it was still
twice as high as output volatility.
There exist mechanisms that reduce both revenue uncertainty and the
transfer of volatility from revenue to expenditure levels, some of which have

Part III: Trade, Competitiveness, and Investment 303
been applied by natural resource-dependent countries. These include market-
based options such as hedging and stabilization funds. The latter have been
favored by developing countries like Venezuela (oil) and Chile (copper), and
have recently been implemented by Algeria. These types of mechanisms need
to be explored throughout the region.
More generally, my advice on volatility follows my road trip analogy. If the
journey takes you through rough terrain, then you are better off insuring the
proper mechanics are in place to handle it than to wait for an accident. And
detours and bypasses, while not ideal, can still ensure that the trip is taken and
the journey is completed.
Finally, I come to my third point about the road to competitiveness. In the
long run, that road is built on raising productivity. Period. As has been the
case since the industrial revolution at least, productivity is the fundamental
source of competitiveness.
Our own institution recently looked at productivity growth in MENA over
the past four decades—from the 1960s to the 1990s. What we found was inter-
esting. While the region experienced negative productivity growth over both
the 1970s and 1980s, there was substantial improvement in productivity
growth in the 1990s. This is the good news.
The more disappointing news is that, despite this almost universal turn-
around in productivity growth, it has failed to generate a comparable private
sector investment response. In fact, investment per worker declined from the
1980s to the 1990s in 8 out of 10 countries. As a result, there has been no
growth “pay-off” from the accumulation of productive, privately owned
capital.
What is preventing the private sector from responding? Clearly, private
investment depends on a host of factors. They encompass the range of
reforms that are needed to create an environment in which private invest-
ments in value-creating industries can thrive. I will focus here on one area in
particular. These are what I call the traffic signals on the road to competitive-
ness: that is, the systems of governance and the related matters of legal and
regulatory reform.
To begin with, there is little doubt that the current systems of governance
in MENA stifle the development of a private sector. Chief among the gover-
nance problems is the issue of state capture, where groups of influential busi-
nesses (whether public or private) exercise effective control over rules and
regulations, and utilize them to profit at the expense of the rest of the private
sector. Glaring examples of this are infrastructure and telecommunications,
but the list extends to taxes, licensing, and manipulating the loopholes with-
in the system.
The privileges extended to some elite firms prevent other entrepreneurs
from entering and competing. But the damage extends beyond competition.
The process of reform also suffers when time and resources are wasted on

Breaking the Barriers to Higher Economic Growth304
maintaining unfair advantages and restricting access to consumers. Rent-
seeking behavior in government is rewarded, reformers are discouraged, and
needed changes in governance are delayed, or worse, derailed altogether.
Related to this, there is the host of reforms that are needed in the area of
the legal and regulatory framework. In many of the MENA economies, there
are some reforms underway to revise and modernize company laws, invest-
ment codes, and customs and tax regulations.Some of these reforms are being
considered in connection with accession to the World Trade Organization or
European Union membership.
But as a whole, the business environment in MENA is plagued by burden-
some regulations. In Egypt, entrepreneurs spend about 30 percent of their
time resolving problems of regulatory compliance. In Morocco,it still takes up
to six months to register a business.
Transparency of regulations is also a problem. In a survey of businesses in
Lebanon, 62 percent of businesses admitted they had paid bribes to state
employees. And 70 percent of respondents knew in advance how much they
would have to pay to get a transaction through the bureaucracy.
To create an enabling environment for the private sector, the MENA region
needs to insure that the road to competitiveness is not full of yield signs, red
lights, and one-way streets. It needs to create the reforms that can eliminate
the gridlock of honking cars and unnecessary U-turns that currently hinder
the speed of the traveler, and hinder the rate of private sector growth.
So, these are just a few of the reforms that come to mind as being essential
for making the road more attractive to domestic and foreign private investors.
In an age of instant communication and supersonic travel, the lure of the
open road has been eclipsed by the conveniences of expediency. But econom-
ic competitiveness cannot be achieved through shortcuts or alternatives. A
successful end must have a successful beginning, and a successful middle, as
well. If this conference has established nothing else, it has confirmed the
responsibility of macroeconomic policy to provide the essential background
forces.

305
Trade, Foreign Direct Investment,
and Development in the
Middle East and North Africa
Farrukh Iqbal*
Mustapha Kamel Nabli
12
Even a casual observer of international development trends cannot fail to
notice that, in the last two decades or so, the MENA region1has lagged most
other regions of the world in both development outcomes (such as growth
and employment) and international integration (such as trade and foreign
investment). This paper argues that these two phenomena are linked, and that
the region’s weak development performance originated in part from its inabil-
ity to engage substantially with the rest of the world at a time when such
engagement was the main engine of rapid economic growth for a large num-
ber of developing countries.
The plan of the paper is as follows. In the next section, we review the devel-
opment performance of the MENA region and show that it has been trapped
in a slow growth and high unemployment situation since the early 1980s. In
the following section, we review trade and foreign investment links and show
that the region has not participated significantly in the huge growth of trade
and investment flows that have taken place in the last two decades. In the
fourth section, we connect these two sets of observations through a review of
*World Bank.
Prepared for the conference The Middle East and North Africa Region: The Challenges of Growth and
Globalization, organized by the International Monetary Fund and held in Washington, DC, on April
7-8, 2004. This paper was drawn from Trade, Investment, and Development in the Middle East and
North Africa: Engaging with the World (World Bank,2003). The main authors of the report were Dipak
Dasgupta and Mustapha Kamel Nabli. Reprinted by permission from the World Bank.

Breaking the Barriers to Higher Economic Growth306
international evidence linking growth and employment to trade openness and
investment climate. The review shows that openness and investment climate
are complementary in their effect on development performance. Indeed, a
good investment climate is a prerequisite for obtaining the full benefits of an
open trade environment. We briefly assess the investment climate in the
MENA region and conclude that, despite some improvements in the last two
decades, the region still lags in this respect. Finally, in the last section, we sim-
ulate the expected impact on growth and employment of further trade and
investment integration for the region.
MENA’s Record in Growth, Employment, and Poverty
Figure 12.1 provides a view of growth performance in the MENA region
between 1963 and 2000.2This extended perspective allows one to see both the
oil-price-supported boom of the 1970s and the slow growth trend thereafter.
For the past decade, the average growth rate of the region has been around 3.3
percent per annum. This translates into a per capita growth rate of around 1.2
percent.
Figure 12.2 provides a comparison of MENA’s growth during the 1990s
with that of other regions. The MENA region’s per capita growth rate of 1.2
percent is worse than that of such regions as Latin America, South Asia, and
East Asia. It is better only than that of Sub-Saharan Africa and the
Europe/Central Asia region. The latter spent most the decade in the throes of
Figure 12.1. GDP Growth Rates for Selected MENA Countries (1963–2000)
16
14
12
10
8
6
4
2
0
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
⫺2
year
percent
Note: Includes Algeria, Egypt, Kuwait, Oman, Morocco, Saudi Arabia, Syria, and Tunisia.

Part III: Trade, Competitiveness, and Investment 307
a wrenching transition away from state-dominated economies. In recent
years, however, several Eastern European countries have posted high growth
rates and appear set to become much more integrated into the world trading
and investment system. This observation applies in particular to such coun-
tries as Hungary, Poland, and the Czech Republic, which have been granted
accession to the European Union.
Weak growth has been reflected as well in high unemployment. While there
is a demographic aspect to this, in that the region’s fertility rates have been rel-
atively high, there is no denying that slow overall economic growth has meant
that the demand for labor, especially in the private sector, has failed to keep
pace with the large numbers of people joining the labor force. Figure 12.3
shows how serious the problem of unemployment has become in the region.
Figure 12.2. Comparative per Capita Growth Rates (1990–2000)
7
6
5
4
3
2
1
0
⫺1
⫺2
East Asia
and Pacic
Europe and
Central Asia
Latin America
and the
Caribbean
Middle East
and
North Africa
South Asia Sub -Saharan
Africa
Source: World Bank, 2003, Appendix Table 1, p. 228.
Figure 12.3. Unemployment Rates in MENA (1980–2000)
16
14
12
10
8
high-income countries
6
4
2
0
1980 1990 2000
percent of labor force
middle-income countries
MENA

Breaking the Barriers to Higher Economic Growth308
The average unemployment rate is over 14 percent currently,with some coun-
tries, such as Algeria, reaching rates as high as 29 percent.
Finally, it is instructive to consider the region’s performance in poverty
reduction. Here the region appears to have done much better than its
growth and employment situation would suggest (see table 12.1). Measured
by the US$1 per person per day criterion, the region’s poverty rate is around
2.8 percent currently. This is the lowest comparedwith other regions.
Measured by the US$2 criterion as well, MENA fares comparatively well,
although the absolute rate of poverty jumps to around 24 percent. However,
the trend over time is not comforting. Between 1990 and 2000, there was an
increase in poverty (by both criteria). If weak growth persists over the next
decade or so, the poverty rate will continue to increase, and the region will
begin to look much worse in comparison with its own past and with sever-
al other regions.
Trade and Foreign Investment Performance
Trade Performance
Overall trade. Countries in the region differ significantly in relative endow-
ments of natural resources and labor. Some are resource-poor with abundant
labor—Egypt, Jordan, Lebanon, Morocco, and Tunisia. Some are resource-
rich with abundant labor—Algeria, Iran, Syria, and Yemen (as a special low-
income case). And some are labor-importing and resource-rich—the GCC
countries. Despite the diversity of country characteristics, trade outcomes are
fairly common throughout the region. The past two decades have largely fea-
tured missed opportunities in trade integration—worse for the resource-rich
and labor-abundant countries, and somewhat better among the resource-
poor countries, with the GCC falling in between. The MENA region clearly
failed to ride the wave of globalization that began in the mid-1980s. While
world trade rose by around 8 percent in the 1990s, MENA’s trade with the
world rose by only 3 percent (figures 12.4 and 12.5).
Table 12.1. Comparative Poverty Reduction Performance (1990-2000)
$1 poverty index $2 poverty index
1990 2000 1990 2000
East Asia 29.4 14.5 68.5 48.3
Europe/Central Asia 1.4 4.2 6.8 21.3
Latin America 11 10.8 27.6 26.3
MENA 2.1 2.8 21 24.4
South Asia 41.5 31.9 86.3 77.7
Africa 47.4 49 76 76.5
Source: Global Economic Prospects, World Bank, 2004.

Part III: Trade, Competitiveness, and Investment 309
Nonoil exports. A significant portion of the trade of the MENA region is made
up of oil and oil-related products whose value fluctuates with the price of oil.
To fully appreciate the role of policy in determining exports, it is necessary to
focus on the performance of the nonoil component of exports. Figure 12.6
shows that while overall merchandise exports have fluctuated between 32 per-
cent and 17 percent of GDP, most likely depending on the price of oil at any
given time, nonoil exports have stayed at a steady and low rate of around 7
percent of GDP since the early 1980s. There is no evidence that the nonoil
exports sector has grown systematically since the 1970s to offset the relative
decline of oil exports and form a substantial new base of export earnings for
the region. Whatever policies have been introduced over this period have
clearly not succeeded in generating a higher ratio of nonoil exports.
Figure 12.4. Trade Performance of MENA Countries
140
120
100
80
60
40
20
0
labor-abundant,
resource-poor
labor-abundant,
resource-rich
MENA
labor-importing,
scarce resource-rich
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1995
1996
1997
1998
1999
1994
trade-to-GDP ratio, constant prices
Source: World Bank, 2003, p.74.
47
42
37
32
27
22
17
12
7
2
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1987
1998
1999
2000
2001
percent
non-oil exports
total merchandise exports
Figure 12.5. Exports (as Percent of GDP) Middle East and North Africa
All Countries

Breaking the Barriers to Higher Economic Growth310
Another way to measure trade performance is to compare across regions.
Figure 12.7 provides such a comparison in the case of nonoil exports. Total
nonoil exports of the MENA region amounted to about $28 billion in 2000
(excluding re-exports). For a middle-income region with nearly 300 million
people and with good resource endowments, this is a small fraction of poten-
tial. Finland, with 5 million people, has almost twice the nonoil exports of the
entire MENA region. And Hungary and the Czech Republic, with populations
Figure 12.6. Comparative Nonoil Exports, 2000
250
200
150
100
50
0
millions of US dollars
MENA10 ECA5 EASIA3 LAC4
Source: World Bank, 2003, p.42.
Note: ECA5 is the Czech Republic, Hungary, Poland, Russia, and Turkey. EASIA3 is Indonesia, Malaysia, and Thailand.
LAC4 is Bolivia, Chile, Mexico, and Brazil.
Figure 12.7. MENA Nonoil Export Potential, Conditioned on per Capita Incomes, Natural Resources, and Population,
2000
underperforming countries
3.5
3.0
2.5
2.0
1.5
1.0
.5
0
high-performing countries
actual to predicted nonoil merchandise exports,
per capita US$
Algeria
Iran
Egypt
Lebanon
Syria
MENA10
Tunisia
Jordan
Saudi Arablia
Morocco
Colombia
Argentina
Brazil
Russian Fed.
South Africa
Turkey
Guatemala
Poland
Bulgaria
El Salvador
Mauritius
Bolivia
Chile
Jamaica
Ecuador
Slovak Rep.
Czech Rep.
Costa Rica
Korea, Rep. of
Mexico
Hungary
Thailand
Philippines
Malaysia
PCNOXs = –134.3 + 0.19*GDPPPPPC – 0.54*PCNatRes – 74.7*Log(Pop)
Adj R2 = 0.60; Sample Size 42
Source: World Bank, 2003, p.43.
Note: Regression is based on 42 countries, but values for 8 low-income countries (Bangladesh, Cameroon, Côte d’Ivoire, Ghana, India, Indonesia,
Pakistan, and Yemen) are not reported because of negative values.

Part III: Trade, Competitiveness, and Investment 311
of about 10 million, each had greater nonoil exports than the region. Nonoil
exports from the MENA region are vastly smaller than those from other sub-
regions with similar populations and resource endowments. For example, a
group of five Eastern European countries—the Czech Republic, Hungary,
Poland, Russia, and Turkey, with a similar population of 270 million—had
nonoil exports of $151 billion, five times more than MENA’s. Three Southeast
Asian countries—Indonesia, Malaysia, and Thailand—had nonoil exports of
$197 billion, seven times more than MENA’s. And four Latin American
countries—Bolivia, Chile, Mexico, and Brazil—had nonoil exports of $213
billion, eight times more than MENA’s .
Nonoil exports relative to potential. The analytical basis for such a comparison
can be extended more carefully by conditioning per capita nonoil exports on
per capita incomes (as a proxy for overall skills and institutional endowments
that influence the capacity to export), and on natural resources endowments
(measured by the value of resource-based exports, mainly oil and minerals)
and population. For some 42 mostly middle-income countries (including the
MENA countries), the results suggest a strong positive association of per capi-
ta nonoil exports with per capita incomes, a negative association with natural
resource endowments and population size. That is expected, since higher skills
and institutional endowments should support higher exports, while greater
natural resource rents should appear in less intensity of effort to export non-
natural resource exports. And larger countries tend to trade less. For the MENA
countries, their nonoil exports are, on average, one-third of their predicted lev-
els. Only Jordan and Morocco had exports close to what would be predicted.
The world’s three biggest underperformers are MENA countries (Algeria, Iran,
and Egypt), and the other MENA countries are all also underperformers.
Manufactured imports relative to potential. Trade has impacts much bigger than
those captured by export performance. When firms can get imported inputs
at world prices and quality, the knowledge embodied in goods and services is
transferred from the rest of the world to the domestic economies—the pro-
ductivity and knowledge-enhancing spillover of trade. When consumers can
buy goods and services produced more efficiently in the rest of the world, they
benefit from lower prices and better quality. But the biggest benefit of trade is
that it allows countries to specialize in the production of goods and services
that rely most more intensively on their most abundant resource—for MENA,
labor—and import more of its least abundant resource—for MENA, capital
and, increasingly, knowledge-intensive goods and services. Barriers that
impede trade therefore impede potential gains in knowledge, consumer wel-
fare, and labor productivity.
The relative openness of countries to imports can be measured by compar-
ing actual imports versus a predicted level of imports conditioned on per

Breaking the Barriers to Higher Economic Growth312
capita incomes (where a higher level of incomes should be associated with a
higher level of imports) and population (to the extent that larger countries
tend to trade less). For the same group of 42 mostly middle-income countries,
per capita manufacturing imports in the MENA region were about half of
their predicted levels in 2000, confirming again that the region trades far less
than its potential (figure 12.8). Once again, Iran, Algeria, and Syria have
extremely low actual levels of imports, implying relatively closed trade
regimes. Others also fall surprisingly below predicted levels, including Egypt,
Jordan, Saudi Arabia, Morocco, and Tunisia. Only Lebanon is more open than
expected.
Trade in services. The MENA region has been losing world market shares in
exports of services. In contrast, comparators in both East Asia and Pacific, and
Eastern Europe and Central Asian regions, have more than doubled their
world market shares (figure 12.9).
Tourism, the main service export of the MENA region, has fluctuated
between 3 and 4 percent of GDP during the 1990s. Some other regions have
done better. Eastern Europe and Central Asian region (ECA4), a competing
destination for tourists from Europe, had a threefold increase in the share of
tourism receipts to GDP, overtaking all other regions. Also, while world
tourism trade has expanded fivefold in the last 20 years, MENA’s market share
has declined from 3.4 percent in 1987 to around 2.6 percent in 2000. But with
a favorable endowment of world heritage sites and a home to some of the
Figure12.8. MENA’s Import Potential, Conditioned on per Capita Income and Population, 2000
3.0
2.5
2.0
1.5
1.0
0.5
0
relatively open to imports
relatively closed to imports
PCMgM = 237.5 + 0.14*GDPPPC – 124.8*LOG(POP)
Adj R2 = 0.65; Sample Size 42
actual to predicted manufacturing imports,
per capita US$
Iran
Algeria
Syria
MENA10
Egypt
Saudi Arabia
Morocco
Tunisia
Jordan
Lebanon
Russian Fed.
Colombia
Brazil
Argentina
South Africa
Cameroon
Bulgaria
Bolivia
Guatemala
Ecuador
El Salvador
Chile
Ghana
Korea, Rep. of
Mauritius
Côte d’Ivoire
Poland
Costa Rica
Slovak Rep.
Thailand
Jamaica
Czech Rep.
Philippines
Hungary
Mexico
Malaysia
Turkey
Source: World Bank, 2003, p.45.
Note: Regression is based on 42 countries; values for 6 low-income countries (Bangladesh, China, India, Indonesia, Pakistan, and Yemen) are not
reported because of negative values.

Part III: Trade, Competitiveness, and Investment 313
world’s key religions and civilizations, the tourism potential of MENA region
remains significant. Regional conflicts may have discouraged tourists from
visiting in larger numbers, but infrastructure and marketing efforts have
proved successful in some countries, such as Jordan.
Foreign Direct Investment (FDI) Performance
The MENA region is not well connected with global investment and produc-
tion chains. This is evident from the limited role played in the region’s
economies by foreign direct investment and by trade in parts and compo-
nents. Integration with global private capital flow markets has also been rela-
tively stagnant, in sharp contrast to comparable country groups. Net FDI
inflows to MENA (measured as a share of PPP GDP) were consistently less
than half a percentage point of GDP for most of the period (figure 12.10). In
the early 1980s this put MENA roughly on par with comparable groups. But
in the next 15 years, the average of the other comparators had risen to between
1.0 to 2.5 percent, while MENA continued to trail at around 0.5 percent.
The MENA region, excluding the Gulf countries, received net inflows of
FDI of about $2.2 billion in 2000—slightly more than 1 percent of the $158
billion to all developing countries, and one-sixth of their share (7 percent) in
the GDP of all developing countries. The group of five Eastern European
countries (Czech Republic, Hungary, Poland, Turkey, and Russia) together
received some $19 billion, nine times more than MENA. The three East Asian
countries (Malaysia, Philippines, and Thailand) received more than US$8 bil-
lion in inflows, four times more than MENA. And the group of four Latin
American countries (Bolivia, Chile, Mexico, and Brazil) received about US$50
billion, more than 22 times the inflows to the MENA region. These compar-
Figure12. 9. MENA’s Share of World Exports of Services
1.4
1.2
1.0
0.8
0.6
0.4
0.2
ECA3
ASE5
LAC4
MENA
0
percent
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
Source: World Bank, 2003, p.85
Note: ECA 4 is Czech Republic, Turkey, Poland, and Hungary. EA5 is China, Indonesia, Republic of Korea, Malaysia, and Thailand. LAC4 is Argentina,
Brazil, Chile, and Mexico.

Breaking the Barriers to Higher Economic Growth314
isons provide some indication of the huge potential for expanding inflows of
FDI to the MENA region. A large part of these inflows came from (neighbor-
ing) high-income Europe. Egypt accounted for about half the MENA total
(US$1.2 billion), and Tunisia and Jordan about a quarter each (US$750 mil-
lion and US$560 million, respectively). The rest received very small amounts
or even had significant outflows (Yemen).
Potential for foreign direct investment. The potential for higher inflows by coun-
try can be determined by conditioning FDI inflows on nonoil trade perform-
ance (measured by nonoil export to PPP GDP ratios), natural resources, and
population. FDI inflows are known to be closely related to trade flows, so the
predicted levels of FDI should be associated with trade. Natural resource
endowments often also lead to higher levels of foreign investment, albeit of
different types of flows. And size may matter, with larger countries expected
to receive higher investment inflows (market-seeking investments); but
because this is already accounted for by measuring FDI inflows relative to
GDP, the residual population variable may or may not be significant. The
results for 42 countries are largely in accord with expectations: trade and nat-
ural resources raise FDI flows, but size turns out to be a negative influence.
How do the actual inflows of FDI to MENA countries look in relation to
expected inflows, once nonoil trade, natural resources, and population are
factored in? First, only Jordan, Lebanon, and Tunisia (small resource-poor
countries) do as well or slightly less well than expected (figure 12.11).
Morocco is a surprise in how little it receives in FDI inflows (at least for
1998–2000); Egypt receives more, possibly given its larger size, but still well
below expected levels. Second, among the countries with large natural
resource endowments, Saudi Arabia is the only country that receives relative-
ly high FDI, led by its oil sector. All the larger oil-producing countries
Figure 12.10. Net FDI Flows to MENA and Other Regions
3.0
2.5
2.0
1.5
1.0
0.5
0
LAC4
ECA4
ASE5
MENA
percentage of PPP GDP
1981–85 1986–90 1991–95 1996–2000
Source: World Bank 2003, p.80.
Note: ECA 4 is Czech Republic, Turkey, Poland, and Hungary. EA5 is China, Indonesia, Republic of Korea, Malaysia,
and Thailand. LAC4 is Argentina, Brazil, Chile, and Mexico.

Part III: Trade, Competitiveness, and Investment 315
(Algeria, Iran, Syria, Yemen) receive very low FDI relative to expected levels.
Third, for the MENA countries as a whole, FDI inflows are only about half
what they should be. In comparison, Chile receives three times as much FDI
as expected, and the Czech Republic twice as much. Others, such as China,
Brazil, and Argentina,receive FDI inflows four to five times their expected lev-
els. The FDI gap for MENA also shows investment climate barriers in these
countries, since the gap in potential already takes into account their very low
nonoil trade. If their nonoil trade were higher and if the investment climate
were better, the region could expect FDI inflows at least four to five times the
current level—or some 3 percent of GDP on average, compared with the cur-
rent 0.5 percent of GDP.
MENA’s Participation in Global Production Chains
The region as a whole has a very large negative trade balance with OECD
countries in trade in parts and components (US$14 billion), with imports
accounting for about 15 times the value of exports to these countries. In con-
trast, China and Malaysia have exports of parts and components to the OECD
that are twice as large as their imports. The MENA region thus participates
very little in global production sharing, exporting primarily low-value fin-
ished goods, and importing parts and components for an inefficient manufac-
turing base—typical of inward-looking, import-substitution economies
(table 12.2).
Figure 12.11. Foreign Direct Investment Potential, Conditioned on Openness, Natural Resources, and Population, 2000
5
4
3
2
1
0
high performers
underachievers
FDIPC = 76 + 0.07*NOXPC + 0.04*NRXPC – 10.9*LOG(POP)
Adj R2 = 0.27; Sample Size 42
actual to predicted net FDI, per capita US$
Algeria
Morocco
Iran
Syria
Egypt
MENA10
Tunisia
Jordan
Lebanon
Saudi Arabia
Cameroon
Bangladesh
Ghana
Pakistan
Turk ey
Côte d’Ivoire
Malaysia
South Africa
Russian Fed.
Philippines
Guatemala
Mauritius
Korea, Rep. of
India
Ecuador
Hungary
Costa Rica
Slovak Rep.
Thailand
El Salvador
Colombia
Mexico
Bulgaria
Jamaica
Bolivia
Czech Rep.
Poland
Chile
China
IBrazil
Argentina
Source: World Bank, 2003, p.49.
Note: Regression is based on 42 countries; values for 2 low-income countries (Indonesia and Yemen) are not reported because of negative values.

Breaking the Barriers to Higher Economic Growth316
Trade Policy, Investment Climate, and Growth Performance
Trade Protection
Of the many factors that affect trade outcomes, price-related ones are usually
among the most important. The prices of tradable goods and services are
strongly affected by tariff levels and non-tariff barriers, as well as by real effec-
tive exchange rates, which are themselves influenced by macroeconomic poli-
cies and conditions. In general, trade protection is high for the developing
countries in MENA relative to their income levels. Comparing over regions
Table 12.2. The Relative Importance of Parts and Components in MENA Countries
Value of parts and components Share of parts and
in OECD trade components in all
($ million) manufacture (%)
Exports Imports Exports Imports
Country 1988 2000 1988 2000 1988 2000 1988 2000
Algeria 6 7 943 994 2.6 2.8 23.5 18.7
Bahrain 39 24 156 208 19.6 6.9 18.1 16.5
Egypt 19 54 1,279 2,142 4.3 3.1 23.5 24.7
Iran 8 18 941 1,311 1.7 2.2 21.1 25.6
Jordan 25 21 301 276 20.2 8.5 21.0 18.4
Kuwait 42 8 437 505 38.0 3.9 15.4 17.0
Lebanon 3 9 80 242 1.5 3.5 7.6 11.8
Libya 10 6 560 369 3.8 1.6 16.3 21.3
Morocco 33 105 421 1,231 2.1 2.5 14.4 19.2
Oman 22 45 208 330 8.3 14.0 15.2 18.8
Qatar 6 17 115 300 11.1 3.7 16.6 20.4
Saudi Arabia 73 213 1,941 3,298 3.6 7.1 13.5 19.0
Syria 1 9 148 338 4.5 2.7 16.2 21.5
Tunisia 72 332 330 859 5.5 7.4 15.8 14.4
United Arab Emirates 31 208 714 2,919 12.2 10.9 16.7 20.8
Yemen Rep. 2 1 61 105 47.0 6.7 15.5 19.6
MENA 391 1,078 8,635 15,430 5.2 5.7 17.1 19.9
Labor-abundant,
resource-poor 151 521 2,411 4,751 4.2 4.7 18.6 19.3
Labor-abundant,
resource-rich 17 35 2,094 2,748 2.4 2.5 21.4 21.9
Labor-importing,
resource-rich 171 508 3,133 7,057 6.1 8.4 14.5 19.7
Memo Items:
China 379 25,409 2,237 16,091 1.9 11.9 11.1 25.6
Japan 34,212 51,583 5,373 17,268 20.4 20.3 13.5 21.1
Republic of Korea 2,841 15,720 5,077 11,128 6.7 18.3 19.4 17.6
Malaysia 452 11,526 1,241 6,959 7.7 22.5 18.8 23.1
Singapore 4,425 9,584 4,492 10,877 32.8 23.4 25.1 22.9
Taiwan, China 5,231 21,369 4,265 11,208 11.5 24.3 17.2 17.1
Source: World Bank, 2003, p. 81.

Part III: Trade, Competitiveness, and Investment 317
and over time, we find that MENA trade barriers have been the slowest to
come down, and there have been episodes as well of reversals in policy during
the 1990s. Exchange rate misalignments, in several countries and over signif-
icant periods, have also affected trade performance.
High protection. Trade protection can be shown by a variety of measures
reflecting tariff rates and the tariff equivalents of non-tariff barriers. Table
12.3 presents data for MENA and several comparator groups, using several
different measures whose derivations are described in the notes to the table.
All the measures show a similar ranking. Key results include:
• Average nontariff barrier protection is higher in MENA than in other
lower-middle-income countries; indeed, it is higher than in all other
regions of the world except Latin America.
• Average MENA trade protection, measured in terms of Anderson-Neary Ideal
Measure, is one-quarter above the comparable average for lower middle-
income countries; it is also higher than in all other comparator groups.
Table 12.3. Trade Protection Indicators for MENA
(Most recent year)
Anderson-
Neary
Simple Weighted Standard NTB Ideal
average average deviation coverage Measure
Algeria 22.4 15.0 14.3 15.8 20.0
Bahrain 8.8 .. .. .. ..
Egypt 20.5 13.8 39.5 28.8 19.0
Iran 4.9 3.1 4.2 .. ..
Jordan 16.2 13.5 15.6 0.0 ..
Lebanon 8.3 12.0 11.2 .. 22.2
Morocco 32.6 25.4 20.5 5.5 25.9
Oman 4.7 4.5 1.2 13.1 4.4
Saudi Arabia 12.3 10.5 3.1 15.6 11.0
Syria 21.0 .. .. .. ..
Tunisia 30.1 26.3 12.6 32.8 19.2
MENA 16.5 13.8 13.6 15.9 17.4
ECA4 12.9 7.2 18.3 12.4 3.5
LAC4 12.2 12.9 6.9 48.4 13.9
EA5 11.3 8.3 17.9 13.5 6.0
LMIC 15.3 12.5 15.0 13.4 13.2
Source: World Bank, 2003, p.103.
Notes: Tariff rates used are most favored nation tariff rates. Nontariff barrier coverage refers to the number of tariff
lines that have at least one nontariff barrier. The Anderson and Neary Ideal Protection Measure is the uniform tariff
rate that must be applied to the free-trade regime, as a compensating variation to return welfare to most recent
year of observation. The comparators are ECA4 (Czech Republic, Turkey, Poland, and Hungary), LAC4 (Argentina,
Brazil, Chile, and Mexico) and EA5 (China, Indonesia, Thailand, Malaysia, and Republic of Korea). LMIC refers to lower-
middle-income countries (with GNI per capita in the range $746-$2975). The comparator numbers are simple aver-
ages of the data for the respective countries they represent.

Breaking the Barriers to Higher Economic Growth318
• Tunisia and Morocco have among the highest protection rates in MENA
(and the world) despite several reform episodes since the early 1980s, while
Egypt has the highest dispersion of tariff rates.
• On the positive side, Oman and Saudi Arabia, two oil-rich countries, have
lower protection than the average for upper-middle-income comparators.
With respect to nontariff barriers, the typical experience in recent years is
one of improvement. In Tunisia, extensive quantitative restrictions (affecting
some 90 percent of domestic output) have progressively been reduced (tex-
tiles, passenger cars, agricultural products) in the 1990s. In Morocco, most
quantitative restrictions have been eliminated. In Algeria, quantitative restric-
tions have been reduced, although temporary reversals occur and prior
authorization lists still exist for some items. In Jordan,removal of quantitative
restrictions was the main item of trade liberalization in 1988 and reduced cov-
erage from 40 percent to 7 percent of production; most remaining quantita-
tive restrictions have been eliminated since 1995. In Egypt, import licensing
was eliminated in 1993 and the scope of quantitative restrictions progressive-
ly reduced. In Lebanon, some import licensing and multiple authorizations
remain. Syria has several lists of goods with import eligibility requirements
(public sector, private sector, and two negative or banned lists) with a general
licensing requirement for all imports.
Exchange Rate Management
During the past three decades, many MENA countries experienced substan-
tial overvaluation of their real exchange rate—around 29 percent a year from
the mid-1970s to the mid-1980s, and 22 percent a year from the mid-1980s to
1999 (table 12.4). In general, the extent of overvaluation did not seem to have
significantly decreased during the 1990s—contrary to the experience of Latin
American, African, or Asian economies.
For the MENA region as a whole, exchange rate policy explains losses in com-
petitiveness and in manufactured exports. Real exchange rate overvaluation has
reduced, on average, the ratio of manufactured exports to GDP by 18 percent a
year. Manufactured exports, which averaged 4.4 percent of GDP from 1970 to
1999, could have reached 5.2 percent of GDP if no overvaluation had taken place.
These losses were more concentrated in the 1970s and 1980s than in the 1990s,
due to the higher overvaluation of the currencies during those two subperiods.
Finally, a significant reason for the persistent misalignment and over-
valuation of exchange rates is the prevalence of pegged or fixed exchange rate
practices. Most MENA countries have had de facto or formal pegged nominal
exchange rates, with only some recent changes towards floating exchange rates
(as in Egypt in 2003). This is related to a general fear of floating. While there
may have been gains in terms of reduced inflation, the tradeoff clearly has
been to hurt GDP growth and exports.

Part III: Trade, Competitiveness, and Investment 319
The Critical Role of the Investment Climate
Trade barriers and overvalued exchange rates impede growth, in part by
restricting domestic expansion into potential export areas, and in part by con-
straining the inward flow of knowledge embodied in new products and serv-
ices. However,they are not the only such impediments. The process of growth
via trade expansion involves a phase of investment supply response that is
influenced by a broad range of considerations that are sometimes grouped
together under the label “investment climate.” They include the transaction
costs encountered in trade-related business activities such as clearing goods
from customs and shipping goods overseas, the expenses involved in routine
business operations for telecommunications and electricity,and the burden of
regulations involved in starting a business, hiring and firing labor, and closing
down a business. They can also refer to aspects of governance, such as corrup-
tion, the strength of property rights, and adherence to the rule of law.3
Recent analyses suggest that openness to trade has limited or no impact on
growth in economies with excessive regulations. Using indices of regulations
derived from employment laws, industrial relations laws, and business regis-
tration and entry procedures, Bolaky and Freund (2003) show that the effect
of trade on growth is more than halved in heavily-regulated economies. They
conclude that the potentially beneficial effects of trade liberalization can be
lost if labor laws prevent the sort of labor re-allocation to more productive
sectors and firms that is necessary for productivity enhancement and growth
to occur. Similarly, if it is very difficult to register new businesses, one would
expect less investment response to new opportunities provided by trade
liberalization.
Table 12.4. Average Exchange Rate Misalignment and Volatility
1975–80/84 Misalignment
(depending on country) (percent per year) Volatility
MENA 29 7.9
Latin America 20 11.2
Africa (CFA) 61 12.7
Africa (non-CFA) 29 11.3
South Asia 43 13
South East Asia 10 5.4
1985–99 Misalignment Volatility
MENA 22 12.4
Latin America 10 12.9
Africa (CFA) 28 14.5
Africa (non CFA) 13 16
South Asia 15 8.3
Southeast Asia 5 8.6
Source:Nabli and Véganzonès-Varoudakis, 2002.

Breaking the Barriers to Higher Economic Growth320
A similar result, with somewhat broader implications, is found by
Dasgupta and others (2003). They show that trade expansion adds significant-
ly to employment in industry (a proxy for good jobs) only in countries where
there are large foreign investment inflows (a proxy for investment climate).
Thus, Figure 12.12 shows a higher elasticity of employment to trade openness
among countries with relatively large investment inflows (left panel) but zero
elasticity among countries with low inflows (right panel). These results may
be interpreted to mean that countries with poor investment climates are
unlikely to benefit significantly from trade openness.
The Investment Climate in MENA
International comparisons suggest that the MENA region lags most other
regions in regard to investment climate considerations. The region tends to
have relatively high transaction costs for starting, operating, and closing busi-
nesses. The following list provides a summary account of some relevant
issues—for which more detailed information can be found in World Bank
(2003).
Starting businesses. The cost of legally registering a new business is influenced
by both the number of steps required and the number of days required to ful-
fill necessary procedures. When measured as a ratio of per capita income, this
cost is higher in MENA than in most other regions.
Contract enforcement. When ranked along a procedural complexity index, the
MENA region has the second highest score among seven regions for which the
relevant data have been compiled (see World Bank, 2004). This index covers
Figure 12.12. Interaction between Trade Openness and Foreign Investment
1.3 0.4
0.3
0.2
0.1
0
0 5 10 15 20 25 30 35 40 45
⫺0.1
⫺0.2
⫺0.3
⫺0.4
⫺0.5
1.0
0.8
0.6
0.4
0.2
00 10203040506070
developing countries
large FDI recipients
developing countries
low FDI recipients
R2 = 0.4639
R2 = 0.0044
non-oil merchandise exports, in percent of GDP non-oil merchandise exports, in percent of GDP
Employment in industry after controlling for factors
other than non-oil merchandise exports (as a ratio of
total working-age population, in logarithm)
Employment in industry after controlling for factors
other than non-oil merchandise exports (as a ratio of
total working-age population, in logarithm)
Source: Dasgupta and others (2003).

Part III: Trade, Competitiveness, and Investment 321
both the number of procedures required in contract enforcement actions and
the number of days. Difficulties and delays in contract enforcement are often
cited by foreign investors as a significant impediment to investment.
Air transport. Despite some progress, the regulatory reform of air transport in
MENA is still at an early stage. Services remain largely provided by state-owned
airlines and airports, with competition restrained by restrictive licensing
regimes for domestic flights and international air service agreements. Efficiency
indicators, such as freight and passenger capacity use, suggest that MENA car-
riers fall short of international standards by an estimated 15 to 20 percent.
Telecommunications. Many developing countries have liberalized their
telecommunications sectors in the last two decades, with Latin America in the
lead, closely followed by South Asia and East Asia and the Pacific. But regula-
tory reform in telecommunications has been slow in MENA, where markets
remain on average less competitive than in other developing countries. Lower-
cost telecommunications could help improve the weak position of MENA
countries in global production networks, boosting the region’s participation
in global trade. Evidence from developing countries over the 1990s suggests
that the share of manufactured exports in GDP increases with improvements
in the overall quality of telecommunications.
Power. In MENA the power sector is still dominated by vertically integrated
public monopolies with poor financial health, low operating efficiency, and
extensive government interference. Regulatory frameworks are not character-
ized by independence, transparency, and accountability. Regulatory and oper-
ational functions are not sufficiently separated to assure potential private
investors and new entrants that the future policy developments in the sector
will be fair and competitively neutral. Until the end of the 1990s MENA’s
reform scorecard in the industry was very low, ahead only of Africa.4
Financial Services. Most MENA countries embarked on financial sector
reform only in the 1990s, almost two decades after East Asia and Latin
America. Banking systems were state-dominated and excessively regulated,
and remain so in Algeria, Iran, and Syria, where state-owned banks still
account for more than 95 percent of domestic bank assets. In some cases,
state-owned banks have extended soft loans to loss-making public enterpris-
es, creating contingent liabilities for the public sector and credit bottlenecks to
private sector investment. Foreign bank presence remains limited.
Conflict and the Investment Climate
The investment climate of a country, or region, for that matter, is clearly
affected not just by the domestic policy context discussed in the preceding sec-

Breaking the Barriers to Higher Economic Growth322
tions but also by what one might call the geopolitical context. This is reflect-
ed most directly in the incidence of conflicts, wars, and politically inspired
trade and investment barriers such as sanctions. The geopolitical context is
likely to have been very important in shaping the manner and extent to which
the MENA region has been able to participate in world trade and investment.
Persistent conflict reduces the scope for trade and investment and thus
lowers growth. In the MENA region, frequent conflicts and militarization are
related to the absence of a lasting peace settlement in the region, but also to
several interstate and intrastate conflicts.5These events raise the risk percep-
tions and willingness to invest in the region among both domestic and foreign
sources. A recent comparative study of conflict-affected developing countries
and a control group of countries not affected (Gupta and others 2002) sug-
gests that the numerous effects include slower GDP growth, falling trade,
sharply reduced tourism, macroeconomic instability, large security expendi-
tures, higher fiscal deficits and inflation, and the crowding out of education
and health spending. The compounded effects appear to be a sharp slowdown
in GDP growth in the preconflict stage (–1% GDP growth change) and an
even sharper decline during conflict (–2%). Given the extent and duration of
conflict in the MENA region, the effects are possibly even greater. Other
research suggests strong contemporaneous negative spillover effects of con-
flict on neighbors—effects as dramatic as those of the conflict within a coun-
try, with the contiguity or nearness (length of common border) a key factor.
These appear to work through general contagion as much as through other
channels (such as trade or labor migration or capital flows). When conflict
does end, there is a sharp rebound in economic activity (+3 percent GDP
change)—signaling the positive effects that ending persistent conflict might
have in the MENA region.
Potential Gains from Trade and Investment Climate Reforms
We have seen in earlier sections that the MENA region exhibits trade and
investment integration ratios that are substantially below the potential levels
estimated on the basis of their income and population levels. We have also
seen that growth and employment can rise significantly if the investment cli-
mate is improved and more trade is enabled. In this section, we simulate the
effect of the region’s moving from its current level of trade and investment
integration to a higher level.
We start with the assumption that the region achieves about half its trade
and investment potential over 10 years. This would raise the nonoil export
ratio of the region from about 6 percent of GDP to about 13 percent by the
end of 10 years—reaching roughly half the average ratio of nonoil exports to
GDP of all developing countries today. The nonoil export growth rate under-
lying this scenario is about 15 percent a year, nearly twice the expected world

Part III: Trade, Competitiveness, and Investment 323
export growth, but the increase in MENA’s share of that market remains very
small because of the very low starting levels. Merchandise imports, now 20
percent of GDP, would also be expected to rise by about 7 percentage points
over the decade (with incremental financing of additional imports from high-
er nonoil exports and higher FDI).
The effect would be to raise the merchandise trade ratio from about 46 per-
cent of GDP (2003) to about 58 percent over a decade (2013) and nonoil mer-
chandise trade from about 26 percent of GDP to about 39 percent. The 50
percen gain in the nonoil merchandise trade ratio over the decade would be a
substantial gain—about half of the average (100 percent) increase for fast-
integrating developing countries in the 1990s. FDI inflows are corresponding-
ly assumed to rise by some 2.5 percent of GDP, and private productive invest-
ment rates (inclusive of FDI) by some 7 percent of GDP (from the current
12.5 percent of GDP to about 20 percent of GDP).
Impact on growth. These broad assumptions are useful for calibrating the like-
ly impacts on growth and productivity with an aggregate source of growth
model (table 12.5). The MENA region has three basic sources of growth other
than exploiting oil (or other non-reproducible assets) more intensively: deep-
ening capital investment per laborer; deepening human capital per laborer;
and increasing total factor productivity. If policies are reasonably successful in
achieving the trade and investment outcomes described above, and a reason-
ably modest turnaround in productivity growth of about 1.4 percent a year is
achieved as a result of greater trade, the net effect would be to raise annual
growth from about 1.4 percent per capita in the 1990s, to about 4 percent per
capita in the coming decade. This is consistent with the experience of other
countries in the world, where correspondingly faster growth was achieved
with greater trade and investment orientation (Dollar and Kraay, 2001).
Table 12.5. Potential for Faster GDP Growth, Accumulation, and Productivity from Trade and Investment Climate
Reforms in MENA, 2003–13
Percentage per annum per laborer, except when noted)
Growth of GDP
Physical Human Total factor per labor (labor Growth of GDP
capital capital productivity force weighted) per capita
1990s 2003–13 1990s 2003–2013 1990s 2003–13 1990s 2000–13 1990–00 2003–13
MENA ⫺0.3 2.4 1.2 1.2 ⫺0.2 1.4 0.7 3.0 1.4 4.3
East Asia 8.4 .. 0.7 .. 3.2 .. 7.0 .. 7.1 ..
South Asia 3.3 .. 0.9 .. 0.9 .. 2.7 .. 3.2 ..
OECD 2.0 .. 0.6 .. 0.4 .. 1.6 .. 2.1 ..
World 4.8 .. 0.7 .. 1.6 .. 3.9 .. 4.7 ..
Source: World Bank, 2003, p.52.
Note: All growth estimates by regions and world are labor force weighted. GDP per capita growth rates differ from GDP per-labor growth rates
because of differences between labor force and population growth rates. “..” means not projected.

Breaking the Barriers to Higher Economic Growth324
The 1990s were marked by a fall in capital per laborer in MENA, primari-
ly because of low and falling rates of growth in private investment relative to
the growth of the labor force. With gains in trade intensification and major
improvements in the investment climate, faster growth in private domestic
investment and foreign direct investment can be expected to lead to a signifi-
cant improvement in capital deepening. Given a fast-expanding stock of labor
force (of about 3.4 percent a year), achieving such gains will be difficult, but
critical. Historically, capital per laborer increased in the region—by as much
as 7.9 percent per annum in the 1970s (driven by public investment)—but
slowed to 2.1 percent per annum in the 1980s and then collapsed to negative
levels in the 1990s. Therefore, business climate reforms to elicit a more robust
private investment response are likely to be central to the success of trade
reforms in the MENA region. Moreover, the shift from public to private
investment will be critical.
Improving total factor productivity (TFP) is also a key driver of growth. It
is reasonable to hypothesize that the total factor productivity gains from
opening trade and investment would be significant. Opening trade would
improve productivity by encouraging shifts in resources to more productive
and internationally competitive activities (the stock effect), by improving
access to higher quality inputs, and by spillovers from FDI and greater com-
petition. World Bank staff estimates suggest, for example, that fast-integrating
countries obtained an additional 1 percent of GDP from productivity gains
over the 1990s compared to only 0.6 percent among slow-integrating coun-
tries (World Bank, 2003, p.53). For the next decade, therefore, it is reasonable
to assume that TFP gain of 1.4 percent per laborer will be achieved in the
MENA region, reversing the negative TFP growth of the previous two decades,
as the region’s economies open up to trade and private investment.
Impact on employment. The employment effects of GDP growth at about 6
percent annual rate (resulting from faster trade integration and improvement
in investment climate) are likely to be substantial. Faster GDP growth will
itself generate equivalently faster growth in employment. However, assumed
technological progress and productivity growth will mean that the gains will
be distributed more widely, and that employment growth will therefore be at
a slower pace. Moreover, likely real wage increases will also dampen employ-
ment growth as firms respond to higher wages. Offsetting these pressures will
be greater labor-intensity of production driven by greater trade orientation of
the production structure. The net effects of these factors is employment
growth of up to 4.5 percent a year. This would be adequate to absorb the new
entrants to the labor force and cut unemployment rates by half over the next
decade.6
Exporting more merchandise goods and manufactures will create new
jobs.Just over the next five years, if the MENA region can achieve15 percent

Part III: Trade, Competitiveness, and Investment 325
real annual nonoil export growth for the region with improved policies, it
would boost the region’s nonoil exports to about US$60 billion from the
present level of US$28 billion (excluding re-exports). This could directly gen-
erate some 2 million additional jobs in such nonoil export activities, and
another 2 million indirectly, as domestic goods and services supply inputs to
these activities, and from the multiplier effects on domestic final demand. So
closing even a small part of the export gap could generate some 4 million
jobs over five years from direct trade effects alone. This simplified projection
abstracts from a much more complex set of factors, but the magnitude is
indicative of the potential.
Examples abound of the effects of expanding trade on jobs. For Mexico,
thanks to NAFTA and radical economic reforms, trade more than tripled,
from $82 billion in 1990 to about $280 billion in 1999, making it the sev-
enth largest trading nation in the world and resulting in a rapid pace of job
creation in manufacturing. From 1994 to 1999, manufacturing employ-
ment grew at almost 16 percent per annum, at almost twice the pace
recorded in the pre-NAFTA years of 1985-93. Job creation has also been
strong in export-processing zones in Mauritius, the Dominican Republic,
and El Salvador (Rama 2001). Indonesia is another country that saw a
major trade and investment policy reform produce employment benefits in
the mid-1980s. Manufactured exports and FDI boomed, as did manufac-
turing employment. Indonesia is especially relevant to some MENA coun-
tries in that it was a resource-dependent economy until the mid-1980s,
when it found a new engine of growth in manufactured exports (Iqbal
2002). Within the MENA region, Tunisian exports of textiles and clothing
have boomed in “offshore companies” that supply foreign markets, and
employment in these industries increased steeply. Morocco is another
example of a country that gained significantly from an initial burst of eco-
nomic reforms in the early 1980s. Manufacturing sector employment and
exports rose sharply in the early 1980s in response to a series of trade and
investment liberalization measures (World Bank, 2003, p. 59). But the
Moroccan boom faltered in the 1990s, as the impact of the initial reforms
package dissipated and macroeconomic policies allowed the real effective
exchange rate to appreciate, hurting both exports and employment in
manufacturing.
Conclusions
The MENA region has been traveling along a slow growth and high unem-
ployment trajectory over the past two decades. To pull out of this path, the
region needs to make three shifts in its sources of growth: from oil to nonoil
sectors, from public state-dominated to private market-oriented investment,
and from protected import-substitution to export-oriented activities.
Intensifying trade and private investment is at the core of all three shifts.

Breaking the Barriers to Higher Economic Growth326
The region has good potential for expanding trade. Exports other than oil
are a third of what they could be, given the characteristics of the region.
Openness to manufacturing imports is half of what would be expected.
Increasing its share of world trade is possible for at least three reasons. First,
the present level of the region’s presence in world markets is small; an increase
is unlikely to be resisted among major trading partners. Second, competitive-
ness based on low wages is possible, since wages in the region are fairly low, in
the bottom half of world wages. (However, for competitiveness to be sustain-
able, overall trade policy and investment climate considerations will have to
be improved). And third, the region is close to a high-income region: across
the Mediterranean is the EU, which can potentially be a source of high
demand for certain regional products.
MENA countries also have great potential for attracting more investment
from abroad and encouraging more private investment at home, both crucial
for trade and development. If exports other than oil were higher, and in a bet-
ter investment climate, domestic private investment in traded goods and serv-
ices would be much higher. And the FDI inflows that the region could expect
would be four to five times what they are today—some 3 percent of GDP, up
from an average of 0.6 percent. On the other hand, the region remains subject
to a geopolitical context featuring high levels of tension and conflict, which
can discourage foreign investment and even trade.
Nevertheless, if only half the region’s trade and private investment poten-
tial were realized over the next ten years, that would be enough to raise its per
capita GDP growth from about 1 percent to about 4 percent a year—half from
more private investment and half from the greater productivity that openness
would encourage.
Expanding trade also holds the promise of substantial dividends in job cre-
ation, because export opportunities would add millions of jobs. For example,
if the region can achieve faster nonoil export growth of about 15 percent a
year over the next 5 years, it would probably be sufficient to generate some 4
million jobs or 4 percent of the labor force, directly and indirectly, in the
export sectors alone. The employment effects are however, conditional on
there being a more favorable investment climate in the region.
Notes
1. The term “MENA region” refers in this paper to Algeria, Djibouti, Egypt, Iran, Jordan, Lebanon,
Morocco, Syria, Tunisia, Yemen, and the six GCC countries, namely, Bahrain, Kuwait, Oman, Qatar,
Saudi Arabia, and the United Arab Emirates. In cases where data are available only for a subset of these
countries, the relevant countries are identified in table or chart notes.
2. A consistent series starting in the 1960s could be obtained only for the eight countries noted in the
figure. However, this set includes most of the bigger economies of the region, with the exception of
Iran and Iraq. From what is generally known of their economic performance over this period, it is
unlikely that their omission affects the conclusions drawn here.

Part III: Trade, Competitiveness, and Investment 327
3. A vigorous debate has arisen around the question of the direction of causality in the link between
trade openness, institutional quality, and growth (see Frankel and Romer,2002, Rodrik, Subramaniam
and Trebbi, 2002 and Dollar and Kraay, 2003).
4. The ranking is based on the following six criteria: Is the existing electric utility functioning on com-
mercial principles? Have laws been passed which would permit unbundling and/or privatization in
part or in all of the industry? Is a regulatory body in place that is separate from the utility and min-
istry? Is there any private sector investment in the industry? Has the core state-owned utility been
restructured? Have any of the existing state-owned enterprises been privatized?
5. The MENA region has seen a large number of conflicts in recent decades, including some 14 years
of civil conflict affecting 8 major countries, and some 15 years of cross-border regional and interna-
tional conflicts affecting 14 countries. Between 1945 and 1999, it had the second highest number of
violent conflicts, compared with other regions (World Bank, 2003, p.91).
6. Note that this is consistent with an implied employment elasticity of about 0.7 with respect to GDP,
which is also fairly uniformly noted in other regions such as Latin America and East Asia.
References
Bolaky, B., and C. Freund. 2003. “Trade, Regulations and Growth.” Unpublished paper. World Bank,
Washington DC.
Dasgupta, D., J. Keller, and T.G. Srinivasan. 2002. “Reforms and Elusive Growth in the Middle East—
What Has Happened in the 1990s?” MENA Working Paper 25. World Bank, Washington, DC.
Dasgupta, D., M. Nabli, C. Pissarides, and A. Varoudakis. 2003. “Making Trade Work for Jobs:
International Evidence and Lessons for MENA.” MENA Working Paper 32. World Bank,
Washington, DC.
Dollar, D., and Kraay, A. 2001. “Trade, Growth and Poverty.” World Bank Policy Research Working
Paper 2615. World Bank, Washington, D.C.
———. 2003. “Institutions, Trade and Growth.” World Bank Policy Research Working Paper 3004.
World Bank, Washington, D.C.
Frankel, J., and D. Romer. 2002. “Does Trade Cause Growth?” American Economic Review 89(3): 379-
99.
Gupta, S., B. J. Clements, R. Bhattacharya, and S. Chakravarti. 2002. “Fiscal Consequences of Armed
Conflict and Terrorism in Law and Middle Income Countries.” Working Paper 02/142.
International Monetary Fund, Washington DC.
Iqbal, F. 2002. “Deregulation and Development in Indonesia: An Introductory Overview.” In F. Iqbal
and W.E. James, eds., Deregulation and Development in Indonesia. Westport, Connecticut: Praeger
Publishers.
Nabli, M., and M.Veganzones-Varoudakis. 2002. “Exchange Rate Regime and Competitiveness of
Manufactured Exports: The Case of MENA Countries.” MENA Working Paper 27. World Bank,
Washington, DC.
Rama, M. 2001. “Globalization, Inequality and Labor Market Policies.” Paper for Annual Bank
Conference in Development Economics (ABCDE-Europe), Washington.
Rodrik, D.,A. Subramaniam, and F. Trebbi. 2002. “Institutions Rule: The Primacy of Institutions over
Geography and Integration in Economic Development.” Unpublished. International Monetary
Fund, Washington, DC.
World Bank. 2003. Trade, Investment and Development in the Middle East and North Africa: Engaging
with the World. Washington, DC: World Bank.
———. 2004. Doing Business in 2004: Understanding Regulation. Washington, DC: World Bank.

329
Making Trade Work for Jobs
International Evidence and Lessons
for MENA
Dipak Dasgupta*
Mustapha Kamel Nabli
Christopher Pissarides†
Aristomene Varoudakis††
13
Can trade expansion help MENA countries step up the pace of job creation?
Despite the short-term costs of adjustment to trade liberalization, in a num-
ber of countries that have successfully integrated into global markets, export-
led growth has eventually brought large employment dividends. The paper
examines the medium-term relationship between international trade and
employment in manufacturing in developing countries. Evidence from 59
developing countries, from the early 1960s to the late 1990s, reveals a positive
medium-term association between employment in manufacturing and open-
ness to trade, after controlling for other structural determinants of employ-
ment. By contrast, an opposite relationship is found in high-income coun-
tries. Countries in MENA find it difficult to make trade a driver of employ-
ment creation and growth, partly because MENA exports are concentrated in
low value-added, slowly growing products, and partly because MENA trade is
poorly linked to global production networks and Foreign Direct Investment
*World Bank. †London School of Economics. ††World Bank. The authors wish to thank Manuel Felix
for providing excellent research support, Martin Rama for making available an international database
on employment and labor market regulations, and T.G. Srinivasan for contributing data on trade.
Comments from Farrukh Iqbal, Kiihiro Fukasaku,Douglas Lippoldt, Andrea Goldstein, and other par-
ticipants in an OECD Development Center seminar are gratefully acknowledged.
Published in the MENA Working Paper series, no. 32, July 2003. Reprinted by permission from the
World Bank.

Breaking the Barriers to Higher Economic Growth330
(FDI) flows. Evidence suggests that while the impact of trade expansion on
employment in manufacturing is highly significant in developing countries
that are large FDI recipients, trade adds little to job creation in countries that
receive small amounts of FDI. To meet the employment challenge, trade lib-
eralization and companion policies would need to strengthen the investment
climate and upgrade the quality of trade-related services to improve the
attractiveness of MENA as a place to invest.
Introduction
Accelerating the pace of job creation is a key challenge in MENA. Across the
region, unemployment is high and the working-age population is growing
fast. Even though a young and fast growing labor force is a valuable asset for
the future, it also presents a serious challenge: How to achieve faster, more
labor-intensive growth, to accelerate job creation, and reduce the currently
very high unemployment rates across the region? Past policies, relying on the
expansion of public sector employment and the use of oil rents to stimulate
domestic demand, migration and growth in agricultural employment, are
running out of steam, calling for more innovative approaches to stimulate
employment growth.
Can trade expansion help MENA countries step up the pace of job cre-
ation? In a number of countries that have successfully integrated into global
markets, export-led growth has eventually brought large employment divi-
dends. But evidence on the impact of trade on employment is not clearcut,
because in developing countries, trade expansion usually relies on trade liber-
alization, which may hurt sheltered sectors in the short term and displace
workers in import-competing industries. Moreover, the reforms that help
expand trade are part of more comprehensive programs, aimed at improving
competitiveness and economic efficiency, that may also entail adjustment
costs. However, trade expansion holds the promise of substantial dividends in
terms of job creation and income growth in the medium term. The delocal-
ization of production in developing countries, in labor-intensive manufactur-
ing such as textiles and clothing, footwear, and food processing, eventually
spurs the demand for labor and boosts workers’ earnings.
The paper examines the medium-term relationship between international
trade and employment in manufacturing in developing countries. The analy-
sis draws on a panel data set from 59 developing countries, spanning five-year
periods from the early 1960s to the late 1990s. Evidence reveals a positive
medium-term association between employment in manufacturing and open-
ness to trade, after controlling for other structural determinants of employ-
ment. By contrast, the opposite relationship is found in high-income countries.
But countries in MENA find it difficult to make trade a driver of employ-
ment creation and growth. After controlling for other structural determinants

Part III: Trade, Competitiveness, and Investment 331
of employment in manufacturing, evidence suggests that trade openness has
contributed less to overall employment creation in manufacturing in MENA,
compared to trends seen elsewhere in developing countries. This is so partly
because MENA exports are concentrated in low value-added, slowly-growing
products, and partly because MENA trade is poorly linked to global produc-
tion networks and FDI flows (Nabli and De Kleine, 2000; Yeats and Ng, 2000;
Petri, 1997a,b). Evidence indeed suggests that while the impact of trade
expansion on employment in manufacturing is highly significant in develop-
ing countries that are large FDI recipients, trade adds little to job creation in
countries that receive only small amounts of FDI.
To step up employment growth, MENA exports would need to be diversi-
fied away from raw materials and resource-based manufactures, toward high
value-added, labor-intensive products, linked more closely with international
production networks and global investment flows. The challenge is to bridge
the “quality gap” in MENA trade, through deeper integration with trade part-
ners and improved attractiveness to investment. To meet this challenge, trade
liberalization will not be enough. Companion policies would be needed to
strengthen the investment climate and relax the “beyond-the-border”
constraints—especially in trade-related services—that increase the cost of
doing business and limit the attractiveness of MENA as a place to invest.
The Employment Challenge in MENA
Accelerating the pace of job creation is a key policy challenge in MENA because
unemployment across the region is among the highest in the world—at above
15 percent in most countries, and close to 30 percent in Algeria. High unem-
ployment hinders the reduction of poverty, adds to inequality, and feeds social
instability in an already fragile region. As a result of slow growth and the slack
in the labor market, real wages fell by 30-50 percent in 1980-90, and have stag-
nated or fallen since. Projections suggest that the required employment growth
in MENA, to reduce the unemployment rate by half over the next 15 years,
would range between 4 and 5 percent per year—well above the average growth
of 2.5-3 percent seen in the past (Figure 13.1.a; also see Dhonte et al., 2001).
Fast employment growth is needed not only because of high actual unem-
ployment, but primarily because of the rapid growth of the working-age pop-
ulation. Although the growth of this population in Arab countries is project-
ed to slow down somewhat over the next 15 years, it will remain significantly
faster than in the rest of the world, adding to labor market pressures (figure
13.1.b). Moreover, reflecting the very low, but rising, participation rates of
women, the labor force is likely to grow even faster in the years ahead. Thus,
unless the pace of employment growth accelerates, unemployment could rise
further across the region. According to estimates, 50 million new jobs need to
be created over the next 10 years to employ expected additional job seekers.

Breaking the Barriers to Higher Economic Growth332
This is four times more than Eastern Europe and Central Asia require, and
about as much as in all of Latin America—a region three times bigger than
MENA in terms of GDP. The employed workforce would need to rise by
almost 60 percent in 10 years—an even stronger increase than in East Asia in
its years of high growth.
The employment challenge is further complicated by important structural
imbalances in MENA labor markets that heighten economic inefficiencies and
exacerbate social tensions. First, unemployment is more severe among female
workers, and is on the rise—exceeding by far levels seen in other middle-
income developing countries (figure 13.1.c). This discourages the participa-
tion of women in the labor force—thus preventing a needed increase in the
very low female participation rates—and denies MENA a significant part of
its productive human resources. Second, compared to other middle-income
countries, unemployment in MENA is much higher among skilled workers
with secondary education, while unemployment of workers with higher edu-
cation remains very high as well (figure 13.1.d). Thus, MENA countries are
losing the benefit of substantial past investments in human capital—a sizeable
opportunity cost in a context where the knowledge-based economy is becom-
ing an increasingly strong driver of growth. The high rates of unemployment
among the educated and female workers are also reflected in severe unem-
ployment among the young and first-time job seekers.
Even more worrisome is the fact that in the face of sluggish job creation, con-
tinuing pressures from population growth, and structural imbalances in the
labor market, the mechanisms that sustained employment in past are running
out of steam. In MENA, public sector employment (in civil service and public
enterprises) expanded rapidly in the 1970s and 1980s in response to the oil
boom, because it was seen as a convenient means of redistributing income, pro-
viding a social safety net, and alleviating the pressures of the fast growing flow
of new entrants into the labor market. Despite efforts to scale back public
employment and pay in the 1990s, prompted by the reversal of the oil boom,
both public sector employment and the wage bill remain higher in MENA than
elsewhere in developing countries, and cannot be relied on to promote employ-
ment in the years ahead (Schiavo-Campo, de Tommaso, and Mukherjee, 1997;
figure 13.2.a). If anything, public employment would have to be downsized, as
in many MENA countries the wage bill in the public sector exceeds that in pri-
vate sector manufacturing—contrary to patterns seen in other developing
countries, including the economies in transition (figure 13.2.b).
Moreover, employment in agriculture, which still accounts for about 30
percent of jobs, on average, in MENA, is declining. If anything, the shift out of
agricultural employment would further intensify pressures on the labor mar-
ket. Migration has also provided substantial relief to MENA labor markets in
the past. However, with slowing growth in high-income countries and tighter
immigration controls, the pace of migration has considerably slowed. For

Part III: Trade, Competitiveness, and Investment 333
example, Egyptian immigrant workers, which account for about 10 percent of
the workforce, have been stagnant since 1997.
Can Trade Expansion Become an Engine of Job Creation?
World merchandise trade has grown rapidly during the 1990s, at an annual
rate of about 9 percent, and is also expected to be a major engine of growth
over the next ten to fifteen years (World Bank, 2001). Thanks to continued
reforms that have enhanced competitiveness, developing countries are gaining
strength in global nonenergy merchandise export markets, with their market
share increasing more that 7 percent over the 1990s (figure 13.3.a). And
despite still-high market access barriers for labor-intensive manufactures such
as textiles and clothing, footwear, and food processing, developing countries’
exports increased sharply in the 1990s, and their export market share now sur-
passes that of high-income countries.
Figure 13.1. Accelerating Job Ceation in MENA—a Multifaceted Challenge
0
1
2
3
4
5
6
Algeria Tunisia Iran
(a) (b)
(c) (d)
Morocco Egypt
estimated 1973–94
required 2000–15
Employment growth (estimated and required,
to reduce unemployment by half)
annual rates, %percent
percent
⫺0.5
0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
world Arab countries European Union
1970–2000
2000–15
Annual growth rate of working-age
population (15-64 years)
0
5
10
15
20
25
high-income
countries
middle-income
countries
MENA
Female unemployment in MENA is severe
(unemployed as % of female labor force)
1980
1990
0
5
10
15
20
25
30
probability of unemployment (%)
secondary tertiary primar y
MENA has the greatest problem in
educated unemployed, 1998
MENA
middle-income
countries
high-income
countries
Source: Authors’calculations; partly based on data from Dhonte et al. (2001).

Breaking the Barriers to Higher Economic Growth334
Trade expansion, especially in the form of rising exports, has been a major
source of growth in developing countries. During each of the past two
decades, developing countries which have had fast export growth—leading to
an increase in the share of non-energy merchandise exports in GDP—have
also had, on average, 1 percent higher real GDP growth (figure 13.3.b). Faster
overall growth is in turn the prerequisite for accelerated job creation. And
global trade in manufactures can be a major driver for employment growth in
the years ahead, as it is expected to increase almost threefold by 2010, com-
pared to the late 1990s. By contrast, trade in nonoil commodities is projected
to double, and trade in fuels is expected to rise by only 50 percent (Riordan et
al., 1997). Thus, from the demand side, there is ample room for further
growth in MENA’s non-energy exports, provided the region succeeds in meet-
ing the challenge of increased competition in global markets.
However, MENA was bypassed by expanding global trade in the 1990s.
Across the region, trade flows stagnated at about 40 percent of GDP on aver-
age, while in other developing regions trade expanded rapidly (figure 13.4.a).
The trade performance of MENA countries is even weaker when non-
hydrocarbon exports are considered separately. While a few countries,
including—Bahrain, Morocco, Tunisia, Jordan, and the UAE, succeeded in
diversifying their exports, non-hydrocarbon merchandise exports remained
in a number of countries at very low levels, below 5 percent of GDP (figure
13.4.b). This is even more a concern in countries with a large labor force—
such as Algeria, Egypt, and Iran—that face high and rising levels of unem-
Figure 13.2. Patterns in Public Sector Employment and Pay
0
2
4
6
8
10
12
14
16
18
20
0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
civilian government employment
in % of total employment
central government wage bill in % of GDP
ratio of public to private wages
in percent
ratio
Public to private wages
0
2
4
6
8
10
12
Morocco
Jordan
Tunisia
Mauritius
Egypt
Algeria
Romania
Hungary
Korea
Brazil
government wages
manufacturing wages
% of GDP
Wages in government and manufacturing
Middle
East and
North
Africa
OECD Europe
and
Central
Asia
Latin
America
and the
Caribbean
Sub-
Saharan
Africa
Asia
(a) (b)
Source: Based on Schiavo-Campo, de Tommaso, and Mukherjee, 1997, and authors’ calculations.
Note: Figure 13.2.a. refers to the early 1990s.

Part III: Trade, Competitiveness, and Investment 335
ployment, because the hydrocarbon sector and downstream industries con-
tribute little to job creation.
The slow integration of MENA countries into global trade reflects bottle-
necks in export capacity, but is also linked to the still-high levels of protection
of domestic markets. Despite some progress in the late 1990s in liberalizing
external trade regimes, partly in connection with the Association Agreements
with the EU and partly as a result of unilateral moves, the average level of tar-
iff protection still remains higher in MENA than elsewhere in developing
countries—with the exception of South Asia (figure 13.5). Regions such as
Latin America, where trade protection used to be as high as in MENA, have
slashed tariffs by more than three times in a decade, while low-income
regions, such as Sub-Saharan Africa, have reduced tariffs to levels below those
Figure 13.3. Export Performance and Output Growth in Developing Countries
0
10
20
30
40
50
60
1991–
92 1998–
99 1991–
92 1998–
99 1991–
92 1998–
99
total non-energy
merchandise agriculture labor-intensive
manufactures
in % of world exports
for each product group
exports from middle-
income countries
exports from low-
income countries
exports from LDC
0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
decreasing export
share in GDP
increasing export
share in GDP
in %, average real
GDP growth per year
(a) (b)
1980-89
1990-99
Source: World Bank, GEP 2002.
Figure 13.4. Trade Patterns in MENA
0
10
20
30
40
50
60
70
80
90
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999
ECA
EAP
SSA
MENA LAC
SA
exports and imports of goods and non-factor
services in % of GDP (in volumes)
0
10
20
30
40
50
Bahrain
Tunisia
Jordan
Morocco
UAE
Syria
Saudi Arabia
Oman
Algeria
Egypt
Kuwait
Yem en
Iran
1986–90
1996–2000
nonfuel merchandise
exports in % of GDP
(a) (b)
Source: Author’s calculations.

Breaking the Barriers to Higher Economic Growth336
in MENA. Besides tariff barriers, “para-tariffs” are also widespread in MENA.
They are often geared to raising revenues for the state, but they also, in effect,
protect domestic companies (Zarrouk, 2000).
The delocalization of production of labor-intensive manufactures in devel-
oping countries has the potential to spur the demand for labor and boost
workers’ earnings. The most visible part of job creation, driven by exports of
labor-intensive manufactures, has been associated with FDI—as for example
in China’s Eastern provinces, or in Mexico’s Maquiladoras. Thanks to the
NAFTA and radical economic reforms, Mexico’s trade more than tripled, from
US$82 billion in 1990 to about US$280 billion in 1999, making Mexico the
seventh largest trading nation in the world. The pace of job creation has been
particularly swift in manufacturing, shared most in the expansion of trade
(figure 13.6.a). Job creation has also been strong in export processing zones in
a number of developing countries—such as Mauritius, the Dominican
Republic, and El Salvador (Rama, 2001).
Indonesia is an earlier (mid-1980s) example of a country that started a major
trade reform effort, whichsubstantially reduced nontariff barriers and impedi-
ments to foreign investment. Manufactured exports and FDI boomed and were
accompanied by rising manufacturing employment rates (figure 13.6.b). The
Indonesia case is especially relevant to some MENA countries, since Indonesia
was “single-engine” economy (oil and natural resources) until the mid-1980s,
when it found a new engine of growth in the form of manufactured exports
(Agrawal, 2002). In MENA, Tunisia is an example of successful diversification
out of resource-based exports. Tunisian exports of textiles and clothing have
boomed in the distortion-free environment for “offshore companies” that sup-
ply foreign markets, while employment in the offshore sector increased steeply.
However, trade expansion may occur in a number of different ways that
affect employment differently—as a result of better access to foreign markets,
Figure 13.5. MENA Markets Still Remain Highly Protected
0
10
20
30
40
50
60
70
80
South
Asia
Middle East
and
North Africa
Sub-Saharan
Africa
East Asia Europe and
Central Asia
Latin
America
1986–90
1991–95
1996–98
Source: Authors’ calculations.

Part III: Trade, Competitiveness, and Investment 337
for example, or of lower international trade barriers or export-oriented FDI
seeking to take advantage of an economy’s comparative strengths. But, in most
cases, sustained trade expansion follows domestic reforms that reshape taxes
and incentives in the economy. Typically, such reforms call for lowering tariff
protection and nontariff barriers, with the aim of reducing the anti-export
bias of protective external trade regimes—especially in countries where nar-
row domestic markets cannot provide sufficient support for industrial
growth. Trade liberalization reduces the anti-export bias, as it helps domestic
producers purchase inputs at internationally competitive cost. By increasing
the profitability of export sectors, trade liberalization helps shift resources to
the uses that give countries the greatest comparative advantage.
Because the previously protected import-substitution sectors are likely to
be capital-intensive, semiskilled (especially female) labor is likely to be under-
utilized. With sizeable amounts of labor remaining underemployed in the
home, or queuing for public sector employment, trade liberalization is likely
to have a net positive impact on employment in the medium term. But in
developing countries, the fear is that massive trade liberalization would erode
rents, expose inefficient industry to competition and cost jobs. Indeed, trade
liberalization may disrupt job creation in the short term for a number of
reasons:
• Lowering trade barriers may initially hurt sheltered domestic producers
and displace unskilled workers in import-competing industries. Though
import-competing industries are usually capital-intensive, in many
middle-income countries—and also in MENA—industries intensive in
unskilled labor are often protected disproportionately, because they face
potentially stiff competition from lower-cost producers (Wood, 1997). For
example, in Morocco, before trade liberalization, the nominal tariff and
Source: Authors’ calculations.
Figure 13.6. Examples of Trade Expansion and Job Creation in Manufacturing
3,000
3,200
3,400
3,600
3,800
4,000
4,200
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
0
100
200
300
400
500
600
persons (OOO)
millions constant pesos, 1995 (000)
exports of goods
(right axis)
employment
(left axis)
Employment in manufacturing and
manufacturing exports/GDP
0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
0
2
4
6
8
10
12
14
16
employment in
manufacturing
(left axis)
exports in manufacturing (% of GDP)
% of working age population
p
ercent
(a) (b)

Breaking the Barriers to Higher Economic Growth338
import license coverage in apparel and footwear was among the higher in
manufacturing (Currie and Harrison, 1997). Similarly, in 1995, Egypt’s
import-weighted tariffs on textiles were about three times higher than
average tariffs for the economy as a whole (Dessus and Suwa-Eisenman,
1998).
• Trade reallocates activity and labor across import-competing and export-
oriented sectors. But while market exit of previously sheltered companies
may be swift, business expansion takes time, and the timing of the net ben-
efits will depend on the flexibility of product and labor markets, and on the
availability of finance. Bottlenecks in access to credit or in availability of
trade-related services (transport, communications) may tame the growth
of export-oriented industries in the medium term. Moreover, the quality of
the investment climate affects investment and, thus, job creation. In some
cases, the investment climate may not be sufficiently attractive—so that
export-oriented companies may lack incentives to expand and absorb labor
released by the contracting, import-competing industries.
• Companion policies such as exchange rate management also affect the
impact of trade policy reform. Exchange rate misalignment has been a fac-
tor in weak export performance in manufacturing in MENA (Nabli and
Veganzones, 2002; Sekkat and Varoudakis, 2002). Morocco is a case in
point: Over the 1980s, Morocco witnessed fast growth in exports and
employment in manufacturing, supported by trade liberalization.
However, over the 1990s, the growth in manufactured exports and employ-
ment ran out of steam, partly owing to deteriorating competitiveness (fig-
ure 13.7.a). The fixed exchange regime implemented in the 1990s helped
achieve stabilization, but led to a 22 percent appreciation of the real effec-
tive exchange rate over the decade, which heightened competitive pressures
on the tradable goods sector (figure 13.7.b). An association of textile pro-
ducers reported the loss of 29,600 jobs in the textile industry (about 12 per-
cent of employment in that industry) since 1999 (IMF, 2001).
Trade liberalization may thus typically lead to an increase in the rate of
unemployment, which may take some time to reverse (figure 13.8.a).
Although unemployment is generally trending downward in the medium
term, its persistence will depend on market flexibility, exchange rate policy,
and other reforms that may accompany trade liberalization. In the adjustment
process, the impact of trade liberalization is difficult to single out, because
other reforms—such as privatization of state-owned enterprises and reforms
of bloated administrations and government agencies—are also likely to gen-
erate employment costs.
Job destruction has been particularly dramatic when trade policy reform
has been associated with large-scale downsizing of state-owned enterprises, as

Part III: Trade, Competitiveness, and Investment 339
in the transition economies where millions of workers had to be made redun-
dant for the restructured enterprises to become profitable as private firms. For
example, in Algeria, an estimated 500,000 workers—about 10 percent of the
labor force—lost their jobs from 1995 to 1999 as a result of only partial
restructuring of nonviable state-owned enterprises.
The increase in unemployment during the adjustment to trade liberaliza-
tion may also be associated with a decline in real wage growth, as a result of
the slack in the labor market (figure 13.8.a). But evidence on the net employ-
ment impact of labor market adjustment is mixed. In Mexico, Revenga (1997)
found that even in the protected sectors, trade liberalization resulted in lower
wages when rents were eroded, rather than lower employment. The wage
reductions were uneven, but a feature of her microdata set was that it showed
Figure 13.7. Growth of Manufactured Exports and Employment in Morocco
Figure 13.8. Adjustment Costs of Trade Liberalization
Employment in manufacturing and manufacturing exports/GDP
1.5
1.7
1.9
2.1
2.3
2.5
2.7
2.9
3.1
3.3
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
employment in
manufacturing
(left axis)
exports in
manufacturing
(% of GDP) (right axis)
% of working-age population
percent
80
100
120
140
160
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
Morocco, real eective exchange rate (1990 = 100)
(a) (b)
Source: Authors’calculations—based on IMF data on the real effective exchange rate,
0
5
10
15
20
25
⫺1 1 3 5 7 9 11 13 15 17 19
year in the reform process
unemployment rate (%)
Chile
Mauritius
Poland
Sri Lanka
Wages and openness to trade
⫺0.5
⫺0.4
⫺0.3
⫺0.2
⫺0.1
0
0.1
0.2
0.3
after
1 year
after
2 years
after
3 years
after
4 years
after
5 years
eect of an extra 1% of GDP in trade...
on the level of wages (in %)
(a) (b)
Source: World Bank.

Breaking the Barriers to Higher Economic Growth340
wage reductions, but no employment reductions—across the board.
Experience in Morocco tells a similar employment story. Although experience
across different occupations differed, trade liberalization surprisingly had no
noticeable impact on either wages or employment. Currie and Harrison
(1997), who studied a large microdata set for Morocco, concluded that the
reduction in economic rents was absorbed by a reduction in profit margins
and improvements in labor productivity, but not by less overall employment.
In the medium term, as the labor-intensive, export-oriented sectors gain
strength, the demand for labor increases, and leads to an increase in real
wages. The net benefit to wage earners appears, on average, after the fourth
year of the adjustment process (figure 13.8.b).
The Medium-Term Impact of Trade Expansion on Manufacturing Employment:
Evidence from Developing and High-Income Countries
To assess the medium-term impact of trade expansion on employment, we
examined evidence from both developing and high-income countries. The
developing country sample includes 59 countries—containing about 140
observations, spanning five-year periods from the early 1960s to the late
1990s. The high-income country sample includes 22 countries and 135 obser-
vations of five-year periods, spanning the same period. In order to remove
shortrun fluctuations, we averaged the data over five-year periods.
Casual inspections of the developing country sample reveals a positive
medium-term association between employment in industry (as a share of the
total working-age population) and openness to trade, but the association is at
first sight weak, because there is considerable variation in employment out-
comes across countries (figure 13.9). Cross-country differences in employ-
ment ratios in manufacturing reflect, indeed, a number of diverse factors
apart from trade that may affect the demand (or the supply) of labor. Such
factors may include the relative sizes of the primary and services sectors—
which depend on the level of development; the overall level of production
capacity and technical skills in manufacturing; the size of the informal econ-
omy (since the reported employment ratios capture employment in the for-
mal sector); the level of real wages; and socioeconomic factors that affect the
participation of women in the labor force.
To account for different structural factors that may affect employment in
manufacturing, we estimated employment equations that include other deter-
minants of the demand for labor, along with a variable for trade effects. The
explained variable in the regressions presented in the tables below is employ-
ment in manufacturing as a percent of working-age population. A number of
controls were used: (i) real labor costs per worker; (ii) a measure of total phys-
ical capital as a ratio to total employment; and, (iii) the real interest rate. The
capital-to-employment ratio captures changes in manufacturing employment

Part III: Trade, Competitiveness, and Investment 341
due to growth in production capacity,and to the shift of employment between
sectors that accompanies economic development (e.g., from manufacturing
to the services sector). All regressions are logarithmic (except for the real
interest rate) and were run using fixed effects. Four different indicators of
trade expansion were used: (i) total trade flows (the sum of exports and
imports); (ii) total exports; (iii) merchandise exports; and (iv) merchandise
exports excluding hydrocarbons. All four indicators are measured relative to
GDP. While the three last indicators are proxies for export-led growth, the first
indicator also accounts for import penetration, and thus also indirectly
reflects the impact of trade liberalization (box 13.1).
The equation fits the data well, with real wages having a negative impact
on employment in manufacturing and capital having a positive impact (table
13.1). The size and significance of the coefficients varies according to the dif-
ferent specifications. All else equal, a 10 percent increase in real labor costs
lowers the industrial employment ratio by an estimated 2 to 3 percent on
average. High real interest rates appear to also depress industrial
employment—though in a statistically less robust way across the different
specifications.
The findings suggest that trade expansion has a positive medium-term
impact on employment in developing countries. All coefficients associated
with trade expansion are statistically significant. This comes true for the var-
ious measures of export performance, but also for the broader measure of
trade openness that accounts for import penetration. All else equal, a 10 per-
cent increase in the share of nonoil merchandise exports in GDP is associated
with an increase in the employment ratio in manufacturing of about 1.4 per-
cent, while the same increase in the share of trade flows in GDP could raise
the employment ratio by an estimated 2.3 percent.
Figure 13.9. Employment in Manufacturing and Trade Flows: Evidence from Developing Countries
0
5
10
15
20
25
30
0 50 100 150 200
employment in industry,
% of working age population
foreign trade ows, % of GDP
Source: Authors’ calculations.

Breaking the Barriers to Higher Economic Growth342
The estimates also accounted for the fact that the measures of trade expan-
sion are endogenous, which could be at the origin of some bias, in that both
trade and employment in manufacturing could be affected by a common set
of factors not included in the regression. One of the specifications (table 13.1,
column 6) uses the black market premium, total world trade as a share of
world GDP, and the country population as instruments for trade expansion.
The results are consistent with the previous specifications, with total mer-
chandise exports remaining a significant determinant of the manufacturing
employment ratio. However, when instrumental variables are used, the signif-
icance of trade expansion is not always robust across specifications.
Box 13.1. Measuring the impact of trade liberalization on employment
Employment effects are likely to be different when trade expansion is the result of a
reform process that restructures implicit and explicit taxes and incentives in the
economy. In developing and transition economies, trade expansion usually follows
a reform process that may well have more important effects on employment than
the growth of trade itself. Usually these other effects are more difficult to assess, and
likely to be more diverse than the reallocation effects of trade. The diversity of the
likely effects explains why the empirical literature has not reached a consensus
about the overall effects of trade on labor market outcomes.
If trade liberalization follows other reforms, should one attribute the employ-
ment effects of the whole process to trade? In an ideal world, the answer is likely to
be no. But in the real world of political economics, a country needs incentives to
reform. The institutions that are dismantled in the reform process shield some sec-
tions of society; these sections are usually the ones with the power to stop the
reform from taking place. The prospect of beneficial trade growth gives incentives
to those in power to push through the reforms, and those that are hurt by the
process to accept them.
In the statistical analysis, it would be difficult to describe the institutional frame-
work of each country and measure the implications of its reform. There is very little
in the literature on measures of institutional rigidity, and what there is usually
applies to single points in time and to the richer countries that keep more com-
plete statistical records. It is therefore not possible to disentangle the employment
effects of institutional reform from those of the trade expansion that follows the
reforms. But our usual measures of trade expansion may not be bad proxies for
reform. A country that dismantles rigid institutions in labor markets when it liberal-
izes trade is likely to experience faster trade growth than one that keeps the rigid
institutions. Trade growth in the statistical analysis picks up both the direct effects
of trade and the indirect effects of reform. For this variable to be a good proxy, it
has to bear a monotonic relation to the degree of reform. Although there is no
research on this point, intuitively it makes sense. Trade growth normally requires
restructuring of employment and countries with flexile labor markets are in a bet-
ter position to take advantage of the new trade opportunities that liberalization
offers.

Part III: Trade, Competitiveness, and Investment 343
The estimates suggest that the medium-term benefits of trade expansion in
terms of employment could be substantial. In MENA, the share of nonoil
merchandise exports in GDP was about 10 percent on average, against 23 per-
cent in East Asia and the Pacific (regression sample statistics). Bridging just
half of this gap in export performance could bring about an estimated 2 per-
cent increase in industrial employment as a share of working-age population.
This would be equivalent to a 4 percent decrease in the average unemploy-
ment rate, as participation in the labor force in MENA amounts to only about
50 percent of the working-age population.
Moreover, the full impact of trade expansion on manufacturing employ-
ment may be underestimated, because the data only account for formal
employment in manufacturing. With the regulatory framework in the labor
market unchanged, an increase in formal employment is likely to also have an
impact on informal manufacturing employment. Higher employment and
incomes in (formal and informal) manufacturing would also boost domestic
expenditure in nontradables, so second-round multiplier effects from trade
expansion could further contribute to economy-wide job creation.
Table 13.1. Determinants of Employment in Manufacturing—Developing Countries
(Dependent variable: ratio of employment in manufacturing to working-age population) (Estimation period: 1960–95)
Employment in manufacturing and openness to trade in developing countries
Explained variable: log of employment
Explanatory variables in manufacturing-to-working-age population ratio
Log of trade-to-GDP ratio 0.229* 0.613**
Log of trade-to-GDP ratio non-MENA
countries 0.247**
Log of trade-to-GDP ratio MENA countries –0.04
Log of exports-to-GDP ratio 0.202*
Log of merchandise exports-to-GDP ratio 0.312**
Log of merchandise exports non-MENA
countries 0.343**
Log of merchandise exports MENA
countries 0.054
Log of merchandise exports (excluding
petroleum)-to-GDP ratio 0.143** 0.201*
Log of labor costs in manufacturing –0.187* –0.192* –0.271* –0.216* –0.339** –0.301* –0.198* –0.266*
Log of total physical capital-to-total
labor force ratio 0.15 0.159 0.396** 0.240** 0.257 0.319* 0.169* 0.370**
Real interest rate –0.005* –0.003
Constant 0.589 0.764 –2.13 0.104 –1.228 –0.481 0.45 –1.845
Observations 140 140 110 134 102 96 140 110
Number of group (country) 49 49 49 45 44 40 49 49
R-squared 0.16 0.16 0.3 0.2 0.36 0.23 0.18 0.32
Source: Data on wages and employment are from Rama and Artecona (2000); data on total physical capital are from Sandeep Mahahjan (2001);
data on merchandise exports including nonoil merchandise exports comes from UN Comtrade database; data on working-age population,
exports, total population, GDP, total world trade come from World Development Indicators, World Bank.
Note: All figures are US $ based. Working-age population is defined as persons between 15 and 64 years old.
*significant at 5 percent; **significant at 1 percent.

Breaking the Barriers to Higher Economic Growth344
It is noteworthy that trade has a different impact on manufacturing
employment in high-income and developing countries. The results presented
in table 13.2 and figure 13.10 suggest that, after controlling for other factors,
trade has a negative impact on manufacturing employment in high-income
countries. The results are statistically significant only for the trade-to-GDP
ratio and total exports, but all coefficients signs are congruent across specifi-
cations. Indeed, in several high-income countries, trade intensification has
gone in tandem with delocalization of production to developing countries,
along with a shift towards areas of comparative advantage in higher-skill
activities in services.
Given the trends of the variables used in the regressions, the story told by
these estimates is consistent with the theoretical argument that in the medi-
um to long run, capital growth increases the demand for labor, but is absorbed
by wage growth, which offsets its impact on employment. In the economy as
a whole, trade should have no role to play in an employment equation, but in
a regression restricted to manufacturing it has a role. In high-income coun-
Table 13.2. Determinants of Employment in Manufacturing—High-Income Countries
(Dependent variable: ratio of employment in manufacturing to working-age population) (Estimation period: 1960–95)
Employment in manufacturing and openness to trade in developing countries
Explained variable: log of employment
Explanatory variables in manufacturing-to-working-age population ratio
Log of trade-to-GDP ratio –2.59*
[2.22]
Log of exports-to-GDP ratio –0.267*
[2.29]
Log of merchandise exports-to-GDP ratio –0.24
[1.76]
Log of merchandise exports (excluding –0.121
petroleum)-to-GDP ratio [1.14]
Real interest rate
Log of labor costs in manufacturing –0.447* –0.443** –0.261 –0.401**
[3.88] [3.84] [2.02] [3.03]
Log of total physical capital-to-total labor 0.297** 0.309** –0.144 0.244**
force ratio [3.69] [3.76] [1.20] [2.80]
Constant 3.828** 3.434** 8.609** 3.491**
[7.15] [5.79] [5.63] [5.79]
Observations 135 135 60 121
Number of group (country) 22 22 21 20
R-squared 0.18 0.18 0.32 0.1
Source: Data on wages and employment are from Rama and Artecona (2000); data on total physical capital are from Sandeep Mahahjan (2001);
data on merchandise exports including nonoil merchandise exports comes from UN Comtrade database; data on working-age population, GDP
come from World Development Indicators, World Bank.
Note: All figures are US $ based. Working-age population is defined as persons between 15 and 64 years old.
*significant at 5 percent; **significant at 1 percent. Absolute value of t-statistics in brackets.

Part III: Trade, Competitiveness, and Investment 345
tries, wage growth alone cannot explain the fall in manufacturing employ-
ment because wages are common across the whole economy. Trade expansion
lies behind the fall in manufacturing employment relative to employment
elsewhere.
The influence of trade expansion on the overall level of employment in
developing countries that dismantle trade barriers is different from its effect
on employment in developed countries. Trade barriers and other institution-
al rigidities have deadweight costs, in addition to disincentives that are associ-
ated with implicit and explicit taxes used to finance them. Removing costly
institutional structures increases national welfare, improves incentives, and
increases both the demand and supply of labor, by attracting more people of
working age into the labor force. Thus, in developing countries, trade expan-
sion promotes manufacturing employment in the medium term, because it
allows these countries to take better advantage of their comparative advantage
in labor-intensive industries.
Why Has Trade Expansion Had a Weak Impact on
Manufacturing Employment in MENA?
Despite the evidence presented so far, there is widespread sentiment that the
impact of trade expansion on manufacturing employment in MENA has been
weak. That perception is, indeed, confirmed by our estimates: As shown in
table 13.1, when estimated separately, the coefficients of trade in the employ-
ment equations turn out not to be significant for the MENA countries. This is
true not only for broad indicators of trade flows, but also for indicators of
export performance. To be sure, employment in manufacturing is higher in
Figure 13.10. Trade Expansion and Manufacturing Employment in Developing and High-Income Countries
R2 = 0.6037
R2 = 0.466
⫺1.8
⫺1.6
⫺1.4
⫺1.2
⫺1.0
⫺0.8
⫺0.6
⫺0.4
⫺0.2
0
0 50 100 150 200 250
Employment in industry unexplained by factors
other than trade openness
(as a ratio of total working-age population; in logarithm)
foreign trade ows in percent of GDP
High-income countries
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
0 20 40 60 80 100 120 140 160 180
foreign trade ows in percent of GDP
Developing countries
(a) (b)
Source: Authors’ calculations.

Breaking the Barriers to Higher Economic Growth346
countries with high nonoil merchandise exports (Tunisia, Morocco, Jordan)
compared to countries with weak export performance (Algeria, Egypt).
However, such differences are largely explained by country-specific factors
(such as, for example, greater female participation in the labor force in
Tunisia) and structural factors other than trade. The employment ratio
responded only modestly to changes in the share of nonoil merchandise
exports in GDP in both groups of MENA countries (figure 13.11). Empirical
evidence suggests that, contrary to the experience elsewhere in developing
countries, when such structural factors are accounted for, trade expansion did
not have a significant impact on industrial employment in MENA.
What might be the reasons for the weak impact of trade on employment in
MENA? A number of factors can be singled out. They are partly related to
what might be called the “quality gap” in MENA trade, and partly to poorly
performing labor market institutions that tame the medium-term benefits
from increased trade openness while exacerbating the adjustment costs.
A number of attributes of MENA trade may account for the weak impact
of trade on employment:
• Non-hydrocarbon exports are concentrated in resource-based, low value-
added products, whose growth has only a weak impact on labor demand
and employment (Petri, 1997). Moreover, in these sectors, world trade is
growing slowly, so that MENA could not take advantage of the strong
growth in world trade growth over the 1990s that gave an impetus to
employment creation elsewhere in developing countries.
• In hydrocarbon-rich countries such as Algeria,exports of manufactures are
concentrated in the downstream energy industries, which are capital-
Figure 13.11. Impact of Trade Expansion on Manufacturing Employment in MENA
0
2
4
6
8
10
12
14
16
18
0 5 10 15 20 25 30
non-oil merchandise exports, % of GDP
Employment in industry (as a ratio of total
working-age population, %)
Tunisia
Algeria
Egypt
Source: Authors’ calculations.

Part III: Trade, Competitiveness, and Investment 347
intensive and have only little impact on employment (for example, refined
gas; fertilizers; plastics).
• MENA trade is poorly integrated into cross-border production-sharing
networks, which have increasingly become a driver of growth in global
trade (Humels et al., 2001; Yeats and Ng, 2000). MENA exports do not,
thus, benefit from rapidly expanding vertical trade. As a result, FDI—the
complement of increased participation in global production networks—
remains limited. The weak responsiveness of foreign and domestic invest-
ment to trade liberalization tames the impact on job creation.
The functioning of the labor markets may also affect the impact of trade on
job creation, because the net effect on employment depends on the response
of wages to the reform process—and thus on the institutional rigidities in the
labor market and the wage setting process. To give some examples, the reforms
and trade expansion that accompanied Spain’s transition to democracy in the
late 1970s did not give rise to more employment because newly-emancipated
trade unions claimed those benefits in the form of higher wages. The integra-
tion of eastern Germany into the western economy in the 1990s tells a similar
story, as the newly liberated labor in the east sought to catch up with their
western counterparts through overvalued wages (and their western counter-
parts supported the big wage rises to stop massive immigration).
In MENA, labor market rigidities are largely associated with the important
role of public sector employment and pay, which sets the stage for real wage
increases and employment conditions in the formal labor market. As shown
in Figures 13.2.a. and 13.2.b., MENA is the developing region with the largest
share of public sector employment and pay—by far larger than the share of
manufacturing employment in total employment or of the share of manufac-
turing wages in GDP. The large share of public sector employment and pay
makes MENA the only region where real wages in the public sector are, on
average, higher than real wages in the private sector. Because of non-
pecuniary benefits of public sector employment and job security, this leads to
queuing for public sector jobs, and puts pressure on private-sector real wages
in order to attract workers (especially skilled workers) when activity is
expanding. The restrictive employment regulations in the private sector for-
mal labor markets, and the high nonwage costs, may further tame the respon-
siveness of job creation to growing trade.
Among the above factors, our estimates highlight, in particular, the role of
FDI as one explanation for the weak impact of trade on jobs in MENA.
Indeed, in developing countries, the impact of trade expansion on employ-
ment is likely to be reinforced by capital flows. Trade liberalization allows large
international corporations to take advantage of the cheaper labor in develop-
ing countries and locate processing plants in them, through direct investment.
FDI flows into developing countries have been identified with higher wages

Breaking the Barriers to Higher Economic Growth348
and with more male/female wage equality.They increase the demand for labor
in the receiving country by increasing the supply of capital. The impact of this
on employment creation is likely to be greater than the impact of the demand-
driven increase in labor demand because although the demand-driven
increase may hit supply bottlenecks, the FDI-induced increase is not likely to.
Indeed, the effect of more FDI on the domestic economy must be beneficial,
unless the higher wages that they pay provoke comparability demands else-
where in the economy, an argument that has not received support in the
empirical literature.1
The estimates reported in table 13.3 confirm the critical role of FDI in job
creation. They are similar to the estimates for developing countries reported
in table 13.1, with the difference that the observations in the sample are split
in two different groups: (i) a group of developing countries with large FDI
inflows and, (ii) a group of small FDI recipients—the cutoff point being the
overall sample median of 0.7 percent of GDP. While the impact of trade
expansion on employment in manufacturing is highly significant in the group
of large FDI recipients (figure 13.12.a), the estimated coefficients of trade turn
out not to be significant in the group of countries that receive only small
Table 13.3. Determinants of Employment in Manufacturing in Developing Countries: The Role of Foreign Direct
Investment
(Dependent variable: ratio of employment in manufacturing to working-age population) (Estimation period: 1990–95)
Employment in manufacturing and openness to trade in developing countries by foreign direct investment level
Explained variable: log of employment
Explanatory variables in manufacturing-to-working age population ratio
Log of trade-to-GDP ratio * high FDI 0.534**
Log of trade-to-GDP ratio * low FDI 0.021
Log of merchandise exports * high FDI 0.270*
Log of merchandise exports * low FDI 0.309
Log of exports of goods and services * high FDI 0.447*
Log of exports of goods and services * low FDI –0.013
Log of merchandise exports excluding
petroleum * high FDI 0.222*
Log of merchandise exports excluding
petroleum * low FDI –0.017
Log of labor costs in manufacturing –0.074 –0.377* –0.054 –0.419* –0.044 –0.383* –0.079 –0.483**
Log of total physical capital-to-total
labor force ratio 0.195 0.377* 0.305 0.573* 0.157 0.396* 0.323 0.496**
Constant –2.172 –0.235 –2.414 –3.452 –1.208 –0.329 –2.328 –0.884
Observations 60 69 53 57 60 69 57 65
Number of group (country) 31 35 27 32 31 35 29 33
R-squared 0.38 0.19 0.36 0.3 0.35 0.19 0.28 0.26
Source: Data on wages and employment are from Rama and Artecona (2000); data on total physical capital are from Sandeep Mahahjan (2001);
data on merchandise exports including nonoil merchandise exports comes from UN Comtrade database; data on working-age population,
exports, total population, GDP, total world trade come from World Development Indicators, World Bank.
Note: All figures are US $ based. Working-age population is defined as persons between 15 and 64 years old.
*significant at 5 percent; **significant at 1 percent.

Part III: Trade, Competitiveness, and Investment 349
amounts of FDI (figure 13.12.b). This is true whatever the indicator of trade
in the regressions.
But MENA countries have lagged considerably behind other developing
regions in attracting FDI, which also explains the limited impact of trade expan-
sion on job creation. MENA has missed in particular the surge in FDI to devel-
oping countries seen in the 1990s, with the share of FDI in GDP remaining
compressed to 0.5 percent, against 2.5 percent, on average, in developing coun-
tries (figure 13.13). Moreover, the structure of FDI in MENA is highly skewed
toward the hydrocarbon sector, thus contributing even less to job creation in
manufacturing and in services. Reflecting the slow pace of restructuring and
privatization of the business sector in MENA, and a weak investment climate,
non-energy FDI stagnated over the 1990s, compared to the levels seen in lower
middle-income countries in Latin America, Central and Eastern Europe, and
East Asia (Nabli et al., 2000; Petri, 1997b; Council on Foreign Relations, 2002).
Rising to the Challenge: Strengthening the Investment Climate and Enabling
Greater Participation in Global Production-Sharing Networks
To step up employment growth, MENA exports would need to be diversified
away from raw materials and resource-based manufactures and toward high
value-added, labor-intensive products, linked more closely with international
production networks and global investment flows. International evidence
reviewed in this paper suggests that, to achieve these goals, and provide a stim-
ulus to employment creation, lowering the still high degree of trade protection
Figure 13.12. Quality of the Investment Climate as a Determinant of the Impact of Trade on Job Creation in
Manufacturing
0
0.2
0.4
0.6
0.8
1.0
1.2
non-oil merchandise exports, in percent of GDP
Employment in industry after controling for factors other than non-oil merchandise
exports (as a ratio of total working-age population, in logarithm)
Developing countries
Large FDI recipients
⫺0.5
⫺0.4
⫺0.3
⫺0.2
⫺0.1
0
0.1
0.2
0.3
0.4
non-oil merchandise exports, in percent of GDP
Developing countries
Low FDI recipients
R2 = 0.4639
R2 = 0.0044
0 05101520253035404510 20 30 40 50 60 70
(a) (b)
Source: Authors’ calculations.

Breaking the Barriers to Higher Economic Growth350
in MENA will not be enough. Companion policies would be needed to help
bridge the “quality gap” in MENA trade, by strengthening the investment cli-
mate and relaxing the “beyond-the-border” constraints that increase the cost of
doing business and limit the attractiveness of MENA as a place to invest.
Improving participation in global production-sharing is key, because a
common pattern of integration in today’s global economy is the increasing
fragmentation of production chains across borders (Arndt and Kierzkowski,
1999). This is reflected in far above average growth of global trade in compo-
nents and partially assembled manufactured goods (Yeats, 2000). Sharp reduc-
tions in the cost of moving goods across borders have enabled firms to better
coordinate production in different locations, and have facilitated exporters’
linkages with vertical production chains that stretch increasingly across bor-
ders (Hummels et al., 2001). Lower logistics costs have resulted from an accel-
erating “logistics revolution”—driven by the more widespread use of contain-
ers in trade; the adoption of “just-in-time” manufacturing techniques;
enhanced supply-chain management; and more widespread use of informa-
tion technology and the internet in logistics. Lower levels of trade protection
have also facilitated the fragmentation of production across borders.
Given the increasing sophistication of the division of labor in the global
economy, efficient trade-related services are becoming key in enabling produc-
ers at various stages of production chains to better coordinate their activities
with intermediate input suppliers located in other countries. Speed, flexibility,
reliability, and low cost of transport and information logistics are particularly
adding value to companies participating in production chains around the
globe. Slow or unpredictable delivery delays the response to new market
opportunities and rapidly changing demand patterns, forcing customers to
hold costly buffer stocks, and making supply chain management ineffective.
Countries that have strengthened their positions in global production chains
have improved their Information and Communication Technology (ICT)
Figure 13.13. MENA’s Share of the Surge in FDI to Developing Countries
⫺1.0
⫺0.5
0.0
0.5
1.0
1.5
2.0
2.5
3.0
1971–75 1976–80 1981–85 1986–90 1991–95 1996–2000
Middle East and North Africa
total to developing countries
Source: Authors’ calculations.

Part III: Trade, Competitiveness, and Investment 351
capabilities; lowered the cost of transport; and created more competitive
finance and insurance markets. Better service delivery has greatly contributed
to reducing the cost of doing business, thus improving the attractiveness of
these countries to both foreign and domestic investment.
Because the location of manufacturing activities has become, competition
to maintain positions has increased. A strong investment climate and logistical
excellence are important parts of all success stories to date. Countries that have
created more open, investment-friendly markets have been able to attract sig-
nificant flows of FDI along with their integration into broader economic areas.
But for this to happen, the reduction of trade barriers had to go in tandem with
broader regulatory reform that improved the attractiveness to investment.
For example, the Central and Eastern European countries (CEECs) that
gained EU accession status carried forward broad-based restructuring pro-
grams, while aligning their regulatory framework to the EU single market.
More ambitious reformers, and countries that were more successful in inte-
grating EU production networks, attracted massive FDI that boosted growth.
By contrast, the Commonwealth of Independent States (CIS) countries lagged
far behind in terms of industrial restructuring, trade expansion, and growth
(World Bank, 2002b). Trade liberalization and regulatory reform in the
CEECs under the European Association Agreements (EAAs) also spurred a
deeper integration of these countries into the EU production-sharing net-
works. CEECs exports of parts have thus increased fourfold from 1993 to
1998—to about US$12 billion, or 14.2 percent of CEECs manufactured
exports (Kaminski and Ng, 2001). The shares of parts in manufactured
exports have thus approached those seen in more integrated countries in
global production sharing, such as Malaysia and Mexico (about 19 percent in
1998).2Integration in EU production networks has been also a factor in
attracting FDI—as evidenced by the positive correlation between FDI per
capita and the share of parts in total exports (figure 13.14).
MENA countries are still poorly integrated in global production-sharing
networks, as reflected by their small share of global FDI flows and trade. The
share of components in manufactured exports remains far below that seen in
other developing countries, such as Singapore, Malaysia, and Taiwan (Yeats
and Ng, 2000). One exception is textile and clothing, especially reflecting
Tunisia’s strong position in EU companies’ outsourcing chains.
Trade liberalization, especially in the countries that have signed the
Association Agreements with the EU (Tunisia, Egypt, Morocco, Jordan, and
most recently Algeria), will help MENA producers improve their competitive-
ness by purchasing inputs at internationally competitive cost. Moreover, MENA
countries could be attractive locations for assembly operations due to low labor
costs and a good quality of human resources. The decrease in tariffs on import-
ed intermediate inputs, scheduled in the first stages of the Association
Agreements, has the potential to increase trade in components across the

Breaking the Barriers to Higher Economic Growth352
Mediterranean and facilitate the integration of MENA countries into EU pro-
duction networks. The planed free trade zone with the EU thus provides a
unique opportunity to MENA to attract more FDI, increase exports, and bene-
fit from knowledge and technology spillovers. This could also help upgrade the
quality of MENA exports toward more high value-added and technologically
advanced products, as in the example of the economies in East Asia.
However, MENA has yet to rise to this challenge. Slow progress to date
partly reflects heightened competition from the transition economies in
Central and Eastern Europe. However, domestic weaknesses, due to the weak
investment climate and the poor quality of backbone services that facilitate
trade, dilute the potential advantages of MENA countries. Bottlenecks in
transport logistics are particularly damaging, but they are not only limited to
poor quality and high cost of transport and information services per se.
Trade-related controls in MENA are associated with burdensome administra-
tive procedures, and create red tape.
Creating an enabling environment for foreign and domestic investment
would be key to reaping the benefits from further trade liberalization in
MENA. In order to encourage transnational companies to extend their supply
chains to MENA through partnerships with domestic companies or new
investment, further progress in lowering trade barriers should go in tandem
with complementary policies in other areas. In particular,trade logistics, trans-
port, and information systems would have to become more flexible, reliable
and sophisticated. This would require ambitious opening up of service markets
to competition—supported by continuous efforts at public enterprise reform
in network industries; privatization; and procompetitive regulation.
The stakes of more ambitious liberalization in services are high for a num-
ber of additional reasons:
Figure 13.14. Integration into EU Trade Networks and FDI in the Central and Eastern
European Countries
Integration into global production networks has been associated with
higher FDI in the CEECs
0
200
400
600
800
1,000
1,200
1,400
1,600
1,800
2,000
0246810121416
parts in percent of total exports, 1997
per capita FDI, 1990–98
Latvia
Romania
Bulgaria
Lithuania
Poland Slovenia
Slovak Rep.
Hungary
Estonia
Czech Rep.
Source: Authors’ calculations.

Part III: Trade, Competitiveness, and Investment 353
• Inefficient and costly services, provided mostly by the public sector, raise
the cost of MENA merchandise exports, limit attractiveness of investment,
while also impeding trade expansion within the region.
• With the right enabling environment in place, liberalization of key
services—such as, for example, telecommunications—may facilitate the
development of export capacity in other services—especially in tourism-
related services and the ICT sector.
• In addition to enhancing export capacity, liberalization in services can cre-
ate more investment opportunities for the domestic private sector, and
help attract more non-debt-creating foreign financing in the form of FDI.
Stepped up investment can offset the short-term adjustment costs stem-
ming from lower protection of import-competing industries.
Procompetitive reforms that facilitate entry by new firms can generate
employment opportunities for skilled and unskilled workers who are now
employed by governments in low-productivity jobs or in threatened import-
competing private manufacturing (Council on Foreign Relations, 2002). Because
services often cannot be traded, increasing access to service markets is likely to
require the entry of foreign competitors through FDI. This will not only lead to
the introduction of new technologies that improve efficiency and competitive-
ness, but also entail the hiring of domestic labor. By creating more opportunities
for the employment of skilled workers, services liberalization would also help
address the structural imbalances in MENA labor markets, especially the excep-
tionally high rates of unemployment for educated workers and the young.
Notes
1. There is some concern that the jobs created by FDI are also less secure, as financial crises or politi-
cal uncertainties can lead to capital flight and job closures. However, experience since the Asian crisis
has shown that FDI and the employment that it generated has been resilient to the financial crisis. FDI
in 2001 was at about the same level as in 1997, despite the collapse of private debt and equity flows.
See World Bank, 2002a.
2. The EU absorbs the lion’s share of CEECs exports of parts—about 79 percent in 1998. At the same
time, 82 per cent of CEECs imports of parts in 1998 originated in the EU. Germany has emerged as
the main trading partner among EU countries, as it takes almost half of CEECs exports of parts.
Among the CEECs, the Czech Republic,Hungary, Slovenia, Estonia, and Slovakia had the highest par-
ticipation of parts in their trade.
References
Agrawal, N. 2002. “The Impact of Liberalization on Employment and Earnings in Indonesia.” In F.
Iqbal and W.E. James (eds.) Deregulation and Development in Indonesia, Westport, Conn.: Praeger
Publishers. Arndt, S., and H. Kierzkowski. 2002. “Introduction.” In S. Arndt and H. Kierzkowski,
eds., Fragmentation and International Trade. Oxford and New York: Oxford University Press.
Council on Foreign Relations. 2002. Harnessing Trade for Development and Growth in the Middle East.
Washington, DC: Council on Foreign Relations.

Breaking the Barriers to Higher Economic Growth354
Currie, J., and A. Harrison. 1997. “Sharing the Costs: The Impact of Trade Reform on Capital and
Labor in Morocco.” Journal of Labor Economics 15: S44–S71.
Dessus, S., and A.Suwa-Eisenmann. 1998.“Trade Integration with Europe, Export Diversification,and
Economic Growth in Egypt.” Technical Paper No. 135. OECD Development Centre, Paris.
Dhonte, P., R. Bhattacharya, and T. Yousef. 2000. “Demographic Transition in the Middle East:
Implications for Growth, Employment, and Housing.” In Z. Iqbal, ed., Macroeconomic Issues and
Policies in the Middle East and North Africa. Washington, DC: International Monetary Fund.
Hummels, D., J. Ishii, and K.-M.Yi. 2001.“The Nature and Growth of Vertical Specialization in World
Trade .” Journal of International Economics 54(1): 75–96.
International Monetary Fund (IMF). 2001. Staff Report.Article IV: Consultation with Morocco. June.
Washington, DC: IMF.
Kaminski, B., and F. Ng. 2001.“Trade and Production Fragmentation: Central European Economies in
EU Networks of Production and Marketing.” Unpublished. World Bank, Washington, DC.
Nabli, M., and A. De Kleine. 2000. “Managing Global Integration in the Middle East and North Africa.”
In B. Hoekman and H. Kheir-El-Din, eds., Trade Policy Developments in the Middle East and North
Africa. Washington, DC: World Bank Institute.
Nabli, M., and M.-A. Veganzones-Varoudakis. 2002. “Exchange Rate Regime and Competitiveness of
Manufactured Exports: The Case of MENA Countries.” MENA Working Paper No. 27. World
Bank, Washington, DC.
Petri, A. 1997a. “Trade Strategies for the Southern Mediterranean.” Technical Paper No. 127. OECD
Development Centre, Paris.
———. 1997b. “The Case of Missing Foreign Investment in the Southern Mediterranean.” Technical
Paper No. 128. OECD Development Centre, Paris.
Rama, M. 2001. “Globalization and Workers in Developing Countries.” Unpublished. June. World
Bank, Washington, DC.
Revenga, A. 1997. “Employment and Wage Effects of Trade Liberalization: The Case of Mexican
Manufacturing.” Journal of Labor Economics 15: S20–S43.
Riordan, E.M., and others. 1997. “The World Economy and its Implications for the Middle East and
North Africa, 1995–2010.” In N. Shafik, ed., Prospects for MENA Economies: From Boom to Bust and
Back? New York: St. Martin’s Press.
Schiavo-Campo, S., G. de Tommaso, and A. Mukherje. 1997. “Government Employment and Pay in
Global Perspective: A Selective Synthesis of International Facts, Policies and Experience.”
Unpublished. World Bank, Washington, DC.
Sekkat, K., and A.Varoudakis. 2002.“The Impact of Trade and Exchange-rate Policy Reforms on North
African Manufactured Exports.” Development Policy Review 20(2).
Wood, A. 1997. “Openness and Wage Inequality in Developing Countries: The Latin American
Challenge to East Asian Conventional Wisdom.” The World Bank Economic Review 11: 33–57.
World Bank. 2001. Global Economic Prospects and the Developing Countries 2002. Washington, DC:
World Bank.
———. 2002a. Global Development Finance 2002, Washington, DC: World Bank.
———. 2002b. Transition: The First Ten Years. Washington, DC: World Bank.
Yeats, A. 1998. “Just How Big Is Global Production Sharing?” Policy Research Working Paper 1871.
January. World Bank, Washington, DC.
Yeats, A., and F. Ng. 2000. “Production Sharing in East Asia: Who Does What, for Whom and Why?”
Unpublished. World Bank, Washington, DC.
Zarrouk, J. 2000. “Para-Tariff Measures in Arab Countries.” In B.Hoekman and H. Kheir-El-Din, eds.,
Trade Policy Developments in the Middle East and North Africa. Washington, DC: World Bank
Institute.

355
Exchange Rate Management within
the Middle East and North Africa
Region
The Cost to Manufacturing Competitiveness
Mustapha Nabli
Jennifer Keller*
Marie-Ange Véganzonès†
14
Perhaps the greatest challenge currently facing the MENA region is the chal-
lenge of creating jobs. A recent World Bank report1estimates that some 100
million new jobs will need to be created in the next two decades to absorb
both the current unemployed as well as the rapidly expanding labor force—
more than doubling the current number of jobs today. If the region is to
ensure that this labor force has both sufficient employment opportunities and
the prospects for real wage growth, GDP growth in the region will need to
more than double from its average of 3 percent per year over the late 1990s to
6-7 percent a year for a sustained period.
Making this possible will require a fundamental transformation in the
region, from public sector-dominated economies to private sector-led
economies, open to international trade, with competitive private sector indus-
tries other than oil becoming the engine for growth and employment cre-
ation. Supporting a competitive private manufacturing sector in MENA will
*World Bank. † Centre d’Etudes et de Recherches sur le Developpement International. Centre National
de la Recherche Scientifique (CNRS), Université d’Auvergne. The authors are grateful to Pedro Alba,
Dipak Dasgupta, Tarik Yousef, and Paolo Zacchia for their comments and suggestions, and to Paul
Dyer and Adama Coulibaly for their analytical support.
Presented at the School of Oriental and African Studies (SOAS) Conference; London, July 2001.

Breaking the Barriers to Higher Economic Growth356
require actions on numerous fronts, including governance, trade, and mone-
tary and fiscal policy actions.
A requisite component of supporting private sector development in MENA
will be appropriate exchange rate management. Evaluations of the economic
policies in developing countries have demonstrated the importance of proper
management of the real exchange rate (RER) in a country's performance.
Empirical evidence consistently indicates that best economic performers are
those countries that have maintained an “appropriate” RER.2Countries that
have properly managed their RER (avoiding substantial RER appreciation)
have been more successful in promoting manufacturing exports.3They have
been more successful in attracting foreign direct investment.4And more gen-
erally, they have experienced higher growth.5
The Middle East and North Africa region has not followed the general
trend worldwide in its choice of exchange rate regimes.Although over the past
decades, countries have progressively adopted more flexible exchange rate
regimes, the majority of the economies in the Middle East and North Africa
continue to maintain de facto fixed exchange rate regimes. While about 65
percent of economies were operating under de facto fixed exchange rate
regimes in 1974 (within MENA, the proportion was somewhat higher, at 77
percent), by the end of the 1990s, only 42 percent of economies outside of
MENA had fixed exchange rate systems. Within MENA, however, that propor-
tion was 60 percent.
What impact, if any, have fixed exchange rate regimes had on RER mis-
alignment and, ultimately, the economic performances in the MENA region?
In this paper, we calculate the level of exchange rate misalignment across a
panel of countries over the 1970-1999 period, and show that the MENA
region has suffered from substantial exchange rate overvaluation which,
though highest over the 1970-1985 period, has persisted into the 1990s. It is
estimated that over the 1985-1999 period, the degree of exchange rate over-
valuation in MENA averaged some 22 percent, higher than for any other
region but CFA Africa. We then calculate the effect that overvaluation of the
exchange rate has had on the competitiveness of nonoil exports. It is estimat-
ed that the overvaluation of exchange rates has reduced the region’s manufac-
turing exports—as a percentage of GDP per year—by about 18 percent over
the 1970-1999 period.
Armed with this information, we discuss the empirical relationship
between the extent of exchange rate misalignment and the choice of exchange
rate regime. From our own calculations, the probability for fixed exchange
rates to become overvalued is substantially higher than for floating regimes,
and the probability for exchange rates under fixed regimes to become serious-
ly overvalued (in excess of 25 percent) is almost twice as high as for flexible
arrangements. We then discuss the reasons behind MENA’s continued reliance
on fixed exchange rate regimes. While the exchange rate choices in the region

Part III: Trade, Competitiveness, and Investment 357
are poorly explained by most traditional models of exchange rate choice, they
in part reflect the interests of the public sector as both producer of oil and
holder of debt, both of which make the government likely to favor fixed
exchange rates over floating ones.
Exchange Rate Management, Overvaluation, and the Costs
to Competitiveness
MENA’s exchange rate management has relied predominantly upon rigid
exchange rates, though not necessarily “officially” fixed exchange rates.6In
part, the reliance on fixed regimes was in response to the rapid inflation many
economies experienced over the late 1980s and early 1990s. Most economies
in MENA opted for a fixed exchange rate regime as the most effective strate-
gy for combating high inflation.
This adoption of fixed exchange rates was successful in contributing to
macroeconomic stability. However, once the immediate threats of high infla-
tion had been averted, only a handful of countries shifted to more flexible
exchange rate arrangements.
One of the arguments for countries that are not contending with high lev-
els of inflation— and thus not necessarily requiring a monetary anchor— to
adopt flexible exchange rate arrangements is that the real exchange rate is less
likely to become overvalued. Overvaluation can negatively affect a country’s
economic performance through a variety of channels. Overvaluation reduces
the profitability of tradables and, in turn, decreases exports. It leads to a
reduction in economic efficiency and a misallocation of resources. By increas-
ing uncertainty and raising the risk of macroeconomic collapse, misalignment
can hinder economic growth through a deterioration of domestic and foreign
confidence and investment, and can act as a catalyst for capital flight.7
Our own estimates suggest that fixed exchange rate regimes are substantial-
ly more likely to become overvalued than flexible regimes. We estimated the
level of exchange rate misalignment for a panel of countries, measured as the
percent difference between the real exchange rate (RER) and its equilibrium
value (ERER).8The RER was modeled following the approach used by
Edwards (1989) and extended by Elbadawi (1994) and Baffes, Elbadawi and
O’Connel (1997).9
From these misalignment estimates, we find that over the 1974-99 period,
the proportion of observations under fixed regimes which were even margin-
ally overvalued was 88 percent, versus 76 percent of flexible exchange rate
regimes. Moreover, the proportion of observations under fixed regimes that
were seriously overvalued (in excess of 25 percent) was 50 percent, almost
twice as high as for flexible regimes (28 percent).
This tendency for fixed exchange rates to become overvalued has impacted
the MENA region significantly, with substantial overvaluation of the real

Breaking the Barriers to Higher Economic Growth358
exchange rate experienced over the past three decades—around 29 percent
per year in the 1970s to the mid-1980s and 22 percent per year from the mid
1980s to 1999 (see table 14.1). In addition, this tendency has not significantly
decreased—contrary to the Latin American, African, or Asian economies in
our sample, which have in general chosen a more flexible exchange rate
regime—with regular devaluation of their currency—as well as more conse-
quent macroeconomic reforms.
Overvaluation and Manufactured Exports
What has been the cost of this greater degree of overvaluation in MENA to
total exports and manufactured exports? To determine, the following model
tests the effects of RER misalignment and volatility on the logarithm of total
and manufactured exports to GDP (log(Xt)):
ln(Xt) = c + b1.GDPgrTP i,t + b2.ln(TOTn i,t) + b3.ln(Inv i,t) + b4.
ln(Roads i,t) + b5.ln(H1 i,t) + b6.RerVol i,t + b7.ln( RerMis i,t) + εt.
where:
(i) GDPgrTP i,t = the rate of growth of GDP of country’s trading partners
(which can have a “pulling” effect on export growth);
(ii) ln(TOTn i,t) = logarithm of terms of trade (in which improvements can
increase the profitability of production for export);
(iii) ln(Inv i,t) = logarithm of investment/GDP (which increases the overall
production capacity, and thereby, export capacity);
Table 14.1. Average Misalignment and Volatilitya
1975/80– 84 (in % per year)bMisalignment Volatility
MENA 29 7.9
Latin America 20 11.2
Africa (CFA) 61 12.7
Africa (non CFA) 29 11.3
South Asia 43 13
East Asia 10 5.4
1985–99 ( in % per year)bMisalignment Volatility
MENA 22 12.4
Latin America 10 12.9
Africa (CFA) 28 14.5
Africa (non CFA) 13 16
South Asia 15 8.3
East Asia 5 8.6
a. Volatility is calculated as the coefficient of variation of the RER over a five-year period.
b. Depending on the countries.

Part III: Trade, Competitiveness, and Investment 359
(iv) ln(Roads i,t) = logarithm of lenth of roads (in km per km2)
(v) ln(H1 i,t) = logarithm of the average number of years of primary school-
ing of adult population.
{Both (iv) and (v) capturing the availability of core physical and human
infrastructures}
(vi) RERVols = volatility of the RER, as a measure of volatility of relative
prices10 (with RER11 volatility increasing the uncertainty of export prof-
itability)
(vii) RERMis = RER misalignment, as a measure of the distortion of relative
prices (overvaluation hampers competitiveness and diverts investment
out of more productive tradable goods sectors).12
The equation was estimated on our panel of 53 countries over 1970/80 to
99. The results from our estimation are shown in table 14.2.
Table 14.2. Estimation Results of the Exports Equations
Dependent variables: ln(Xmanuft) and ln(Xtott)
Variable Manufactured exports Total exports
GDPgrTP i,t 2.83 1.48
(1.9) (2.52)
ln(TOTn i,t)⫺1.4 0.1
(0.81) (2.49)
ln(Inv i,t) 0.87 0.30
(5.8) (8.69)
ln(Roads i,t) 0.08 0.10
(1.4) (3.48)
ln(H1 i,t) 1.92 0.26
(11.13) (5.66)
RerVol -0.27 -0.1
(0.80) (1.21)
Ln(RerMis) ⫺0.72 ⫺0.10
(5.75) (2.75)
Year 1974 0.25
(1.65)
Year 1975 0.34
(1.7)
Intercept -1.14
(9.05)
Adjusted R20.81 0.13
Fischer test 31.7 78.3
Haussmann test 12.4 0.20
Note: Student t statistics are within brackets. The numbers of observations are 816 and 964. Cointegration of the variables
was tested using Im, Pesaran, and Shin (1997) critical values of ADF tests in the case of heterogeneous panel data. (see
table A-2 in Annex 2). The equations were estimated by using the fixed effect method in the case of manufactured exports
and the random effect method in that of total exports..

Breaking the Barriers to Higher Economic Growth360
Our estimations confirm a significant negative impact of ER mismanage-
ment (in the form of overvaluation) on total and manufacturing export per-
formance. According to our estimations, a 10 percent increase in the level of
misalignment lowers the ratio of manufactured exports to GDP by 7.2 per-
cent, and the ratio of total exports to GDP by 1 percent. Overall, for the
MENA region, this RER overvaluation during the 1970-99 period reduced, on
average per year, manufacturing exports to GDP by 18 percent.
In terms of individual countries in MENA, losses have been important in
Jordan and Morocco in the 1970s and 1980s, because of the more diversified
export base of these economies. This is also the case of Tunisia in the 1990s,
despite a low level of overvaluation. In the major oil-exporting countries
(Algeria and Iran), losses appear small because of the low level of manufac-
tured exports (table 14.3). The large overvaluation, however, has certainly
contributed to the low diversification of these economies.
Overall, overvaluation represents a large cost to the region. Developing a
competitive private sector depends upon ensuring appropriate prices.
Profitability of production hinges on prices: prices of inputs that go into the
production process, and the price that can be obtained in the market for out-
put. Overvaluation damages competitiveness because it artificially alters the
price ratio between tradables and nontradables, and the region’s producers of
tradable goods find they are less able to compete with either imported goods
or with other countries’ exports.
Economies that, in reality, have cost advantages in labor and domestically
produced inputs begin altering their production processes and substituting
for capital equipment and imported inputs. And the greater the overvaluation
that takes place, the more difficult it becomes for otherwise competitive firms
Table 14.3. Cost of Overvaluation on Manufactured Exports
(Selected MENA Countries)
Algeria Egypt Iran
Mis (%) ExpM (%)aCostbMis (%) ExpM (%)aCostbMis (%) ExpM (%)aCostb
1970–79 79 3 ⫺1.7 15 27 ⫺2.9 42 3 ⫺0.9
1980–89 59 1.5 ⫺0.6 22 19 ⫺3244⫺0.7
1990–99 8 3.3 ⫺0.2 19 37 ⫺2.4 84 7 ⫺4
1970–99 49 2.6 ⫺0.8 15 27.6 ⫺2.7 49 4.5 ⫺1.8
Jordan Morocco Tunisia
Mis (%) ExpM (%)aCostbMis (%) ExpM (%)aCostbMis (%) ExpM (%)aCostb
1970–79 57 26 ⫺10.5 49 16 ⫺5.7 25
1980–89 31 43 ⫺9.4 8 39 ⫺2.4 3 49 ⫺1
1990–99 9 49 ⫺3.1 1 53 ⫺3.7 16 75 ⫺8.7
1970–99 25 39.1 ⫺7.7 21 36.1 ⫺3.9 9 49.6 ⫺4.8
a. ExpM: manufactured exports as percent of total exports.
b. Cost of overvaluation as percent of total exports.

Part III: Trade, Competitiveness, and Investment 361
to maintain their competitive edge, and the more it discourages new firms
from entering the market. At a time when encouraging an export-oriented,
nonoil private sector in MENA is critical, there is little room for excessive
exchange rate overvaluation.
Exchange Rate Regime Choice
While fixed exchange rates significantly increase the incidence of overvalua-
tion, and subsequently the cost to manufacturing exports, the question
emerges, why does the region continue to rely upon rigid exchange rate
arrangements? Is misalignment a justified cost that the MENA region must
pay to maintain stability in other macroeconomic fundamentals? In other
words, have MENA countries been choosing the appropriate exchange rate
arrangements?
The question is complex. A great deal of research has been devoted to
improving our understanding of how exchange rate regime choices are made,
with two major branches of research emerging. The first branch has produced
models of exchange rate choice based solely upon economic factors.13 In this
framework, the optimal regime could be determined as the one that mini-
mizes fluctuations in output, the price level, or some other macroeconomic
variable. The other branch of research has focused on the political economy.
While there have been several arguments within this general framework, they
have generally focused on the relationship between domestic political institu-
tions and exchange rate decisions.14
But while both economic and political economic models have substantial-
ly improved our understanding of exchange rate choice across countries, tra-
ditional models have been less successful in explaining the de facto15 exchange
rate regime decisions within MENA. Standard models of exchange rate choice
that incorporate both structural and political characteristics, when applied to
MENA economies, result in incorrectly predicted exchange rate regimes twice
as often as for non-MENA economies. In table 14.4, several conventional
models of exchange rate regime choice are outlined, which incorporate a
broad range of structural and political economy variables. In none of these
standard models are the exchange rate regimes adopted in MENA as well pre-
dicted as for other countries. In most cases, the difference in the proportion
of observations correctly predicted between non-MENA economies and
MENA economies is substantial.exchange rate regime decisions within MENA.Standard models ofexchange rate choice that incorporate both structural and politi
cal characteristics,when applied to MENA economies,result in incorrectly predicted exchange rate regimes twice as often as for non-MENA economies.In table 14.4,several conventional models ofexchange rate regime choice are outlined,which incorporate a broad range ofstructural and political economy variables.In none ofthese standard models are the exchange rate regimes adopted in MENA as well predicted as for other countries.In most cases,the difference in the proportion ofobservations correctly predicted between non-MENA economies and MENA economies is substantial.
Where do the standard models fail in predicting the exchange rate regimes
within MENA? The majority occur for fixed exchange rate observations.
Model A, for example, correctly predicts 31 percent of the MENA observa-
tions of floating regimes. On the other hand, it does not correctly predict any
fixed regime observations (that compares with non-MENA economies, with
55 percent correct predictions for floating regime observations and 10 percent

Breaking the Barriers to Higher Economic Growth362
correct predictions for fixed regime observations). Model B correctly predicts
31 percent of MENA’s floating regime observations, but only 6 percent of its
fixed regime observations (compared with 57 percent and 21 percent for non-
MENA observations of floating and fixed regimes, respectively). Overall, and
especially for the cases where fixed regimes have been adopted, according to
traditional models for exchange rate choice, the MENA region isn’t getting it
right.
Why the MENA countries, which would be predicted along structural vari-
ables lines to adopt flexible regimes, have chosen to maintain fixed exchange
rate arrangements may have something to do with the public sector’s person-
al interests.
Recent research advances the proposition that countries with high
unhedged foreign currency-denominated debt, and a correspondingly high
exchange rate risk exposure (such as economies in the MENA region) have an
incentive to peg (see Calvo and Reinhard, 2000; Hausman, Panizza, and Stein,
2000). For governments with high publicly guaranteed external debt, fixed
regimes may be preferred as a means to better control the direction of tempo-
rary currency appreciation (or depreciation), since by allowing overvaluation
of the currency, it permits at least a temporary reduction in foreign-
Table 14.4. Correct Predictions of Exchange Rate Regime under Alternate Model Specificationsa
Non-MENA MENA
Independent % correct % correct
Model variablesbNo. obs predictionscNo. obs. Predictions
A Size, Openness, CVEX,
CVGDP 895 37 106 16
B Size, Openness, CVEX,
CVGDP, CAPCONT,
RESERVES, INFLAT 895 43 106 19
C POLSTABLE, HISTGROW,
Size, Openness, CVEX,
CVGDP, CAPCONT,
RESERVES, INFLAT 674 41 91 15
D HISTGROW,
Size, Openness, CVEX,
CVGDP, CAPCONT,
RESERVES, INFLAT 779 42 97 17
a. Dependent variable exchange rate regime arrangement (1=fixed; 0=floating). Probit estimation for the 1985-2000 period.
b. SIZE = lagged ratio of GDP in constant US$ to US. OPENNESS = average ratio of (imports + exports) / GDP for five years prior to observation
year. CVEX = Coefficient of variation of exports for period (t-5) to (t-1). CVGDP = Coefficient of variation of GDP for period (t-5) to (t-1). CAPCONT =
dummy variable for whether country had controls on movement of capital. RESERVES = lagged ratio of international reserves to imports. INFLAT
= average inflation rate over five year period prior to observation year. POLSTABLE = lagged index of political stability (ICRG). HISTGROW = aver-
age lagged log growth, period (t-5) to (t-1). All estimations performed over the 1985-2000 period.
c. Exchange rate choice was modeled through probit estimation, with fixed exchange rates having a value of 1, and flexible regimes a value of 0.
Probit estimations will produce a probability of adopting a fixed regime, ranging from 0 to 1. A fixed exchange rate regime was said to be correct-
ly predicted if the estimated probability of adopting a fixed exchange rate regime was greater than 0.6. A flexible exchange rate regime was said
to be correctly predicted if the estimated probability of adopting a fixed regime was less than 0.4. Estimated probabilities between 0.4 and 0.6
were categorized as not definitively predicted by the model.

Part III: Trade, Competitiveness, and Investment 363
denominated debt payments (albeit while sacrificing the competitiveness of
some national industries in the process). Continued build-up of further debt,
at the same time, allows an overvalued fixed exchange rate to be maintained,
since it permits the government to continue to borrow foreign currency to
sustain the current account deficit and meet the excess demand.
In addition, the public sector in many MENA economies (and most oil-
based economies) has interest in its own business, namely the export of natu-
ral resources. Oil still represents the major source of income and a dominant
source of foreign exchange for the oil-producing economies of MENA. Shares
of oil in current total exports ranged from a high of more than 95 percent in
Yemen, to a low of less than 1 percent in Jordan and Morocco in at the end of
the 1990s. In 10 of the 16 countries of MENA listed below in figure 14.1, oil
revenues account for more than 70 percent of total export revenues, and in 12
of the 16, it accounts for more than 45 percent of export revenue.
What is striking about oil economies, in general, is the reliance they have
maintained on fixed exchange rate regimes. For economies in which over 50
percent of export revenue emanated from natural resource extraction in 1997,
some 83 percent had fixed exchange rate arrangements in place. That com-
pares with fixed exchanges rate regimes being adopted in only 38 percent of
economies in which oil represented less than 25 percent of exports.
Conventional economic models approach the desire for fixed or flexible
regimes by agglomerating the interests of the tradable goods sector together.
The problem with this approach is that the various industries within the
exporting sector are assumed to have concurrent interests. There is reason to
believe this may not always be the case.
oil exports as a share of
total exports (%)
Algeria
Bahrain
Egypt
Iran
Jordan
Kuwait
Lebanon
Libya
Morocco
Oman
Qatar
Saudi Arabia
Syria
Tuni si a
United Arab Emirates
Yem e n
0
25
50
75
100
Figure 14.1. Share of Oil in Total Exports, 1999
Source: United Nations CONTRADE data, as reported in WITS trade data warehouse.

Breaking the Barriers to Higher Economic Growth364
The manufacturing sector, with relatively elastic worldwide demand, is
likely to be more dependent upon competitiveness (and thus, more likely to
lobby for floating exchange rates) than is the natural resources sector.Floating
exchange rate mechanisms may entail greater short-term volatility, but better
prevent long-term appreciation of the exchange rate. As a result, it allows
exporters to better achieve external competitiveness through efficiency, by
leaving to the market forces of supply and demand the exchange rate
determination.
On the other hand, the natural resource extraction sector is assumed to face
more inelastic demand and depend less upon imported raw materials. As a
result, currency appreciation has weaker impact on profits (and depending
upon the elasticity of demand, may result in even higher profits). Thus, the
natural resource-exporting sector is more likely to prioritize stability in
exchange rates, and the potential gains from currency overvaluation (under
control of policy makers under a fixed regime, at least for the short-term),
than the potential for competitiveness.
In an attempt to better understand the exchange rate regime decisions in
MENA, we estimate exchange rate choice according to traditional models of
exchange rate choice for a large sample of economies, but include public exter-
nal debt and the divergent interests of the oil versus manufacturing sectors.
The Empirical Results. Several standard exchange rate regime choice models are
augmented with proxies for the importance of oil revenues to the public sec-
tor, the importance of external debt payments to the public sector, and the
ability of the manufacturing sector to lobby the government for flexible
regimes. Debt payments are measured by the ratio of public external debt to
GDP, lagged one period. The importance of the oil sector is captured as the
value of oil exports to total exports, lagged one period. The lobby power of the
manufacturing sector is measured as an interactive between the size of the
manufacturing export sector in GDP and the concentration of manufacturing
exports within the sector, with the expectation that larger or more concentrat-
ed manufacturing sectors (in terms of the industries represented) are more
able to have an effective voice in influencing the government’s exchange rate
regime choice. The manufacturing lobby variable is also lagged a period.
Other variables include political stability, size, degree of openness, meas-
ures of external and domestic variability, inflation, reserves, and capital con-
trols. Annex 3 provides a more detailed explanation of the variables included
in the estimations, and the expectations about their influence on exchange
rate regime choice.
To investigate whether the incorporation of additional explanatory vari-
ables significantly improves the predictive power of traditional models on the
exchange rate regime choices within MENA, standard probit models of
exchange rate choice are compared with augmented models (table 14.5),

Part III: Trade, Competitiveness, and Investment 365
incorporating variables for external debt and for the interests of the oil sector
versus the lobby power of the manufacturing sector. To avoid simultaneity
problems, the oil sector and manufacturing sector variables are included in
separate estimations.
By then comparing the predicted exchange rates from the models to the
actual exchange rate choices in MENA and outside of MENA, versus models
incorporating the public sector’s and the manufacturing sector’s interests, it is
possible to measure the improvement of fit.
Table 14.5. Standard and Augmented Models of Exchange Rate Choice
Dependent variable: Exchange rate regime (1=fixed; 0=floating)
Independent Model 1 Model 2 Model 3
variables (a) (b) (c) (a) (b) (c) (a) (b) (c)
Size ⫺4.97 ⫺2.91 ⫺4.32 ⫺5.58 ⫺3.13 ⫺4.75 ⫺4.76 ⫺2.48 ⫺4.17
(⫺4.75) (⫺2.75) (⫺4.05) (⫺4.80) (⫺2.74) (⫺4.12) (⫺3.76) (⫺2.23) (⫺3.48)
Open 0.01 0.02 0.14 0.01 0.01 0.01 0.04 ⫺0.00 ⫺0.01
(5.08) (5.40) (5.23) (3.17) (3.05) (3.01) (0.92) (⫺0.15) (⫺0.31)
CVEX ⫺0.12 0.25 ⫺0.29 0.09 0.43 ⫺0.09 0.12 0.99 0.39
(⫺0.21) (0.44) (⫺0.50) (0.16) (0.72) (⫺0.14) (0.14) (1.08) (0.43)
CVGDP 0.37 0.30 0.40 0.81 0.62 0.80 0.71 0.56 0.89
(0.79) (0.63) (0.84) (1.65) (1.24) (1.61) (1.04) (0.80) (1.30)
CAPCONT 0.23 0.23 0.20 0.44 0.13 0.02
(2.35) (2.37) (2.04) (0.38) (1.09) (0.17)
RESERVES ⫺0.15 0.53 0.10 0.34 0.84 0.56
(⫺0.22) (0.78) (0.15) (0.44) (1.05) (0.71)
INFLAT ⫺0.71 ⫺0.80 ⫺0.80 ⫺0.81 ⫺1.01 ⫺0.96
(⫺5.33) (⫺5.73) (⫺5.79) (⫺5.02) (⫺6.00) (⫺5.71)
POLSTABLE 0.00 0.01 0.01
(⫺0.40) (2.13) (1.63)
HISTGROW 0.04 ⫺0.01 ⫺0.03
(⫺2.12) (⫺0.30) (⫺1.63)
OILX 0.50 0.57 0.83
(2.90) (3.22) (4.07)
MANLOBBY ⫺1.53 ⫺1.60 ⫺2.16
(⫺6.97) (⫺7.15) (⫺7.06)
PUBEXDEBT 0.00 0.08 0.06 0.14 0.27 0.39
(0.05) (1.08) (0.89) (1.98) (2.28) (3.16)
No. obs 1001 1001 1001 1001 1001 1001 765 765 765
MENA
Proportion right 16 35 16 19 37 32 15 44 26
Proportion wrong 16 19 12 10 14 11 9 13 9
No prediction 67 45 73 71 49 57 76 43 65
NON-MENA
Proportion right 36 49 41 43 57 46 41 61 52
Proportion wrong 13 15 16 11 17 14 13 14 15
No prediction 50 36 43 46 27 41 46 25 34

Breaking the Barriers to Higher Economic Growth366
The incorporation of debt, combined with a measure of the interests of the
oil sector or the lobby power of the manufacturing sector, significantly
increases the predictive power of the exchange rate models under all three
model classifications, not only for MENA economies,but for economies over-
all. The improvement in predictive power of these augmented models explain
MENA’s exchange rate arrangements is substantial, especially so for models
incorporating the lobby power of the manufacturing sector.
Conclusions
Against the overall trend throughout the world, the majority of the MENA
region has continued to maintain de facto fixed exchange rate arrangements.
Empirical analysis suggests that fixed exchange rates are associated with
greater levels of exchange rate misalignment in the form overvaluation, which,
in turn, reduce competitiveness for nonoil exporters. In MENA, manufactur-
ing exports—as a percentage of GDP per year—were reduced by some 18 per-
cent over the 1970-99 period as a result of the region’s substantial overvalua-
tion of its currency. At a time when developing a strong, export-oriented pri-
vate sector outside of oil is critical in MENA, there is no room for excessive
currency overvaluation.
We find that MENA’s choice of exchange rate regime—predominantly
leaning toward rigid exchange rate arrangements—is less a reflection of struc-
tural characteristics of the economies than it is a reflection of the political
economy. With a large public sector, which has individual interests as produc-
er of oil and holder of external debt, the interests of the economy are often at
odds with the interests of the political economy. If the manufacturing sector
has enough power, however, it may lobby effectively for flexible exchange rate
regimes.
Notes
1. Unlocking the Employment Potential in the Middle East and North Africa: Toward a New Social
Contract; World Bank; 2003.
2. That is, its equilibrium real exchange rate value (ERER). See Williamson,1985; Harberger, 1986;and
Collins, 1997.
3. See Balassa, 1990, for empirical evidence among Sub-Saharan African economies.
4. Goldberg 1993; Goldberg and Kolstad 1995; and Cushman 1985; 1988.
5. Edwards, 1988; Cottani, Cavallo and Khan, 1990; and Ghura and Grennes, 1993.
6. In many cases of exchange rate systems officially classified as flexible, for example, there has been
considerable “management” of the exchange rate. De facto exchange rate regimes, according to Levy-
Yeyati/Sturzenegger (2000), are determined by looking at the actual behavior of three variables close-
ly related to exchange rate behavior: exchange rate volatility, volatility of exchange rate changes, and
volatility of reserves. External liabilities and government deposits were netted out from the reserves
data, in order to consider only changes with a counterpart in monetary aggregates, an especially

Part III: Trade, Competitiveness, and Investment 367
important correction for both oil-producing countries and countries with large privatization pro-
grams. The LYS dataset of exchange rate regimes has subsequently been amended. This paper’s analy-
sis reflects use of the earlier de facto classification system, and will need to be revised.
7. In addition to misalignment, variability of the RER has been found to have negative consequences
on growth (Ghura and Grennes, 1993; Grobar, 1993; Cushman, 1993; and Gagnon, 1993).
8. This estimate represents a new contribution to the study of exchange rate policy in MENA coun-
tries, since previous studies are sparse (Domac and Shabsigh, 1999; Mongardini, 1998; Sundararajan,
Lazare and Williams, 1999).
9. This model estimates the RER as a function of both “fundamental” factors in the medium to long
term (terms of trade, investment, capital flows, and trade openness) and less persistent factors in the
short term (macroeconomic policies, nominal devaluations, and others). Following the estimation of
the RER, the ERER could be computed. Using the estimated RER, the ERER is computed by eliminat-
ing the effects of transitory variables, and using estimates of “sustainable” values of the fundamentals.
10. Calculated as the coefficient of variation of the RER over an eight-year period. To compute this
indicator, some economists use more or less sophisticated regression techniques, such as the variance
of the residual of the regression of the RER on a time trend, or an ARCH modelization RER behavior.
However, from an empirical point of view, all these various measures are highly correlated and the
standard deviation or the coefficient of variation measures perform as well as more sophisticated ones
(see Kenen and Rodrik, 1986 or Grobar, 1993).
11. In addition to misalignment, variability of the RER has been found to have negative consequences
on growth (Ghura and Grennes ,1993; Grobar, 1993; Cushman, 1993; and Gagnon, 1993)
12. RER misalignment can also disrupt exports by increasing RER uncertainty. Our measure of RER
misalignment comes from our estimation of the ERER (see previous section).
13. The earliest literature focused strictly on the structural characteristics of the economy,such as eco-
nomic openness, country size, and labor mobility. From these characteristics, the optimal exchange
rate arrangement is determined (Dreyer, 1978; Heller, 1978; Holden et al., 1979; Wickman, 1985;
Savvides, 1990). Later research in this branch of the literature has focused on country-specific shocks
emanating from both the international and domestic community (Fischer, 1977. Savvides, 1990).
14. Within that concentration, arguments have centered around policy discipline and credibility
(Kydland and Prescott, 1977; Barro and Gordon, 1983; Flood and Isard, 1989; Giavazzi and Pagano,
1988; Rogoff, 1985), policy-making capability (Eichengreen, 1992; Simmons, 1994), and constraints
on future governments (Rogowski, 1987; Edwards, 1996).
15. As opposed to the de jure classification of exchange rate regimes, we have examined the de facto
classification of exchange rate regimes (Levy-Yeyati and Sturzenegger, 2000), which looks at the actu-
al behavior of three relevant variables to exchange rate behavior: exchange rate volatility, volatility of
reserves, and volatility of exchange rate changes. This new classification of exchange rate regimes
refines the analysis substantially.
References
Balassa, B. 1990. “Incentive Policies and Export performance in Sub-Sahara Africa.” World
Development 18 (3).
Baffes, J., I. A. Elbadawi, and S.A. O’Connell. 1997. “Single-Equation Estimation of the Equilibrium
Real Exchange Rate.” Mimeo. Swarthmore College, Swarthmore, PA. May.
Barro, Robert J. and David B. Gordon. 1983.“A Positive Theory of Monetary Policy in a Natural Rate
Model.” Journal of Political Economy 104: 589–610.
Beveridge, S., and C. R. Nelson. 1981. “A New Approach to Decomposition of Economic Time Series
into Permanent and Transitory Components with Particular Attention to Measurement of the
‘Business Cycle.’” Journal of Monetary Economics (7).

Breaking the Barriers to Higher Economic Growth368
Caramazza, Francesco,and Jahangir Aziz. 1998.“Fixed or Flexible? Getting the Exchange Rate Right in
the 1990s,” Economic Issues 13. International Monetary Fund, Washington, DC.
Clark, P. B. and R. MacDonald. 1998.“Exchange Rates and Economic Fundamentals: A Methodological
Comparison of BEERs and FEERs.” IMF Working Paper WP/98/67. International Monetary
Fund,Washington, DC. May.
Cordon, M. 1984. “Booming Sector and the Dutch Disease Economies: Survey and Consolidation.”
Oxford Economic Papers 36 (3): 359–380.
Cottani, J. A., D. F. Cavallo, and M. S. Khan. 1990. “Real Exchange Rate Behaviour and Economic
Performance in LDCs.” Economic Development and Cultural Change 39.
Cuddington, J. T., and L. A. Winters. 1987. “The Beveridge-Nelson Decomposition of Economic Time
Series—A Quick Computational Method.” Journal of Monetary Economics (19).
Cushman, D. O. 1985. “Real Exchange Rate Risk, Expectations, and the Level of Direct Investment.”
The Review of Economics and Statistics 67: 297–308.
———. 1988. “Exchange Rate Uncertainty and Foreign Direct Investment in United States.”
Weltwirtschaftliches Archiv, Review of World Economics 124 (2): 322–336.
———. 1993.“The Effects of Real Exchange Rate Risk on International Trade.” Journal of International
Economics 15.
Dreyer, J. S. 1978. “Determinants of Exchange Rate Regimes for Currencies of Developing Countries:
Some Preliminary Results.” World Development 6: 437–45.
Domac, I., and G. Shabsigh. 1999. “Real Exchange Rate Behaviour and Economic Growth: Evidences
from Egypt, Jordan, Morocco and Tunisia.” IMF Working Paper WP/99/40. International
Monetary Fund, Washington, DC. March.
Edwards, S. 1988. Exchange Rate Misalignment in Developing Countries. Baltimore, MD: The Johns
Hopkins University Press.
———. 1989. Real Exchange Rate, Devaluation and Adjustment: Exchange Rate Policy in Developing
Countries. Cambridge, MA: MIT Press.
———. 1992. “Oil and the Exchange Rates in Venezuela: Historical Experience and Policy Options.”
Mimeo. UCLA and NBER.
———. 1994. “Real and Monetary Determinants of Real Exchange Rate Behaviour: Theory and
Evidence from Developing Countries.” In Estimating Equilibrium Exchange Rates, ed. J.
Williamson. Institute of International Economics, Washington, DC.
———. 1996.“The Determinants of the Choice Between Fixed and Flexible Exchange Rate Regimes.”
NBER Working Paper 5756. NBER, Cambridge, MA.
Elbadawi, I. A. 1994. “Estimating Long-Run Equilibrium Real Exchange Rates.” In Estimating
Equilibrium Exchange Rates, ed. J. Williamson. Institute of International Economics, Washington
DC.
———. 2002.“Real Exchange Rates Policy and Non-Traditional Exports in Developing Countries.” In
Non-Traditional Export Promotion in Africa: Experiences and Issues, ed. G. Helleiner. Palgrave, New
Yo r k .
Engel, R. F., and C W. J. Granger. 1987. “Cointegration and Error: Representation, Estimation and
Testing.” Econometrica (35).
Fischer, S. 1977. “Stability and Exchange Rate Systems in a Monetarist Model of the Balance of
Payments.” In The Political Economy of Monetary Reform, ed. R. Aliber. London: Macmillan Press.
Frieden, Jeffry A. 1991. “Invested Interests: The Politics of National Economic Policies in a World of
Global Finance.” International Organization 45: 425–51.
Funke, N. 1996.“Vulnerability of Fixed Exchange Rate Regimes.” Economic Studies 26: 1. Organisation
for Economic Co-operation and Development, Paris, France.
Gagnon, J. E. 1993. “Exchange Rate Variability and the Level of International Trade.” Journal of
International Economics.

Part III: Trade, Competitiveness, and Investment 369
Gelb, A., et al. 1988. Oil Windfalls: Blessing or Curse? New York, NY: Oxford University Press for the
World Bank.
Ghura, D., and T. J. Grennes. 1993. “The Real Exchange Rate and Macroeconomic Performances in
Sub-Saharan Africa.” Journal of Development Economics 42.
Goldberg, L. S. 1993. “Exchange Rate and Investment in United States Industry.” The Review of
Economics and Statistics 75 (4). November.
Goldberg, L. S., and C. D. Kolstad. 1995. “Foreign Direct Investment, Exchange Rate Variability and
Demand Uncertainty.” International Economic Review 39 (4). November.
Gowa, J. 1983. Closing the Gold Window. Ithaca, NY: Cornell University Press.
Grobar, L. M. 1993. “The Effect of Real Exchange Rate Uncertainty on LDC Manufactured Exports.”
Journal of Development Economics 14.
Harberger, A. 1986. “Economic Adjustment and the Real Exchange Rate.” In Economic Adjustment
Exchange Rates in Developing Countries, ed. S. Edwards and L. Ahamed. Chicago, IL: University of
Chicago Press.
Heller, H. R. 1978. “Determinants of Exchange Rate Practices.” Journal of Money, Credit, and Banking
10: 308–21.
Holden, P., M. Holden, and E. Suss. 1979, “The Determinants of Exchange Rate Flexibility: An
Empirical Investigation.” Review of Economics and Statistics 61: 327–333.
Im, K. S., M. H. Pesaran, and Y. Shin. 1997. “Testing for Unit Roots in Heterogeneous Panels.” DAE
Working Paper Amalgamated Series No. 9526. Cambridge, U.K.: University of Cambridge.
Levy-Yeyati, E., and F. Sturzenegger. 2000.“Exchange Rate Regimes and Economic Performance.” Draft
mimeo. Buenos Aires, Argentina: Universidad Torcuato Di Tella.
Kenen, P. B., and R.Rodrik. 1996.“Measuring and Analysing the Effects of Short Term Volatility in Real
Exchange Rates.” Review of Economics and Statistics 68.
Kydland,F.,and E. Prescott. 1977.“Rules Rather than Discretion: The Inconsistency of Optimal Plans.”
Journal of Political Economy 85: 473–91.
Mongardini, J. 1998. “Estimating Egypt Equilibrium Real Exchange Rates.” IMF Working
PaperWP/98/5. International Monetary Fund, Washington, DC. January.
Pinto, B.1990. “Black Market Premia, Exchange Rate Unification and Inflation in Sub-Saharan Africa.”
The World Bank Economic Review 3 (3).
Razin, O., and S. M. A. Collins. 1997. “Real Exchange Rate Misalignments and Growth.” NBER
Working Paper No. 6174. NBER, Cambridge, MA. September.
Rizzo, Jean-Marc. 1998. “The Economic Determinants of the Choice of an Exchange Rate Regime: A
Probit Analysis.” Economic Letters 59: 283–87.
Rodrik, Dani. 1994. “Getting Intervention Right; How South Korea and Taiwan Grew Rich.” NBER
Working Paper Series No. 4964. NBER, Cambridge, MA. November.
Rogoff, K. 1985. “The Optimal Degree of Commitment to an Intermediate Monetary Target.”
Quarterly Journal of Economics 100: 1169–90.
Rogowski, R. 1987.“Trade and the Variety of Democratic Institutions.” International Organization 41
(2): 203–33.
Savvides, A. 1990.“Real Exchange Variability and the Choice of Exchange Rate Regime by Developing
Countries.” Journal of International Money and Finance 9: 440–54.
Stein, J. L. 1994. “The Natural Exchange Rate of the U.S. Dollar and Determinants of Capital Flows.”
In Estimating Equilibrium Exchange Rates, ed. J.Williamson. Institute for International Economics,
Washington, DC.
Sundararajan, V., M. Lazare, and S. Williams. 1999. “Exchange Rate Unification, the Equilibrium Real
Exchange Rate, and Choices of Exchange Rate Regime: The Case of Islamic Republic of Iran.”IMF
Working Paper WP/99/15. International Monetary Fund, Washington, DC. January.

Breaking the Barriers to Higher Economic Growth370
Wickman, P. 1985. “The Choice of Exchange Rate Regime in Developing Countries: A Survey of the
Literature.” IMF Staff Papers 32: 248–88. International Monetary Fund, Washington, DC.
Williamson, J. 1985. “The Exchange Rate System.”Policy Analyses in International Economics, No. 5.
Institute of International Economics, Washington, DC.
———. 1994. “Estimates of FEERs.” In Estimating Equilibrium Exchange Rates, ed. J. Williamson.
Institute for International Economics, Washington, DC.

Part III: Trade, Competitiveness, and Investment 371
Annex 14.1. Calculating RER Misalignment
Modeling the Long-Run Equilibrium of the RER
The long-run equation explaining the RER behavior is based on Edwards
(1994), which has developed a dynamic model of RER determination for a
small, open economy with a single nominal exchange rate system. The model
allows for both real and nominal factors to play a role in the short run. In the
long run, only real factors—the “fundamentals”—influence the ERER. In our
case, the long-run relation is specified as follows1:
ln (et) = c +a1.ln(Invt) + a2.ln(Opent) + a3.ln(TOTt) + a4.Capinft+ εt. (1)
with:
- et= RER. This indicator is used as a proxy for the ratio of the price of
nontradable goods (PDt) to the price of tradable goods (Pwt.Et0, Et.
being the nominal ER in local currency/US$ and Pwt the world
prices);
- Invt= Investment ratio to GDP;
- Opent= Indicator of trade openness, measured as the sum of import and
export divided by GDP;
- TOTt= External terms of trade, measured as the ratio of export to import
prices (in dollars);
- Capinft= capital inflows calculated as the net change in reserves minus the
trade balance scaled by GDP2
c= intercept, a1to a4= parameters, t = time index and (t= error term.
Following Edwards (1989), we assume that—in the long term—an
increase in the investment rate results in an augmentation in the demand and
in the relative price of nontradables—thus appreciating the real exchange
rate. This assumption implies, however, that investment is predominantly
constituted of nontradable products (such as services and construction) and
not of tradable goods (such as equipment). It can also be due to the multi-
plier effect of the investment, which increases the aggregate demand for non-
tradable products, principally. Conversely, the RER is positively affected by
trade restrictions, for which the ratio of imports plus exports to GDP is a
proxy. The impact of the terms of trade on the RER is more ambiguous, since
there are two opposite effects: An increase in the relative price of export
goods to import goods leads to an appreciation of the RER if the income
effect, which results in higher demand for nontradables, dominates the sub-
stitution effect—associated with a decline in the relative cost of imported
intermediate goods used in the production process of nontradables. Finally,
an increase in capital inflows, either officially or not, involves stronger
demand for both tradable and nontradable goods. They, therefore, lead to a

Breaking the Barriers to Higher Economic Growth372
higher relative price of nontradables, and conversely appreciates the RER, as
needed for domestic resources to be diverted toward production in the non-
tradable sector to meet increased demand.
A complementary equation has been estimated in order to take more into
consideration the characteristics of some MENA countries. The idea is that in
a certain number of countries, including Egypt, debt relief should have led to
an appreciation of the ERER. For this purpose we have added to equation (1)
the ratio of the debt service to total external trade (imports + exports,
DebtServ).
ln (et) = c +a1.ln(Invt) + a2.ln(Opent) + a3.ln(TOTt) +
a4.Capinft+ a5. ln(DebtServt) + εt.(2)
The existence of these long-term relationships implies that variables of
Equations (1) and (2) are cointegrated. It is therefore required to determine
the order of integration of the series. Table A-1 in annex 2 provides the results
of the Augmented-Dickey-Fuller (ADF) tests of the data for our sample of 53
countries during 1970-80 (depending on the countries) to 1997. We used the
Im, Pesaran, and Shin (1997) methodology—which provides critical values of
ADF tests in the case of heterogeneous panel data. The results indicate that the
series are stationary at either the 1 percent or 5 percent levels, when allowed
to run Equations (1) and (2). We then used the Engel and Granger (1991)
method to test for cointegration between the variables of Equations (1) and
(2). Cointegration tests have been based on the residuals of the two equations.
ADF tests conclude, still using Im, Pesaran, and Shin (1997) critical values,
that residuals are stationary.
Hence, Equations (1) and (2) describe the long-run relationship
between RER and a number of fundamental variables. The equations were
estimated on an unbalanced panel of 51 countries, among which 17 are
African countries (8 CFA and 9 non CFA), 13 Latin America countries, 10
Asian countries, 10 MENA countries, plus one country3(see annex 2 for
the list of countries). The results of the regressions—using the White esti-
mator to correct for the heteroscedasticity bias—are presented in table A1.
The equations were estimated by using the fixed effect methodology.4The
estimated regressions explain a fairly large amount of the observed varia-
tion of the RER.
Estimated relationships between RER and its fundamentals are consistent
with theory: An increase in investment and in capital income, or an improve-
ment of the terms of trade, results in a RER appreciation, which indicates, in
the latter case, that the income effect dominates the substitution effect.
Conversely, the opening of the economy and the increase in the debt service
lead to a RER depreciation.

Part III: Trade, Competitiveness, and Investment 373
Calculating RER misalignment
RER misalignment is measured as the ratio of the RER and its equilibrium
value (ERER):
MIS = (RER / ERER)
Thus, when the RER is higher than its equilibrium value (when the curren-
cy is overvalued), misalignment takes a value greater than 1. But when the
RER is lower than its equilibrium value (undervalued), misalignment takes a
value less than 1.
The estimations of the long-term relationship between the RER and its
fundamental determinants have been used to compute the ERER.5For this
purpose, the “sustainable” or “equilibrium” values of the fundamental vari-
ables had to be assessed. The idea is that the deviation of the fundamental
variables from their “equilibrium”—in addition to the variations of the short-
term economic policy variables (see the estimation of the error correction
model through Equation 4 in annex 3)—leads to a misalignment of the RER.
The “permanent” values of the four fundamental variables, that is, Invt,Opent,
TOTt, and Capinft,were computed using moving averages of the series over a
three years period. This simple method was possible because our series were
stationary.6
Following this methodology, excessive trade protection, unexpected appre-
ciation of the terms of trade or increase in investment and capital flows, in
Table 14A1. Estimation Results of the Cointegrating Equations (1) and (2)
Dependent variable: ln(et)
Variable Eq (1) Eq (2)
ln(Invt)0.16 0.10
(3.6) (2.6)
ln(Opent)-0.64 -0.72
(12.5) (13.7)
ln(TOTt)0.14 0.21
(3.1) (4.41)
Capinft0.34 0.44
(3.6) (4.3)
ln(DebtServt) -0.18
(9.9)
Adjusted R20.61 0.65
Fischer test 28.3 25.7
Haussmann test 28 15.5
Source: Authors’estimations.
Note: Student tstatistics are within brackets. The number of observations used in Equations (1) and (2) are respec-
tively 1183 and 1062. Data have been compiled from WDI, GDF, GDN, and LDB World Bank databases.

Breaking the Barriers to Higher Economic Growth374
comparison to the “normal” or long-term trend in the economy, lead to an
overvaluation of the RER. It can also be shown from the estimation of the
error correction model that in the short run, nominal devaluations (Dev), the
black market premium (BMP), and inflation (Infl) explain the deviations of
the RER from the ERER.

Part III: Trade, Competitiveness, and Investment 375
Annex 14.2. List of Countries in Estimations of Exports/GDP
AFRICA
MENA CFA NonCFA ASIA LATIN AMERICA
Bahrain (BHR) Burkina Faso (BFA) Botswana (BWA) South East Asia Argentina (ARG)
Algeria (DZA) Côte d’Ivoire (CIV) Gambia, The (GMB) Indonesia (IDN) Bolivia (BOL)
Egypt (EGY) Gabon (GAB) Kenya (KEN) Korea, Rep.(KOR) Brazil (BRA)
Iran (IRN) Cameroon (CMR) Madagascar (MDG) Malaysia (MYS) Chile (CHL)
Jordan (JOR) Gambia, The (GMB) Mozambique (MOZ) Philippines (PHL) Colombia (COL)
Kuwait (KWT ) Niger NER) Mauritius (MUS) Thailand (THA) Costa Rica (CRI)
Malta (MLT) Senegal (SEN) Malawi (MWI) South Asia Ecuador (ECU)
Morocco (MAR) Togo ( TGO) Nigeria (NGA) Bangladesh (BGD) Guatemala (GTM)
Syria (SYR) Tanzania (TZA) India (IND) Mexico (MEX)
Tunisia (TUN) China (CHN) Peru (PER)
OTHER COUNTRIES Sri Lanka (LKA) Paraguay (PRY)
Israel (ISR) Pakistan (PAK) Uruguay (URY)
Venezuela, RB (VEN)

Breaking the Barriers to Higher Economic Growth376
Annex 14.3. Modeling Exchange Rate Choice
In this paper, we empirically tested the hypothesis that for each economy, the
public sector’s determination of exchange rate regime is a decision-making
process that weighs three factors: the overall structural characteristics of the
economy, its personal interests in minimizing its current external debt pay-
ments and maximizing natural resources revenue (both better achieved under
fixed exchange rates), and the degree to which lobby pressures by the manu-
facturing sector can sway the public sector. The greater the lobby power of the
manufacturing sector,the more likely the public sector will be to adopt a float-
ing exchange rate regime.
A central problem throughout the literature in the testing of models of
exchange rate regime choice has been the utilization of de jure exchange rate
regimes. Most empirical analysis has used the published exchange rate
regimes from the IMF’s Exchange Arrangements and Exchange Restrictions:
Annual Report. The report classifies economies, according to their exchange
rate arrangement, into three broad groups: (i) those whose currency is pegged
to a single currency or currency composite; (ii) those whose exchange rate sys-
tem has limited flexibility, in terms of a single currency or group of curren-
cies; and (iii) those with more flexible exchange rate systems. Unfortunately,
in many countries, exchange rates that are officially flexible have been subject
to considerable official “management.” Indeed, as Calvo and Reinhart (2000)
and others have emphasized, many countries that claim to have floating
exchange rates do not in practice allow the rate to float freely, but use interest
rate and intervention policies to affect its behavior.
Within the approaches pioneered by Holden, Honden, and Suss (1979) to
characterize the de facto exchange rate regimes economies employ, a major
contribution was provided by Levy-Yeyati and Sturzenegger (2000, hereafter
LYS), who developed a database of exchange rate classifications by looking at
the actual behavior of the main relevant variables, as opposed to the tradition-
al classification compiled by the IMF. The LYS classification is based on three
variables closely related to exchange rate behavior to determine the de facto
exchange rate regime: exchange rate volatility, volatility of exchange rate
changes, and volatility of reserves. The empirical results from this paper were
based on the original reclassification of exchange rates by LYS for their 2000
paper.That dataset has recently been amended, and the results from this paper
will have to be revised.
Structural variables in the analysis include many of the variables suggested
by the optimal currency areas (OCA) literature. Two of these are the country
size and the degree of openness of the economy, with the expectation that
smaller countries that are more open tend to favor fixed exchange rate
regimes. In our estimations, size (SIZE) is measured by the log of GDP, rela-
tive to the United States, lagged one period, and openness (OPEN) is defined

Part III: Trade, Competitiveness, and Investment 377
as the average share of exports to GDP for the five- year period prior to the
observation year. In addition, the OCA literature would suggest that the vul-
nerability of an economy’s output to shocks affects its choice of exchange rate
regime. Thus, we have included two indices of the extent of external and
domestic shock variability. The first, CVEX, is the coefficient of variation of
real exports for the five-year period prior to the observation year. Likewise,
CVGDP was constructed as the coefficient of variation of real GDP for the
five-year period prior to the observation year. Each of these indicators was
assembled from World Bank data.
Several control variables suggested by the literature were also included in
our analysis. One, capital controls (CAPCONT) is suggested in the literature
on capital liberalization and financial openness, with countries with capital
controls more likely to have fixed exchange rates. If a government controls the
movement of international capital, it can insulate itself from the internation-
al price movements and will be more able to maintain a pegged regime. A
dummy variable was included if the country in question had controls on the
international movement of capital, and the data were assembled from the
IMF’s Annual Report on Exchange Arrangements and Exchange Restrictions.
An additional control variable suggested by the political economy literature
(Edwards 1996) is the historical rate of inflation. Theory would predict that
countries with a history of rapid inflation will have a lower probability of
maintaining a pegged regime, and will thus tend to favor the adoption of a
more flexible system. In most other studies, the history of inflation is meas-
ured as the average rate of inflation for some period prior to the observation
year. We constructed a slightly different index, INFLAT. This was constructed
by taking the average rate of inflation over the five-year period prior to the
observation, and determining the proportion of years in which inflation
exceeded 30 percent.This was meant to better capture the variable which deci-
sion makers might consider important in determining their exchange rate
regime.xx One might question whether a policy maker makes a serious dis-
tinction between whether his economy experienced average inflation rates of
140 percent a year versus 100 percent a year. Even taking logs of past inflation
rates would retain these ordinal differences in the inflation rate, which from
the point of view of a policy maker may lose their significance at some level of
inflation. On the other hand, the past probability of the economy facing
episodes of high or runaway inflation (above 20 percent a year, in our estima-
tions) may play a serious role in considerations for exchange rate policy in the
future.
In addition to these variables, we have incorporated variables that proxy
the public sector’s weight put on personal interests (in minimizing external
debt payments and maximizing natural resource export revenue) versus its
ability to be lobbied by the manufacturing sector.For the reasons given above,
countries that have higher levels of external debt will have a greater tendency

Breaking the Barriers to Higher Economic Growth378
to opt for a fixed exchange rate regime over a floating one, since there is an
economic pay-off to allowing the currency to become overvalued, in terms of
lower foreign currency debt payments. In our estimations, we included PUB-
XDEBT, which is the lagged value of public and publicly guaranteed external
debt to GDP.
To measure the public sector’s interest in petroleum revenues, we include
the size of the petroleum sector (measured by its share of total exports), OILX,
to measure the public sector’s personal exporting interests, again lagged one
period, and assembled using World Bank data. Lobby power in the manufac-
turing sector, MANLOBBY, is measured as an interactive between the share of
manufacturing exports in GDP and the concentration of manufacturing
exports among the top three products, constructed using data from the
United Nations’ COMTRADE (which allows for analysis of trade by commod-
ity). Like the size of the petroleum sector in exports, the manufacturing lobby
variable was lagged one period. Because of significant reporting errors on the
part of some economies (particularly the GCC) in terms of exports, exports
of oil, exports of manufacturing goods, and total exports (to determine shares
in total exports) by each economy were recomputed by aggregating world
imports from each economy in the various sectors as a share of total world
imports from the economy in question.
Our estimations were performed over the 1985-1999 period.
Appendix Notes
1. The short-run dynamic of the RER has also been estimated through an error correction model
Equation (4) in Annex 3. Results are shown in table A.3.
2. An increase in net capital inflows may result from (i) an autonomous augmentation in foreign aid,
foreign voluntary lending, or FDI; (ii) an increase in borrowing due to the removal of domestic capi-
tal controls; (iii) a fall in the world interest rates; or (iv) an increase in public borrowing to finance the
fiscal deficit.
3. The countries have been selected on the criteria of their level of income per capita. To preserve a
kind of coherence in the sample, we have chosen (most of the time) intermediate income countries
that would be comparable to the ones of the MENA region.
4. This is supported by the data, as shown by the Fischer test of equality of intercepts across countries,
and is preferable to the random effect methodology, as revealed by the value of the Haussmann test.
5. In the rest of the document, it is Equation (2) that has to be used to calculate the misalignment.
6. Other trials consist of an “economic” determination of these “sustainable” levels, inspired by
Edwards, 1988, which, for example, takes as sustainable value for openness the average of the three
higher values of the variable, or in the case of capital inflows, zero, if the rate of growth of the econo-
my is inferior to the international interest rate, which means, in this case, that that borrowing is not
sustainable—did not give better results as far as misalignment is concerned. They are not presented
here.
Our calculations of misalignment appear, however, in some cases, to underestimate the level of mis-
alignment as generally perceived in the different countries. We thereby have adjusted our estimates by
scaling them up, according to the difference between our calculations of ERER and its level in periods
in which the actual RER was considered to be at the equilibrium. The RER was considered to be close

Part III: Trade, Competitiveness, and Investment 379
to its equilibrium in periods following devaluations and structural adjustment, where the balance of
payment was also close to the equilibrium. For example, it has been considered that RER equilibrium
took place in 1989 in the case of Morocco. This period was 1991 and 1994-95 for Algeria; 1993-94 for
Egypt; 1995 for Iran; 1992 for Jordan; and 1980, 1994, and 1997 for Tunisia.
Some more sophisticated calculations exist, when a variable has a unit root, in using time series tech-
niques introduced by Nelson (1981), where variables are decomposed into a random walk with a drift
and a stationary component. Unlike the trend stationary model-based decomposition, this technique
allows the steady-state growth path of the series to shift over time. Fluctuations around the shifting
permanent path reflects cyclical effects.

381
How Does Exchange Rate Policy
Affect Manufactured Exports in
MENA Countries?
Mustapha Kamel Nabli
Marie-Ange Véganzonès-Varoudakis*
15
Recent assessments of economic policies and performance in developing
countries have underlined the crucial issue of the management of the real
exchange rate (RER). It has been shown that the best performers are countries
that have maintained an “appropriate” RER—that is, one close to the equilib-
rium real exchange rate (ERER) (Williamson, 1985; Harberger, 1986; Razin
and Collins, 1997). In particular, all countries that have been successful in pro-
moting manufactured exports have avoided real exchange rate overvaluation.1
In fact, RER misalignment—especially overvaluation—is damaging to eco-
nomic performance because it decreases the profitability of production and
the export of tradable goods. In this way, RER misalignment leads to a reduc-
tion in economic efficiency and a misallocation of resources. In addition, by
increasing uncertainty and raising the risk of macroeconomic collapse, RER
misalignment can also hinder growth by deterring domestic and foreign
investment and contributing to capital flight. These negative effects of mis-
alignment on growth and export performance have been shown by Edwards
(1988), Cottani et al. (1990), and Ghura and Grennes (1993) for different
groups of developing countries.
*Centre d'Etudes et de Recherches sur le Développement International and Centre National de la
Recherche Scientifique, Université d’Auvergne, France.
Published in Applied Economics (2004), vol. 36, no. 19. Reprinted by permission from Taylor and
Francis.

Breaking the Barriers to Higher Economic Growth382
In addition to misalignment, inconsistencies among macroeconomic,
trade, and exchange rate policies increase the variability of the RER—which
in turn can affect growth. Higher RER volatility sends confusing signals to
economic agents. It increases the uncertainty of long-term investments and
the profitability of producing tradable goods. The sensitivity of export per-
formance to RER volatility has been highlighted in the case of various
economies by Ghura and Grennes (1993), Grobar (1993), Cushman (1993),
and Gagnon (1993).
The harmful effect of RER misalignment on exports by the MENA
economies is well confirmed by our study. We show that during the past three
decades MENA countries experienced substantial RER misalignment, with a
net tendency to overvaluation of their RER. This had a significant negative
impact on the export growth of manufactured products, though the effect was
less significant when total exports are considered. This appears to have result-
ed in slower economic growth, as manufactured products exports have
become a major factor of economic growth in developing economies, many of
which have now successfully entered world markets.2
The findings bring new empirical evidence to the subject of misalignment
and of export growth in the case of MENA countries, on which little previous
work has been undertaken. Our results were obtained through the estimation
of an export equation on a panel of 53 developing countries (see annex 1),
among which 10 are MENA economies. The calculations cover the period
1970–80 to 1999, during which tremendous changes in trade and exchange
rate policies are observed.
The first step was to provide an accurate measure of the gap (or misalign-
ment) between RER and its equilibrium level (ERER). The estimates of ERER
and of RER misalignment are based on a reduced form approach. RER behav-
ior is modeled using an equation that includes both the role of fundamental
factors in the medium to long term (for example,, terms of trade, investment,
capital flows, trade openness), and the less persistent impact of short-term
variables (for example,, macroeconomic policies, nominal devaluations). The
ERER is then computed, using this equation, by eliminating the effect of tran-
sitory variables and using estimates of sustainable or long-term values of the
fundamental variables. This approach, initiated by Edwards (1989), has been
extended by Elbadawi (1994) and Baffes et al. (1997).
The use of this approach also represents a new contribution to the study
of exchange rate policy in MENA economies, since the few previous studies
were generally based on a time series approach (Mongardini, 1998; Domac
and Shabsigh, 1999; Sorsa, 1999; Sundararajan, Lazare, and Williams, 1999;
Achy, 2001).3In addition to using panel data estimations techniques in com-
parison to times series econometrics,4our calculations allow some compar-
ative analysis among the different regions, as well as among the MENA coun-
tries themselves.

Part III: Trade, Competitiveness, and Investment 383
The paper is organized as follows. In the second section, panel data calcu-
lations of the RER’s longterm equilibrium are presented. In the third section,
the misalignment and volatility of the MENA countries’ RER are discussed
and compared to those of other regions. The fourth section presents our esti-
mation of the impact of RER misalignment and of RER volatility on the
export performance of the economies. It is illustrated that the misalignment
has had a negative influence on the MENA countries’ manufactured exports.
The fifth section concludes.
Modeling the Longterm Equilibrium of the RER
The longterm equation explaining RER behavior is based on Edwards (1994),
who developed a dynamic model of RER determination for a small, open
economy with a single nominal exchange rate system. The model allows for
both real and nominal factors to play a role in the short run. In the long run,
only real factors—the fundamentals—influence the ERER. In the present case,
the longterm relationship is specified as follows:5
ln (et) = c + a1.ln (Inv i,t) + a2.ln (Open i,t) + a3.ln (TOT i,t) +
a4.Capinf i,t + a5.ln (DebtServt)+ εi,t.(1)
with:
et= bilateral RER between the country concerned and the United
States, measured as the ratio of the consumption price index in the
country (PDt) to the wholesale price index in the USA (Pwt), mul-
tiplied by the nominal exchange rate in local currency / US$ (Et.).
These price indices are used, respectively, as proxies of the price of
nontradable goods (PDt) and the price of tradable goods (Pwt.Et,);
RER t= (PDt) / (Pwt.Et,);
- Invt= investment ratio to GDP;
- Opent= indicator of trade openness, measured as the sum of imports and
exports, divided by GDP.
As an improved measure of trade openness, we have substituted for Opent,
a policy-induced trade openness indicator TPt, which consists of adjusting
Opentfor the “natural trade openness” of the economy, a metric comprising
the size of the country and the distance from markets (see Frankel and Romer,
1999). This is Equation (1’).
TOTt= external terms of trade, measured as the ratio of export to import
prices (in dollars);
- Capinft = capital inflows, calculated as the net change in reserves minus the
trade balance scaled by GDP;6

Breaking the Barriers to Higher Economic Growth384
- DebtServt= debt service to total exports;
- c = intercept, a1to a4= parameters, t = time index, and εt= error
term.
Following Edwards (1989), it is assumed that,in the long term, an increase in
the investment rate (Invt) results in an increase in the demand for and relative
price of nontradable products—thus increasing Keep as appreciating, JK the
real exchange rate. This assumption implies that as the investment rate grows,
investment is increasingly constituted of nontradable products (for example,,
services and construction), and relatively less of tradable goods (for example,,
equipment). This effect can also be due to the multiplier effect of the invest-
ment, which raises the aggregated demand mainly for nontradable products.
The RER is positively affected by trade restrictions, which implies a nega-
tive sign on the coefficient on the proxy for trade openness, measured as the
ratio of imports plus exports to GDP (Opent). The same negative sign is
expected for the improved measure of policy-induced trade openness (TPt).
The impact of the terms of trade (TOTt) on the RER is more ambiguous,
since there are two opposite effects. That is, an increase in the relative price of
exported goods to imported goods leads to an appreciation of the RER only if
the income effect (which results in higher demand for nontradables) dominates
the substitution effect (which is associated with a decline in the relative cost of
imported intermediate goods used in the production process of nontradables).
An increase in capital inflows (Capinft) involves stronger demand for both
tradable and nontradable goods. Increased inflows, therefore, lead to a higher
relative price of nontradables, and conversely, appreciate the RER—as is nec-
essary for domestic resources to be diverted toward production in the non-
tradable sector to meet increased demand. On the other hand, a rise in the
debt service (DebtServt), which captures the important impact of debt relief in
many MENA countries, contributes to depreciating the RER.
The existence of this long-term relationship implies that the variables of
Equations (1) and (1’) are cointegrated. It is therefore necessary to determine
the order of integration of the series. Table 15A.1 in Annex 15.2 provides the
results of the Augmented-Dickey-Fuller (ADF) tests of the data for our sam-
ple of 53 developing countries7over the time period. The Im et al. (1997)
methodology, which provides critical values of ADF tests in the case of hetero-
geneous panel data, is used. The results indicate that the series are stationary
at either the 1 percent or 5 percent level, which allows Equations (1) and (1’)
to be run.
Hence, Equations (1) and (1’) describe the longterm relationship between
RER and a number of fundamental variables. The equations were estimated
for the panel of 53 developing countries, of which 10 are MENA economies.8
The results of the regressions—using the White estimator to correct for the
heteroscedasticity bias—are presented in Table 15.1. The equations were esti-

Part III: Trade, Competitiveness, and Investment 385
mated using the fixed-effect methodology. The estimated regressions explain
a fairly large amount of the observed variation of the RER.
Estimated relationships between RER and its fundamentals are consistent
with theory (Edwards, 1989): an increase in investment and in capital income,
or an improvement of the terms of trade, result in a RER appreciation. This
indicates, in the latter case, that the income effect dominates the substitution
effect. Conversely, the opening of the economy and the increase in the debt
service lead to a RER depreciation.
RER Misalignment
The misalignment (MIS) of the real exchange rate (RER) is measured as the
percent difference between the RER and its equilibrium value (ERER):
MIS = (RER / ERER) –1
The estimations of the long-term relationship between the RER and its
fundamental determinants have been used to compute the ERER based on
Equation (1’). For this purpose, the sustainable, or equilibrium, values of the
fundamental variables had to be assessed. The idea is that the deviation of the
fundamental variables from their equilibrium—in addition to variations in
Table 15.1. Estimation Results of the Cointegrating Equations (1) and (1’)
Dependant variable: ln(et)
Variable Eq (1) Eq (1’)
ln(Invt)0.09 0.11
(2.0) (2.3)
ln(Opent)-0.71
(14.4)
ln(TPt) -0.32
(6.7)
ln(TOTt)0.23 0.24
(4.9) (4.8)
Capinft0.45 0.5
(4.5) (4.7)
ln(DebtServt) -0.18 -0.14
(9.9) (7.5)
Adjusted R20.63 0.55
Fischer test 25.9 19.1
Haussmann test 20 18.8
Source: Authors’ estimations.
Note: Student tstatistics are within brackets. The number of observations used in eq (1) and (2) are, respectively,
1092 and 1080. Data have been compiled from World Development Indicators, Global Development Finance,
Global Development Network, and Live Data Base (World Bank).

Breaking the Barriers to Higher Economic Growth386
short-term economic policy variables (see the estimation of the error correc-
tion model through equation (A3-1) in annex 3) — leads to a misalignment
of the RER. The permanent values of the five fundamental variables—Invt,
Opent, TOTt, Capinft, ,DebtServt—were computed using moving averages of
the series over a three-year period. This simple method was possible because
the series are stationary.10
Following this methodology, excessive trade protection, unexpected appre-
ciation of the terms of trade, an increase in investment and in capital flows, or
a reduction of the debt service,—in comparison to the “normal” or long-term
trend in the economy, lead to an overvaluation of the RER. It can also be shown
from the estimation of the error correction model (table A-3 in annex 3) that,
in the short term, nominal devaluations (Dev), black market premiums
(BMP), and inflation (Infl) explain the deviations of the RER from the ERER.
The results confirm that, during the past three decades, MENA countries
have experienced substantial overvaluation of their RER: 29 percent per year
on average from the mid-1970s to the mid-1980s, and 22 percent from the
mid-1980s to 1999 (see table 15.2). In general, the extent of overvaluation has
not significantly decreased—unlike in Latin America, Africa, or Asia.
Overvaluation remains higher in MENA than in the other regions, except
Communauté Francophone d’Afrique (CFA) Africa (see next section for the
experience of individual MENA countries).
On the other hand, exchange rate volatility has generally been lower in the
MENA economies than in the other regions of our sample (see table 2). This
can surely be explained by the less flexible exchange rate regimes of the MENA
countries. This conclusion should, however, be nuanced. In particular during
the second subperiod (1985–99), the volatility of the exchange rate in the
Table 15.2. Average Misalignment and Volatility
1975–80 to 1984 (percent per year) Misalignment Volatility
MENA 29 7.9
Latin America 20 11.2
Africa (CFA) 61 12.7
Africa (non-CFA) 29 11.3
South Asia 43 13
East Asia 10 5.4
1985 to 1999 (percent per year) Misalignment Volatility
MENA 22 12.4
Latin America 10 12.9
Africa (CFA) 28 14.5
Africa (non-CFA) 13 16
South Asia 15 8.3
East Asia 5 8.6
Source: Author’s calculations.

Part III: Trade, Competitiveness, and Investment 387
MENA region is not very different from that in Latin America, and is higher
than that in Asia.
RER Management and Manufactured Export Performance
Manufactured Exports in the MENA Countries
Table 15.3 presents data on the performance of some MENA countries in terms
of manufactured exports. Over the last three decades, the success of these coun-
tries in increasing exports and in diversifying their economies varied widely.
Tunisia and Jordan have been the most successful in increasing their export
of manufactures. Tunisian manufactured exports rose, on average, from 24.5
percent of total exports in the 1970s to 75 percent in the 1990s (4.6 percent to
21.2 percent of GDP). If Jordan’s performance seems less impressive than
Tunisia’s, its increase in manufactured exports as a percentage of GDP is, in
fact, comparable to Tunisia’s (although Jordan’s level of exports to GDP
remains lower). Morocco also significantly increased its exports during the
1970s and 1980s, but these gains slowed in the 1990s.
In Egypt, manufactured exports increased slowly throughout the period,
growing from 27.1 percent of total exports in the 1970s to only 36.6 percent
in the 1990s (and, in fact, decreasing from 3.1 to 2.4 percent of GDP). The two
major oil-exporting countries, Algeria and Iran, showed the most dismal per-
formance, with manufactured exports remaining negligible throughout the
period.
Modeling Exports of Manufactured Products
Overvaluation should have had a cost for the MENA countries that we would
like to quantify. As seen previously, manufactured exports should have suf-
fered from RER misalignment and volatility. The following model is used to
test for these effects:
ln(Xt) = c +b1.GDPgrTP i,t + b2.ln(TOTn i,t) + b3.ln(Inv i,t) +
b4.ln(Roads i,t) + b5.ln(H1 i,t)+ b6.RERVol i,t + (2)
b7.ln(RERMis i,t)+ εt.
Table 15.3. Average Manufactured Exports of Selected MENA Countries
Algeria Egypt Iran Jordan Morocco Tunisia
%X %GDP %X %GDP %X %GDP %X %GDP %X %GDP %X %GDP
1970–79 3.0 0.6 27.1 3.1 2.9 0.6 25.8 1.9 16.0 2.1 24.5 4.6
1980–89 1.5 0.3 19.2 1.5 4.0 0.3 42.7 5.4 39.4 6.0 49.4 11.7
1990–99 3.3 0.8 36.6 2.4 6.6 1.5 48.9 9.5 52.9 7.5 74.9 21.2
Source: Authors’ calculations.
Note: For the first subperiod, four values were missing for Iran (1970, 71, 72, and 73). For the third sub-period, two values were missing for Iran
(1991 and 92) and one for Jordan (1996).

Breaking the Barriers to Higher Economic Growth388
The model explains exports to GDP in logarithmic form by:
– the GDP growth rate of the trade partners (GDPgrTPi,t), which can have a
“pulling” role in exports
– the logarithm of the terms of trade ln( TOTn i,t), the improvement of which
increases the profitability of production for export
– the logarithm of the ratio of investment to GDP [ln(Inv i,t)], which is con-
ducive to an increase in overall production capacity,and thus to an increase
in export capacity
– the availability of core infrastructure, measured by the logarithm of the
length of roads ln(Roads i,t) in km per km2, as well as the availability of
human capital, approximated by the logarithm of the average number of
years of primary schooling of the adult population [ln(H1 i,t)]
– the volatility of relative prices, approximated by the volatility of the RER
(RERVol) and calculated as the coefficient of variation of the RER over a
five- year period.11 RER volatility increases uncertainty regarding the prof-
itability of producing tradable goods
– the distortions in relative prices, as measured by RER misalignment
(RERMis), where overvaluation hampers competitiveness and diverts
investment out of the more productive tradable goods sectors. RER mis-
alignment can also disrupt exports by increasing RER uncertainty.
In addition, the heterogeneity of the sample is controlled for by consider-
ing country dummy variables. These variables reflect differences in the quali-
ty of institutions or different endowments in natural resources—which can be
the origin of large discrepancies in the natural propensity to export. The
hypothesis of country dummy variables is supported by the data for manufac-
tured exports12 (see table 15.4 below). A time dummy variable is also intro-
duced for the years 1974–75, corresponding to the first oil shock.
Econometric Results
Equation (2) was estimated on the panel of 53 developing countries from
1970–80 to 1999, for both total (Xtott) and manufactured exports (Xmanuft).
Manufactured exports are more sensitive to competitiveness problems. They
should consequently be more negatively influenced by RER overvaluation.
Because of missing data for some variables, the model was finally estimated on
two unbalanced panels of, respectively, 943 and 837 observations.13 Results
are shown in table 15.4.
The estimations confirm the negative impact of exchange rate misalign-
ment on total and manufactured export. The coefficient is rather strong in the
case of manufactured exports (–0.72). It remains significant for total exports
(–0.10). The weaker elasticity in the latter case can be explained by the fact
that total exports include products that are less sensitive to competitiveness,

Part III: Trade, Competitiveness, and Investment 389
such as oil products and other primary goods, which are often owned and
managed by governments.
For the MENA region as a whole, exchange rate policy helps explain losses
in competitiveness and in manufactured exports. During the whole period,
RER overvaluation reduced the ratio of manufactured exports to GDP by 18
percent per year on average. Manufactured exports, which averaged 4.4 per-
cent of GDP from 1970 to 1999, could have reached 5.2 percent of GDP if no
overvaluation had taken place. These losses were more concentrated in the
1970s and 1980s than in the 1990sdue to the higher overvaluation of the cur-
rencies during those two subperiods.
Some countries with a more diversified export base, such as Jordan and
Morocco (see table 15.5) faced high losses during the 1970s and 1980s.
Because of its high level of manufactured exports, Tunisia still incurred a large
loss during the 1990s, despite a relatively low level of misalignment. But in
Table 15.4. Estimation Results of the Exports Equations
Dependent variables: ln(Xmanuft) and ln(Xtott)
Manufactured exports Total eExports
Variable ln(Xmanuft)ln(Xtott)
GDPgrTP i,t 2.83 1.48
(1.9) (2.52)
ln(TOTn I,t)⫺1.4 0.1
(0.81) (2.49)
ln(Inv i,t) 0.87 0.30
(5.8) (8.69)
ln(Roads i,t) 0.08 0.10
(1.4) (3.48)
ln(H1 i,t) 1.92 0.26
(11.13) (5.66)
RERVol ⫺0.27 ⫺0.1
(0.80) (1.21)
Ln(RERMis) ⫺0.72 ⫺0.10
(5.75) (2.75)
Year 1974 0.25
(1.65)
Year 1975 0.34
(1.7)
Intercept ⫺1.14
(9.05)
Adjusted R20.81 0.13
Fischer test 31.7 78.3
Haussmann test 12.4 0.20
Source: Authors’ estimations.
Note: Student tstatistics are within brackets. The number of observations used in the regressions are, respectively,
816 and 964. Data have been compiled from World Development Indicators, Global Development Finance, Global
Development Network, and Live Data Base (World Bank).

Breaking the Barriers to Higher Economic Growth390
these countries, RER misalignment either declined significantly or remained
low during the 1990s, as the countries saw a continuous rise in diversification
of their manufactured exports.
In the major oil-exporting countries, Iran and Algeria,the large overvalu-
ation of the currency certainly contributed to the low diversification of their
exports away from oil. But the losses, as measured here, appear small, given
the low initial level of manufactured exports, which can be explained by the
structure of their economies.
The estimations fail, however, to show a significant impact of RER volatil-
ity on either manufactured or total exports of the countries. This finding does
not confirm the empirical results of several studies of different groups of
economies (see in particular Ghura and Grennes, 1993; Grobar, 1993;
Cushman, 1993; Gagnon 1993).
The results also highlight that total, as well as manufactured, exports are
positively influenced by the GDP growth rate of the trade partners, the ratio
of investment to GDP, and the physical and human infrastructure (measured
respectively by the length of the road network and by the level of primary edu-
cation of the population).14
The pulling effect of the trade partners’ GDP growth rate is particularly
strong in the case of manufactured exports (elasticity of 2.8 against 1.5 for
total exports). This result goes in the direction expected. The income elastici-
ty is higher for manufactured products than for other products in the
economy.
The same conclusions can be drawn for human capital, which improves the
profitability of investment and the competitiveness of manufactured exports
much more than in other sectors of the economy. The effect of primary edu-
cation is particularly strong (elasticity of 1.9, compared to 0.26 for total
Table 15.5. Cost of Misalignment on Manufactured Exports, Selected MENA Countries
DZA EGY IRN
ExpMaMis CostbExpMaMis CostbExpMaMis Costb
1970–79 3 1.79 ⫺1.7 27 1.15 ⫺2.9 3 1.42 ⫺0.9
1980–89 1.5 1.59 ⫺0.6 19 1.22 ⫺3.0 4 1.24 ⫺0.7
1990–99 3.3 1.08 ⫺0.2 37 1.09 ⫺2.4 7 1.84 ⫺4.0
1970–99 2.6 1.49 ⫺0.8 27.6 1.15 ⫺2.7 4.5 1.49 ⫺1.8
JOR MAR TUN
ExpMaMis CostbExpMaMis CostbExpMaMis Costb
1970-79 26 1.57 ⫺10.5 16 1.49 ⫺5.7 25
1980–89 43 1.31 ⫺9.4 39 1.08 ⫺2.4 49 1.03 ⫺1.0
1990–99 49 1.09 ⫺3.1 53 1.10 ⫺3.7 75 1.16 ⫺8.7
1970–99 39.1 1.25 ⫺7.7 36.1 1.21 ⫺3.9 49.6 1.09 ⫺4.8
a. ExpM = manufactured exports as percent of total exports.
b. Cost = cost of overvaluation as percent of total exports.

Part III: Trade, Competitiveness, and Investment 391
exports). This makes education a key variable for the competitiveness of the
manufacturing sector in the developing world.
Manufactured exports, however, are not sensitive to improvement in terms
of trade, which were supposed to provide an incentive to produce for the
export sector. This could reflect the fact that the terms of trade measures
include prices of exports of agriculture and mining products, which are not
included in manufacturing.
Conclusion
In this paper, it is shown that during the 1970s and 1980s MENA countries
were characterized by a significant overvaluation of their currencies.
Overvaluation decreased in the 1990s, probably due to some degree of flexibi-
lization of the exchange rate regime, or to better macroeconomic manage-
ment. Misalignment remained, however, higher than in other developing
countries (but CFA Africa). This may be explained by the delay in adopting
more flexible exchange rates and in reforming the economy.
In fact, although many economies have progressively adopted more flexible
exchange rate regimes—leading to a better management of their RER—most
MENA countries are still implementing fixed or adjustable-peg exchange rate
policies. In addition, if the shift toward a more open economy has begun in
several countries of the region, this process needs to be deepened, since the
current situation reduces manufactured exports competitiveness and weakens
the incentive for exporters to increase their penetration of foreign markets.
This is partly the case for oil-exporting countries, which have failed to address
the volatility of their economies, and in which diversification of exports is still
very low. But this lack of trade openness also explains the low diversification
of other MENA countries in the 1970s and 1980s.
The study also illustrates that overvaluation has had a cost for the region in
terms of competitiveness. Manufactured exports, in particular, have been
affected by the overvaluation of the exchange rate. These findings confirm
recent assessments of economic policies and performance in developing
countries, which underscore the crucial issue of the management of the real
effective exchange rate. The results corroborate the findings of Edwards
(1988); Balassa (1990) and Cottani et al. (1990) for different groups of devel-
oping countries.
Notes
1. See, for example,Balassa (1990) and Reinhardt (1995) for empirical evidence in both developed and
developing countries.
2. In fact, export diversification—through promotion of manufactured exports—is an important fac-
tor of sustained growth for different reasons. First, income elasticity of demand is higher for manufac-

Breaking the Barriers to Higher Economic Growth392
tured goods than for primary products. In this way, growth in foreign income is expected to increase
the growth prospects of country’s manufactured exports. Second, price elasticity of both demand and
supply is presumed to be higher for manufactured goods than for primary commodities. This implies
a stabilizing effect on the terms of trade and more stable growth of exports over time. Third, develop-
ment of the manufacturing sector involves substantial prospects for dynamic productivity gains
through economies of scale, learning effects, and externalities among firms and industries. See
Nishimizu and Robinson (1986) for cross-country evidence at a two-digit industry level of positive
correlation between export growth and total factor productivity (TFP) changes.
3. See Sekkat and Varoudakis (2002) for a panel data approach to assessing the misalignment of North
African countries.
4. The comparative advantage of panel data regressions compared to time series estimations can be
seen first, in the double dimension of the sample (time series-cross section), which improves estimates
by adding information, and second, in the country dummy variables, which generally ask for an
important degree of freedom, and which improve the results of the estimations.
5. The short-term dynamic of the RER has also been estimated through an error correction model
(equation (A3-1) in annex 3. Results are shown in table A.3.
6. An increase in net capital inflows may result from (i) an autonomous augmentation in foreign aid,
foreign voluntary lending, or foreign direct investment (FDI); (ii) an increase in borrowing due to the
removal of domestic capital controls; (iii) a fall in world interest rates; or (iv) an increase in public bor-
rowing to finance the fiscal deficit.
7. of which 19 are African countries (8 Communauté Francophone d’Afrique and 11 non-CFA), 13 are
Latin America countries, 10 are Asian countries, and 11 are MENA countries.
8. The countries were selected based on level of income per capita. To preserve a kind of coherence of
the sample, we generally chose intermediate-income countries so they could be compared to countries
in the MENA region.
9. The use of the fixed-effect methodology is supported by the data, as shown by the Fischer test of
equality of intercepts across countries, and is preferable to the random effect methodology,as revealed
by the value of the Haussmann test (see table 15.1).
10. Other attempts have consisted of an “economic” determination of these “sustainable” levels
(inspired by Edwards, 1988).We took as a sustainable value for openness the average of the three high-
er values of the variable, and in the case of capital inflows, zero, if the rate of growth of the economy
was inferior to the international interest rate-which meant that borrowing was not sustainable. Those
calculations are not shown here because they did not give better results as far as misalignment is con-
cerned.
Our calculation of misalignment has been adjusted according to a base year, where the RER could
be considered close to its equilibrium level. This has been the case especially in periods following
devaluation and structural adjustment, when the balance of payments was also close to equilibrium.
For example, it has been considered that RER was close to equilibrium in 1989 in Morocco; in 1991
and 1994-95 in Algeria; in1993-94 in Egypt; in1995 in Iran; in 1992 in Jordan; and in 1980, 1994, and
1997 in Tunisia. The method used to determine the probability of the RER being in equilibrium was
to identify a period of time when the difference between the observed and the sustainable value of the
fundamental variables was very small.
Some more sophisticated calculations consist-when a variable has a unit root-of using time series
techniques introduced by Beveridge and Nelson (1981), where variables are decomposed into a ran-
dom walk with a drift and a stationary component. This technique, unlike the trend stationary model-
based decomposition, allows the steady-state growth path of the series to shift over time. Fluctuations
around the shifting permanent path reflect cyclical effects.
11. To compute this indicator, some economists use more or less sophisticated regression techniques,
such as the variance of the residual of the regression of the RER on a time trend, or an ARCH mod-
elization of RER behavior. However, from an empirical point of view, all these measures are highly cor-
related, and the standard deviation or the coefficient of variation measures perform as well as more
sophisticated ones (see Kenen and Rodrik, 1986; Grobar, 1993).

Part III: Trade, Competitiveness, and Investment 393
12. As shown by the value of the Fischer test of equality of intercepts across countries, and by the value
of the Haussmann test as far as the random effect method is concerned (table 15.4).
13. Before proceeding to the estimation of Equation (2), the degree of integration of the series enter-
ing into the regression and the existence of a long-term relationship among them have been tested.The
results of the ADF tests of the variables of Equation (2) - using Im et al. (1997) critical values - are
shown in Table A.2, Annex 2.
14. Surprisingly,in the case of roads, the elasticity for manufactured exports is weakly significant. This
may be due to the fact that, in several MENA countries, oil exports represent an important percentage
of total exports (as well as of GDP). In this case, it can be assumed that oil exports have led to the con-
struction of good physical infrastructure.
References
Achy, L. 2001. “Equilibrium Exchange Rate and Misalignment in Selected MENA Countries.” EUI
Working Papers, Mediterranean Programme Series, RSC no. 2001/42. European University
Institute, Florence, Italy.
Balassa, B. 1990. “Incentive Policies and Export Performance in Sub-Sahara Africa.” Wor l d
Development 18(3).
Baffes, J., I. A. Elbadawi, and S. A. O’Connell. 1997. “Single-Equation Estimation of the Equilibrium
Real Exchange Rate.” Unpublished. Swarthmore College, Swarthmore, PA (May).
Beveridge, S. and C. R. Nelson. 1981. “A New Approach to Decomposition of Economic Time Series
into Permanent and Transitory Components with Particular Attention to Measurement of the
Business Cycle.” Journal of Monetary Economics 7.
Cottani, J. A., D. F. Cavallo, and M. S. Khan. 1990. “Real Exchange Rate Behavior and Economic
Performance in LDCs.” Economic Development and Cultural Change 39.
Cushman, D. O. 1993. “The Effects of Real Exchange Rate Risk on International Trade.” Journal of
International Economics 15.
Domac, I. and G. Shabsigh.1999. “Real Exchange Rate Behavior and Economic Growth: Evidence from
Egypt, Jordan, Morocco and Tunisia.” IMF Working Paper WP/99/40 (March). International
Monetary Fund, Washington, DC.
Edwards, S. 1988. Exchange Rate Misalignment in Developing Countries. Baltimore: The Johns Hopkins
University Press.
———. 1989. Real Exchange Rate, Devaluation and Adjustment: Exchange Rate Policy in Developing
Countries. Cambridge, MA: MIT Press.
———. 1994. Real and Monetary Determinants of Real Exchange Rate Behavior: Theory and
Evidence from Developing Countries. In Estimating Equilibrium Exchange Rates, ed. by J.
Williamson.Washington DC: Institute of International Economics.
Elbadawi, I. 1994.“Estimating Long-term Equilibrium Real Exchange Rates.”In Estimating Equilibrium
Exchange Rates, ed. by J.Williamson.Washington DC: Institute of International Economics.
Frankel, J.A. and D.Romer. 1999.“Does Trade Cause Growth?” The American Economic Review 89(3).
Gagnon, J. E. 1993. “Exchange Rate Variability and the Level of International Trade.” Journal of
International Economics 34(3-4).
Ghura, D.and T. J. Grennes. 1993.“The Real Exchange Rate and Macroeconomic Performances in Sub-
Saharan Africa.” Journal of Development Economics 42.
Grobar, L. M. 1993. “The Effect of Real Exchange Rate Uncertainty on LDC Manufactured Exports.”
Journal of Development Economics 14.
Harberger, A. 1986. “Economic Adjustment and the Real Exchange Rate. In Economic Adjustment
Exchange Rates in Developing Countries, ed. by S. Edwards and L. Ahamed.Chicago: University of
Chicago Press.

Breaking the Barriers to Higher Economic Growth394
Im, K. S., M. H. Pesaran, and Y. Shin. 1997. “Testing for Unit Roots in Heterogeneous Panels.” DAE
Working Paper Amalgamated Series, no. 9526. Cambridge University, Cambridge, UK.
Kenen, P. B. and R.Rodrik. 1996.“Measuring and Analysing the Effects of Short-Term Volatility in Real
Exchange Rates.” Review of Economics and Statistics 68.
Mongardini, J. 1998. “Estimating Egypt Equilibrium Real Exchange Rates.” IMF Working Paper,
WP/98/5 (January). International Monetary Fund, Washington, DC.
Nishimizu, M., and S. Robinson. 1986. “Productivity Growth in Manufacturing.” In Industrialization
and Growth: A Comparative Study, ed. by H. Chenery, S. Robinson, and M. Syruin.World Bank
Research Report. Oxford: Oxford University Press.
Razin, O.and S. M. A.Collins. 1997.“Real Exchange Rate Misalignments and Growth.” NBER Working
Paper no. 6174 (September). National Bureau of Economic Research, Cambridge, MA.
Reinhardt, C. 1995. “Devaluation, Relative Prices, and International Trade.” IMF Staff Paper No. 42.
International Monetary Fund, Washington, DC.
Sekkat K. and A. Varoudakis. Forthcoming, 2002. “The Impact of Exchange and Trade Policy Reforms
on Manufactured Exports in North Africa.” Development Policy Review.
Sorsa P. 1999. “Algeria: The Real Exchange Rate, Export Diversification, and Trade Protection.” IMF
Working Paper No. 99/49 (April). International Monetary Fund, Washington, DC.
Sundararajan, V., M. Lazare, and S. Williams. 1999. “Exchange Rate Unification, the Equilibrium Real
Exchange Rate, and Choices of Exchange Rate Regime: The Case of Islamic Republic of Iran.”IMF
Working Paper.WP/99/15 (January). International Monetary Fund, Washington, DC.
Williamson, J. 1985. “The Exchange Rate System.” Policy Analyses in International Economics 5.
Washington, D.C.: Institute for International Economics.

Part III: Trade, Competitiveness, and Investment 395
Annex 15.1
List of Countries in the Sample
MENA AFRICA ASIA LATIN AMERICA
CFA Non-CFA South East Asia
Bahrain (BHR) Burkina Faso (BFA) Botswana (BWA) Indonesia (IDN) Argentina (ARG)
Algeria (DZA) Côte d’Ivoire (CIV ) Gambia, The (GMB) Korea, Rep.(KOR) Bolivia (BOL)
Egypt (EGY) Gabon (GAB) Kenya (KEN) Malaysia (MYS) Brazil (BRA)
Iran (IRN) Cameroon (CMR) Madagascar (MDG) Philippines (PHL) Chile (CHL)
Jordan (JOR) Ghana (GHA) Mozambique (MOZ) Thailand (THA) Colombia (COL)
Kuwait (KWT ) Niger (NER) Mauritius (MUS) South Asia Costa Rica (CRI)
Malta (MLT) Senegal (SEN) Malawi (MWI) Bangladesh (BGD) Ecuador (ECU)
Morocco (MAR) Togo (TGO) Nigeria (NGA) India (IND) Guatemala (GTM)
Syria (SYR) Tanzania (TZA) China (CHN) Mexico (MEX)
Tunisia (TUN) Sri Lanka (LKA) Peru (PER)
Other countries Pakistan (PAK) Paraguay (PRY)
Israel (ISR) Uruguay (URY)
Venezuela, RB (VEN)

Breaking the Barriers to Higher Economic Growth396
Annex 15.2
Table 15A.1. Augmented Dickey-Fuller (ADF) Unit Root Tests
Equations (1) and (1’)
Variable ADF statistic k(1) Critical value(2) ADF test
RER
ln(et)⫺1.73 1 ⫺1.69** I(0)
Fundamentals
ln(Invt)⫺1.92 1 ⫺1.82* I(0)
ln(Opent)⫺1.69 1 ⫺1.69** I(0)
ln(TPt)⫺3.77 1 ⫺1.82* I(0)
ln(TOTt)⫺2.15 1 ⫺1.82* I(0)
Capinft⫺2.79 1 ⫺1.82* I(0)
DebtSevt
Other variables
Deft⫺2.43 1 ⫺1.82* I(0)
p⫺2.76 1 ⫺1.82* I(0)
Deprt⫺3.07 1 ⫺1.82* I(0)
BMPt⫺2.69 1 ⫺1.82* I(0)
Source: Authors’calculations.
(1) kis the number of lags in the ADF test.
(2) Im et al.(1997) critical values (respectively, *1 and **5 percent level).
Data have been compiled from World Development Indicators, Global Development Finance, Global Development
Network, and Live Data Base (World Bank).
Table 15A.2. Augmented Dickey-Fuller (ADF) Unit Root Tests
Equation (2)
Variable ADF statistic k(1)kCritical value(2) ADF test
Ln(Xmanuf i,t )⫺1.76 ⫺1.69**
GDPgrTP i,t ⫺3.69 1 ⫺1.82* I(0)
ln(TOTn i,t)⫺2.15 1 ⫺1.82* I(0)
ln(Inv i,t)⫺1.92 1 ⫺1.82* I(0)
ln(Roads i,t)⫺3.65 1 ⫺1.82* I(0)
ln(H1 i,t)⫺1.86 1 ⫺1.82* I(0)
RERVol ⫺2.83 1 ⫺1.82* I(0)
Ln(RERMis) ⫺2.24 1 ⫺1.82* I(0)
Source: Authors’calculations.
(1) kis the number of lags in the ADF test.
(2) Im et al.(1997) critical values (respectively, *1 and **5 percent level).
Data have been compiled from World Development Indicators, Global Development Finance, Global Development
Network, and Live Data Base (World Bank).

Part III: Trade, Competitiveness, and Investment 397
Annex 15.3
Short-Term Dynamics of the RER
Since the variables are cointegrated, the short-term dynamic adjustment of
the RER toward its equilibrium level can be estimated through an error cor-
rection model. The estimated equation is as follows:
Δln(ei,t) = –a [ln(ei,t -1) –ln(e( i,t- 1)]
+ a’ Δln(ei,t -1 )
+ b1.Δln(Inv i,t) + b2.Δln(Open i,t)
+ b3.Δln(TOT i,t) + b4.Δln(Capinf i,t)
+ b5.Δln(DebtServ i,t)
+c1.Δln(Inv i,t -1) + c2.Δln(Open i,t -1) (A3-1)
+ c3.Δln(TOT i,t -1) + c4.Δln(Capinf i,t -1)
+ c5.Δln(DebtServ i,t -1)
+ d1.Depr i,t +d2.Depr i,t -1
+ e1.Infl i,t + e2.Inf i,t -1
+ f1.Def i,t + f2.Def i,t -1
+ g1.BMPt+g2.BMP i,t -1 + ε2t.
In addition to the error correction term—that is, , the lagged error term of
the cointegrating equation (ln(et-1) - ln(e(t-1 )(, and lagged variables of
Equations (1) and (1’) in first differences—indicators of fiscal policy (fiscal
deficit as percentage of GDP, Def) and of exchange rate policy (nominal
depreciation, Depr, and black market premium, BMP), as well as inflation
(Infl) are included. The assumption is that the adjustment path of the RER
toward its equilibrium level may be affected (accelerated or slowed down) by
short-term economic policies, including capital controls (for which BMP is a
proxy), nominal exchange rate depreciation, and fiscal policy, of which infla-
tion can be a consequence. Table 15A.3 below shows the estimates of the error
correction model.
Nominal devaluations show a short-term impact on the RER, which is in
the expected direction, and significant. The change in the official nominal
exchange rate (NER) hence captures the strong temporary effect that devalu-
ation may produce on the RER due to price rigidities.
In addition, these estimations highlight the role of other short-term eco-
nomic policies through the black market premium (BMP) and inflation (Infl).
These variables (Infl,BMP), by leading to a rise in the price of nontradable
goods, appreciate the RER and lead to its overvaluation. Although public
deficit does not show a significant effect, it can be captured by the inflation
variable, the effect of which is strong and which is also supposed to be a proxy
for other inappropriate policies.

Breaking the Barriers to Higher Economic Growth398
Table 15A.3. Estimates of the Error Correction Model
Dependant variable: Δln(et)
Eq (1) Eq (1’)
Variable Elasticity Student Elasticity Student
ε1t-1 ⫺0.13 (7.29) ⫺0.2 (9.7)
Δln(Invt)0.04 (1.42) 0.2 (0.78)
Δln(Opent)⫺0.27 (6.97) ⫺0.5 (14.53)
Δln(TOTt)0.1 (2.7) 0.1 5 (4.8)
Δ(Capinf) 0.006 (1.27) 0.25 (3.8)
Δln(DebtServt) 0.02 (1.81)
Δln(Invt-1)0.01 (0.33) 0.03 (1.2)
Δln(Opent-1)0.06 (1.57) 0.02 (0.5)
Δln(TOTt-1)0.02 (0.72) 0.04 (1.4)
Δ(Capinf t-1)0.78 (1.81) ⫺0.33 (5.1)
Δln(DebtServt-1) 0.04 (2.1)
Δln(et-1)0.06 (1.64) 0.16 (4.9)
Depr ⫺0. 22 (18.0) ⫺0.04 (10.9)
Deprt-1 ⫺0.05 (8.0) 0.006 (1.4)
Inflt0.19 (17.8) 0.04 (10.4)
Inflt-1 0.05 (7.91) ⫺0.007 (1.6)
Deft-1 0.05 (0.38)
Deft0.05 (0.44)
BMPt0.006 (2.5) 0.12 (5.5)
BMPt-1 0.21 (0.86) ⫺0.003 (1.47)
D- W 1.74 2.03
Source: Authors’ estimations.
Note: Student tstatistics are within brackets. The sample includes, respectively, 640 and 828 observations over the
1970–99 period. * ε1t-1 is the lagged error term of the cointegrating Equation (1). Data have been compiled from
World Development Indicators, Global Development Finance, Global Development Network, and Live Data Base
(World Bank).

399
Public Infrastructure and
Private Investment in the
Middle East and North Africa
Pierre-Richard Agénor*
Mustapha K. Nabli
Tarik M. Yousef†
16
According to conventional wisdom, the poor growth and employment per-
formance of the Middle East and North Africa (MENA) region is primarily
the result of the “slow, uneven, and hesitant pace” of structural reforms
launched in the late 1980s and early 1990s (see World Bank 2003a) and
Richards and Waterbury (1996).1In particular, governments in the region
continue to dominate most economies, with pervasive involvement in pro-
duction, labor markets, banking systems, and social services. Despite downsiz-
ing efforts, the share of MENA’s public sectors in output and employment still
exceeds averages for developing and industrialized countries. In addition to
the size and scope of government intervention, private sector development in
the region continues to be stifled by limited progress in building market-ori-
ented institutions and in integrating the region into the world economy (see
World Bank 2003b). As a result, the economic recovery of the 1990s was weak,
labor productivity remained low, and unemployment rates continued to
increase.
* School of Social Studies, University of Manchester, United Kingdom, and Centre for Growth and
Business Cycle Research.† School of Foreign Service, Georgetown University. The authors are grateful
to Nihal Bayraktar for technical support as well as to the editor, three referees, and participants at var-
ious seminars for helpful comments on an earlier draft.
World Bank Policy Research Working Paper No. 3661. July 2005. Reprinted by permission from the
World Bank.

Breaking the Barriers to Higher Economic Growth400
The absence of dynamic private sectors in MENA’s economies has been
especially felt in the area of investment. Observers in the early 1990s had taken
the view that the public sector has “overinvested” and that public investment
competes with, rather than fosters, private investment (see Page 1998 and
World Bank 1995). But the decline in public investment rates was not always
compensated by a rise in private investment. As a result, capital accumulation
rates on a per worker basis stagnated in the past two decades (see Nabli and
Keller 2002). Indeed, with the exception of Sub-Saharan Africa, MENA has the
lowest private investment ratios among developing regions. Moreover, in
countries where public investment levels remained high, the productivity of
capital was limited. The very nature of the network utilities that were built to
provide infrastructure services (vertically and horizontally integrated state
monopolies) often resulted in weak delivery of services such as electricity, nat-
ural gas, telecommunications,railroads, and water supply.Common problems
included low productivity, high costs, bad quality, insufficient revenue, and
shortfalls in maintenance spending (see World Bank 1994 and Kessides 2004).
Low productivity of public investment is consistent with recent growth
accounting exercises showing that the contribution of physical capital accu-
mulation to growth in MENA countries has declined over time—despite the
fact that there has been no attempt in this literature to explicitly separate pub-
lic and private capital accumulation. 2
The extent to which public investment (especially in infrastructure) com-
plements or crowds out private investment, and the role of quantity versus
quality in the productivity of public investment in MENA, remain largely
unknown. To date, there have been few empirical studies focusing on these
issues in the region.3In an early paper on Egypt, for instance, Shafik (1992)
found that public investment tends to crowd out private investment through
its effect on credit markets, and to crowd it in through investment in infra-
structure. Everhart and Sumlinski (2001), using panel regression techniques
and a proxy for the quality of public investment, found no significant effect of
public investment on private investment in MENA. Dhumale (2000), using a
model that accounts for credit to the private sector and the accelerator effect,
found that public investment in infrastructure appeared to have a crowding
out effect in oil-exporting countries, and a crowding-in effect in the non-oil-
exporting countries. Mansouri (2004) found that public capital had a positive
effect on private investment in Morocco.
However, existing studies are lacking in at least three respects. First, they
seldom make a clear distinction between the flow effect of public investment,
and the stock effect of public capital. But this is crucial, given that the trans-
mission channels are substantially different. Second, these studies do not
always account for the simultaneous relationships between public investment
and capital, private capital formation, and other variables like output growth,
relative prices, and private sector credit. Third, the treatment of dynamics in

Part III: Trade, Competitiveness, and Investment 401
these studies is sometimes crude, if not inexistent. All three issues are
addressed in this paper, which assesses quantitatively the impact of public
infrastructure on private capital formation in three MENA countries: Egypt,
Jordan, and Tunisia.
The remainder of the paper is organized as follows. The next section pro-
vides a brief overview of direct and indirect channels through which public
infrastructure (flows and stocks) may affect private investment. The third sec-
tion describes the vector autoregression (VAR) model that we use to assess the
links between public infrastructure and private capital formation. The fourth
section examines the data and the construction of our quality measures of the
public capital stock in infrastructure. Estimation results and variance decom-
positions are discussed in the fifth section. In the last section, impulse
response functions are computed to assess the dynamic effects of a shock to
public infrastructure expenditure and the public capital stock. The last section
draws together some policy implications of our analysis.
Public Infrastructure and Private Investment: Transmission Channels
Public infrastructure investment and capital can affect private investment
through various channels. For the purpose of this study, and given the empir-
ical technique that we use later on, it is convenient to classify these channels
into two broad sets of effects: complementarity and crowding-out effects, and
output and relative price effects.
Complementarity and Crowding-Out Effects
The complementarity effect asserts that public capital (as opposed to public
investment) in infrastructure may stimulate private physical capital formation
because of its impact on private activity. By raising the marginal productivity
of private inputs (both labor and capital), it raises the perceived rate of return
on, and increases the demand for, physical capital by the private sector.4
Alternatively, a complementarity effect between public capital in infrastruc-
ture and private investment may operate through installation costs. This idea,
formalized for instance by Turnovsky (1996) in a growth context, is based on
the view that the availability (and quality) of public capital in infrastructure
affects some of the costs that firms may incur when investing. For instance, a
better road network may reduce expenses associated with the construction of
a new factory or the transportation of heavy equipment to a new production
site. In large countries, the impact on unit production costs and the produc-
tivity of private capital can be substantial (Cohen and Paul 2004).
Of course, the positive effect of public capital on the marginal productivi-
ty of private inputs may hold not only for infrastructure but also for public
capital in education and health, which may enhance the productivity of labor.
Other components of current public spending, related, for instance, to the

Breaking the Barriers to Higher Economic Growth402
enforcement of property rights, can also increase the productivity of the econ-
omy and exert a positive indirect effect on private investment. But infrastruc-
ture capital may have a particularly large effect in countries where initial
stocks are low and basic infrastructure services (electricity and communica-
tions, for instance) are lacking. Moreover, whereas gains in education and
health tend to accrue to individuals, increases in public infrastructure assets
tend to have economy-wide spillover effects.
Nevertheless, to the extent that public investment in infrastructure dis-
places or crowds out private investment, its net positive impact on private
capital formation can be highly mitigated. Such crowding-out effects tend to
occur if the public sector finances the increase in public investment through
an increase in distortionary taxes—which may raise incentives for private
agents to evade taxation, or reduce the expected net rate of return to private
capital, and therefore the propensity to invest. A similar effect on private cap-
ital formation may occur if the increase in public infrastructure investment is
paid for by borrowing on domestic financial markets, as a result of either
higher domestic interest rates or a greater incidence of rationing of credit to
the private sector.5Moreover, if an investment-induced expansion in public
borrowing raises concerns about the sustainability of public debt over time
(that is, the perceived risk of default), and strengthens expectations of a future
increase in taxation, the risk premium embedded in interest rates may
increase.6By negatively affecting expected after-tax rates of return on private
capital, the increase in the cost of capital may have a compounding effect on
private investment. Private investors may revise downward their investment
plans because of anticipated hikes in tax rates to cover the increase in govern-
ment investment.
Moreover, it is important to note that the productivity and complementar-
ity effects of public infrastructure assets are significant only to the extent that
the services derived from them are of sufficient quality. Having roads is a good
step, but if these roads are filled with potholes, they may not do much to
reduce installation costs. Likewise, an erratic supply of electricity may be tan-
tamount to not having electricity at all. Quality considerations are therefore
important when assessing the benefits associated with public capital. This is
indeed one of the main features of the empirical methodology developed in
this paper.
Indirect Output and Relative Price Effects
Public investment and capital in infrastructure may also affect private capital
formation indirectly, through changes in output and relative prices. As noted
earlier, public capital in infrastructure may increase the marginal productivi-
ty of existing factor inputs (both capital and labor), thereby lowering margin-
al production costs and increasing the level of private production. In turn, this
scale effect on output may lead, through the standard accelerator effect, to

Part III: Trade, Competitiveness, and Investment 403
higher private investment (see Chirinko 1993). Moreover, if there are exter-
nalities associated with the use of some production factors (for instance,
learning-by-doing effects resulting from a high degree of complementarity
between physical capital and skilled labor), a positive growth effect may also
result.
Public infrastructure can also affect private investment indirectly through
its “flow” effect on the price of domestic consumption goods relative to the
price of imported goods, that is, the (consumption-based) real exchange rate.
An increase in public investment in infrastructure, for instance, will raise
aggregate demand and domestic prices (in addition to stimulating output).
The real exchange rate will tend to appreciate,thereby stimulating demand for
these goods and dampening domestic activity. The net effect on output may
be positive or negative, depending on the intratemporal elasticity of substitu-
tion between domestic and imported goods. If this elasticity is low (as one
would expect in the short run), the net effect on output may be positive, so
that private investment may indeed increase. At the same time, the apprecia-
tion will tend to lower the relative price of imported capital goods, resulting
in a drop in the user cost of capital and an increase in private investment. This
relative price effect may be particularly important in developing countries
where a large fraction of capital goods used by the private sector are imported.
In addition to these effects, changes in domestic prices and the real
exchange rate induced by an increase in the flow of public investment in infra-
structure may affect private investment through both demand- and supply-
side effects on output. On the demand side, the increase in domestic prices
may lower private sector real wealth and thus expenditure; if this effect is suf-
ficiently large (relative to the increase in public spending) to entail a fall in
domestic absorption, firms may revise their expectations of future demand
and lower investment outlays, through a “reverse” accelerator effect. On the
supply side, the real appreciation may lead to a shift in resource allocation
toward the nontradable goods sector, thereby stimulating investment in that
sector and depressing capital formation in the tradable goods sector. The net
effect may be a lower growth rate of output, and thus lower investment as a
result of an expected reduction in demand growth. At the same time, howev-
er, the real appreciation tends to lower the real cost of imported intermediate
inputs, thereby stimulating output and private investment.
It is important to note that both the direction and the strength of the var-
ious effects described above can vary over time, and depend to a very large
extent on the environment in which private investors are operating. For
instance, the relationship between public and private investment may be one
of substitution in the short run, and one of complementarity in the long run,
depending on how “productive” public investment is. In the short term, the
crowding-out effect may predominate (because the pool of resources available
to finance public and private investment is limited), whereas the complemen-

Breaking the Barriers to Higher Economic Growth404
tarity effect may prevail in the long term, as a result of strong supply-side
effects. Thus, using dynamic models is essential to study the relationship
between public infrastructure and private capital formation, beyond the need
to account for gestation lags. At the same time, it is important to control for
indirect effects that operate through changes in output, the real exchange rate,
and possibly interest rates or credit.
Finally, it is also worth noting that there may be a feedback effect through
public investment itself; indeed, to the extent that the rise in private invest-
ment stimulates output and leads to higher tax revenue, public investment
may increase further, as a consequence of the additional resources at the dis-
posal of the public sector. These dynamic and feedback effects are key reasons
for choosing a VAR framework for our empirical analysis, as discussed next.
VAR Specification
The foregoing discussion suggests that it is important, in assessing the link
between public infrastructure and private investment, to account for both the
flow and stock of public infrastructure, and to control for simultaneous inter-
actions between these variables and output, the real exchange rate, and finan-
cial variables (either interest rates or private sector credit). Accordingly, we
opted to use a VAR approach. VAR models offer a number of advantages over
the specification and estimation of a structural model. First, in developing
countries in general, it has proved difficult to estimate robust structural mod-
els of private investment (see Agénor 2004). VAR models offer a way of ana-
lyzing the dynamic relationship between our two main variables without hav-
ing to fully specify a structural model of private capital formation. The lumpy
nature of much infrastructure investment implies that the full impact of
investment in, say, roads or telecommunications, may be felt only after sever-
al years; VAR models allow us to take into account delayed responses, even
with a parsimonious lag structure. Second, VAR models explicitly recognize
the endogeneity of public infrastructure investment and capital—which may
result, as noted above, from the feedback effect of private investment on out-
put (through tax revenue). Third, VAR models provide a convenient common
framework for examining investment behavior in a cross-country study.
Using a uniform single regression model would amount to imposing strong
restrictions on specification and the direction of causality among the vari-
ables. As a result, models of this type tend to be prone to misspecification
errors resulting from “missing variables” bias and the neglect of dynamic
feedbacks—a particularly important problem when the purpose of the study
is to conduct simulation experiments.
The use of VAR models to study the impact of public investment on private
capital formation is by no means new. For instance, Mittnik and Neumann
(2001) examined the impact of public investment using impulse response

Part III: Trade, Competitiveness, and Investment 405
functions derived from a VAR consisting of public investment, private invest-
ment, public consumption, and output. Ghali (1998) used a VAR (or, more
precisely, a vector error correction model) with real GDP, public investment,
and private investment. Ligthart (2000) used an unrestricted VAR in output,
public capital, private capital, and employment for Portugal. Belloc and
Vertova (2004), using a vector error-correction approach, found a comple-
mentarity relationship between public and private investment, and a positive
effect of investment on output in six out of seven highly indebted poor coun-
tries. Finally, Voss (2002) specified a VAR with ratios of public and private
investment to GDP, the growth rates of the relative prices of public and pri-
vate investment goods, the real interest rate, and the growth rate of GDP.
All these studies, however, suffer from three major limitations in terms of
their specification: (i) they do not generally make a distinction between the
flow of public investment and the stock of public capital; (ii) they do no
always account for potential crowding-out effects; and (iii) they do not
account for indirect effects of public investment on private capital formation
through the real exchange rate. Belloc and Vertova (2004), for instance, used a
trivariate VAR, with no control for factors other than output. As emphasized
earlier, the channels through which public infrastructure affects private
investment involve both “flow” effects (which operate through aggregate
demand, relative prices, and the financial sector), and “stock” effects (which
operate through both the demand and the supply sides).
Our VAR improves on existing studies in all three respects. We include the
following variables in our specification: the flow of public capital expenditure
on infrastructure as a share of GDP, the stock of public capital in infrastructure
as a share of GDP, private capital formation as a share of GDP, the ratio of pri-
vate sector credit to GDP, real GDP growth, and the rate of change of the real
exchange rate. The actual growth rate of output is used as a proxy for expect-
ed changes in aggregate demand, and captures dynamics associated with the
accelerator effect. Changes in private sector credit account for possible crowd-
ing-out effects associated with government spending through changes in cred-
it rationing. We chose a credit variable instead of Interest rates, because these
rates remained largely under government control for much of the estimation
period.7Changes in the real exchange rate account for both the relative price
effect of an increase in domestic absorption, and indirect effects on the user
cost of capital and the price of imported inputs, as discussed earlier.8
To assess whether the stock of public capital in infrastructure should be
included in the VAR (in addition to the associated flow), we performed an
exogeneity test based on estimating both the “unrestricted” and “restricted”
VAR models (that is, with and without the public capital stock).9To calculate
variance decompositions and identify impulse response functions, we use the
standard Choleski decomposition. Specifically, to implement this decomposi-
tion, the disturbances in the model are assumed to follow the following causal

Breaking the Barriers to Higher Economic Growth406
ordering: credit-to-GDP ratio; public infrastructure spending ratio; public
infrastructure capital ratio; the rate of change of the real exchange rate; the
growth rate of GDP; and the private investment ratio. The reasoning behind
this ordering structure is that whereas public expenditure decisions or the
public capital stock can affect private sector investment decisions in the short
run (within one period), the reverse is not true. Thus, public expenditure on
infrastructure does not depend contemporaneously on private investment, an
assumption that we take to be consistent with treating public investment as
exogenous (at least with respect to private investment) in structural models.
Public expenditure on infrastructure naturally precedes the public capital
stock. The real exchange rate and output growth are both assumed to respond
immediately (within a year) to changes in public investment and the public
capital stock. Private sector credit is considered the most “exogenous” vari-
able, with the implicit view being that it is largely under the control of (risk-
averse) banks.10
Finally, it is important to note that in our VAR model, flows and stocks are
entered as ratios, not as levels.Because we use the Perpetual Inventory Method
(PIM) to calculate our estimates of capital stocks, there is a recursive (or
autoregressive) relation between flows and stocks. However, this relation
holds only for the levels of the variables; in our VAR model, we consider the
ratios of investment (public and private) to GDP as well as public capital to
GDP. An exogenous shock to, say, the public investment-GDP ratio may or
may not lead to a change in the same direction of the public capital-GDP
ratio, because the response of output is endogenous in the VAR. In particular,
if both the public capital stock and GDP increase as a result of the shock (the
former arguably in a mechanical fashion, as a result of the autoregressive
nature of PIM), their ratio would remain unchanged.
The Data
We begin by examining the data on private and public investment in the three
countries in our sample. Next, we consider the evolution of public capital
expenditure on infrastructure, and describe how these flows are converted
into stocks. We then explain how our basic indicators of the quality of public
capital in infrastructure are constructed, and how they are used to derive a
composite indicator.
Overall Trends
Figure 16.1 shows the evolution of public and private investment ratios to
GDP since the mid-1960s in Egypt, the mid-1970s in Jordan, and the early
1970s in Tunisia. The share of public investment in GDP has displayed sub-
stantial volatility over time in all three countries, but has been on a downward
trend in Egypt and Jordan since the late 1980s. Private investment ratios have

Part III: Trade, Competitiveness, and Investment 407
Figure 16.1. MENA Countries: Public and Private Investment
0
5
10
15
20
25
30
35
40
1965 1967 1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001
Egypt
0
5
10
15
20
25
30
35
40
45
50
1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002
Jordan
0
5
10
15
20
25
30
35
40
1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000
private
public
Tunisia
% of GDP% of GDP% of GDP

Breaking the Barriers to Higher Economic Growth408
at the same time been subject to large fluctuations, most significantly during
the 1980s and 1990s—a period characterized also by large fluctuations in
GDP in the region as a whole. In Egypt, following a steady increase from the
mid-1960s to the late 1980s, the share of private investment in GDP has aver-
aged 10 percent. In Jordan and Tunisia, private investment ratios have
declined significantly since the peaks of the early 1990s, fluctuating in recent
years between 12 and 15 percent.
Flows and Stocks of Public Infrastructure
National Accounts data on public investment in infrastructure are generally
not available. For the purpose of our study, we used government budget data
published in the IMF’s Government Finance Statistics (GFS) Yearbook to build
an estimate. Specifically, as discussed in the Annex, we calculated capital
expenditure on “core” infrastructure by adding capital outlays on fuel and
energy, water transportation, and transportation and communications. Of
course, all capital expenditures on these categories do not necessarily repre-
sent “investment,” as conventionally defined; some components of these out-
lays may be related to current spending, such as maintenance operations.
However, we did not have sufficient information to refine the GFS estimates.
The data indicate that the evolution of total capital expenditure as a pro-
portion of GDP, and capital expenditure on infrastructure, both as a share of
total public expenditure (including current spending), and as a share of total
capital expenditure only, display a declining trend since the early 1980s for
Jordan and the late 1980s for Tunisia. But for all three countries, the behavior
of the ratio of total capital expenditure to GDP is consistent with the evolu-
tion of total public investment illustrated in Figure 16.1. As a proportion of
total capital outlays, the share of capital expenditure on infrastructure has
averaged 30 percent for Jordan, and almost 40 percent for Tunisia, in recent
years. For Egypt, by contrast, there are large fluctuations during the past 40
years, but no clear trends. In recent years, the share of capital expenditure on
infrastructure in total capital outlays has fluctuated between 45 and 50
percent.
Using our flow data on infrastructure spending, we calculated the stock of
public capital in infrastructure, using the perpetual inventory method and the
GDP deflator to estimate real values in constant local prices. This procedure
is, of course, subject to limitations.11 Differences in the efficiency of the pub-
lic sector and the price of infrastructure capital, in particular, mean that the
same level of capital spending on infrastructure may yield very different
results across countries. We account for quality ex post by using various indi-
cators, as discussed below. More generally, the use of perpetual inventory
methods to calculate stocks from expenditure flows may introduce systemat-
ic errors in stock estimates. At the same time, however,it should be noted that
adequate price variables for public infrastructure are difficult to construct

Part III: Trade, Competitiveness, and Investment 409
(given that infrastructure services are often provided free of charge), and that
errors in estimating initial stocks (a common problem with this methodolo-
gy) tend to become less significant over time. In addition, the alternative of
using actual stocks of infrastructure (such as roads, electricity production, or
water supply) was not feasible because of lack of data; the only complete series
that we had at our disposal was that of electricity production, but it was felt
that using it as a the sole indicator of quality of the overall stock of infrastruc-
ture was not warranted.
Quality Indicators
In assessing the impact of the public capital stock in infrastructure on private
investment, it is important to account not only for the absolute amount of
that stock, but also for the quality (or efficiency) with which public capital is
used.12 A common procedure for estimating the quality of public infrastruc-
ture capital is to calculate the index proposed by Hulten (1996) His compos-
ite measure of public capital efficiency is based on four basic indicators: main-
line faults per 100 telephone calls for telecommunications; electricity genera-
tion losses as a percent of total electricity output; the percentage of paved
roads in good condition; and diesel locomotive utilization as a percentage of
the total rolling stock. In practice, researchers have found that these individ-
ual quality indicators tend to be highly correlated with the quantities of each
type of infrastructure.13 Thus, much of the variation in infrastructure quality
may be well captured by variations in its quantity.Thus,much ofthe variation in infrastructure quality may be well captured by variations in its quantity.
The individual quality indicators proposed by Hulten (1996) are subject to
limitations. For instance, electric power losses include both “technical” losses,
reflecting the quality of the power grid, and outright theft; in general, the
breakdown between the two components is not available. Moreover, these
series tend to fluctuate significantly over time, and these fluctuations are not
always easy to interpret as changes in quality as opposed to, say, measurement
errors or “abnormal” shocks (due, say, to bad weather). More importantly in
the present case, these indicators were not all available for our group of coun-
tries, and when they were, many data points were missing. Despite using a
combination of local and international sources—including the World
Development Indicators of the World Bank and Canning’s (1998) database on
physical infrastructure stocks—we were unable to “piece together” complete
series, with a sufficient amount of overlapping with our private investment
series.
We therefore followed another approach, which consisted in, first, defining
two alternative quality indicators and, second, combining them to create a
composite indicator. Our first individual indicator is an “ICOR-based” meas-
ure. Aggregate ICORs (calculated as the ratio of total domestic investment
divided by the change in output) are commonly viewed as a measure of the
efficiency of investment. Here we apply this idea to public infrastructure only,

Breaking the Barriers to Higher Economic Growth410
by defining an ICOR coefficient as public capital expenditure on infrastruc-
ture divided by the change in GDP. We then invert this measure and take a
three-year moving average, in order to smooth out the behavior of the series
over time.
Our second indicator is an “excess demand” measure. Our premise is that, if
growth in the demand for infrastructure services tends to exceed growth in
supply, pressure on the existing public capital stock will intensify and quality
will deteriorate. To construct these indicators, we proceeded in two steps. First,
we calculated individual indicators of “excess demand” for three categories of
infrastructure services: electricity generation; the number of telephone main-
lines; and the percentage of paved roads in the road network. To estimate
demand for infrastructure service h, we applied the annual growth rate of real
GDP per capita to the stock of public capital in hat the base period. We used
elasticity values of unity in each case.14 To estimate supply of infrastructure
service h, we used the actual stock of h. We then calculated individual indica-
tors of excess demand for each component of infrastructure services by taking
the ratio of supply to “predicted” demand. This ratio gives, therefore, an indi-
cator of adequacy between supply and demand; a fall in the ratio would indi-
cate excessive pressure on existing infrastructure, and therefore, a deterioration
in quality. Second, we calculated a “composite” excess demand indicator for
each country. To do so, we used the same procedure as Hulten (1996) did to
calculate his quality index, that is, we standardized each of the three series (by
subtracting the mean and dividing by the standard error) and calculated the
unweighted, arithmetic average of the standardized series.
Continuous annual time series for all three of the infrastructure services
referred to earlier were not available for the whole estimation period. For
instance, data on roads (in terms of kilometers per capita) were available only
since 1990. We therefore calculated the composite indicator with all the infor-
mation available in the base period to begin with, and added additional series
as they became available. Again, a three-year moving average was used, in
order to smooth out spikes possibly associated to measurement errors or ran-
dom events.
The evolution of both indicators is displayed in figures 16.2, 16.3, and 16.4.
Results for the ICOR-based indicator suggest that, whereas quality seems to
have improved in recent years in Tunisia, it has deteriorated in Egypt and
Jordan. By contrast, results for the composite excess demand indicator suggest
that quality improved in all three countries in recent years.
Differences in the behavior of our two quality indicators may appear prob-
lematic, given that there is no strong a priori reason for choosing among
them. We thus follow the logic of the approach proposed by Calderón and
Servén (2004b), who define several standard quality indicators—based on
electricity losses, percentage of paved roads, and telephone faults, given that
they use stocks of electricity, roads, and telephones in their regressions—and

Part III: Trade, Competitiveness, and Investment 411
use principal components analysis to "summarize" the information contained
in all of these series. Specifically, we applied principal components analysis to
the two quality series that we defined earlier.15 The results show that the first
principal component explains 64 percent of the total variance of the underly-
ing variables for Egypt, 68 percent for Jordan, and 55 percent for Tunisia. After
Figure 16.2. Egypt: Variance Decomposition of Private Investment in Percent of GDP
0
20
40
60
80
100
1 2 3 4 5 6 7 8 9 10111213 14 15161718 19202122 2324
public spending on infrastructure (in % of GDP) public capital in infrastructure (in % of GDP)
credit to private sector (in % of GDP) growth rate of real exchange rate (in %)
growth rate of real GDP (in %) private investment (in % of GDP)
Figure 16.3. Jordan: Variance Decomposition of Private Investment in Percent of GDP
0
20
40
60
80
100
1 2 3 4 5 6 7 8 9 101112131415161718 192021222324
public spending on infrastructure (in % of GDP) public capital in infrastructure (in % of GDP)
credit to private sector (in % of GDP) growth rate of real exchange rate (in %)
growth rate of real GDP (in %) private investment (in % of GDP)

Breaking the Barriers to Higher Economic Growth412
renormalization, we use a weighted average of the two principal components,
using as weights the proportion of total variance explained by each compo-
nent, as our quality indicator. The results show that the first principal component explains 64 percent ofthe total variance ofthe underlying variables
for Egypt,68 percent for Jordan,and 55 percent for Tunisia.After renormalization,we use a weighted average ofthe two principal components,using as weights the proportion oftotal variance explained by each component,as our quality indicator.
Estimation and Variance Decompositions
As noted earlier, our VAR model consists of public capital expenditure on
infrastructure as a share of GDP; private investment as a share of GDP; the
growth rate of real GDP; the ratio of private sector credit to GDP; the rate of
change of the real exchange rate; and possibly the public capital stock in infra-
structure. Prior to estimation, we examined the stationarity properties of each
of these variables with Augmented Dickey-Fuller and Phillips-Perron unit
root tests. The results (which are available upon request) indicated that all
series are either stationary or trend stationary. As a result, we included an
exogenous time trend in the VAR.
The first step in the estimation was to verify that the public capital stock in
infrastructure “belongs” to the VAR. To do so, we applied the exogeneity test
described earlier. The results indicated that the null hypothesis (exclusion of
the public capital stock from the VAR) was soundly rejected by the likelihood
ratio test. Moreover, because the crowding-out effects associated with changes
in public investment flows could be quantitatively small, we also tested for
whether the flow of public capital spending on infrastructure, as opposed to
the stock of public capital, should be excluded from the VAR. The results,
based on the likelihood ratio tests, again led to a rejection of the null
hypothesis.
Figure 16.4. Tunisia: Variance Decomposition of Private Investment in Percent of GDP
0
20
40
60
80
100
1 2 3 4 5 6 7 8 9 101112 131415 16171819 202122 2324
public spending on infrastructure (in % of GDP) public capital in infrastructure (in % of GDP)
credit to private sector (in % of GDP) growth rate of real exchange rate (in %)
growth rate of real GDP (in %) private investment (in % of GDP)

Part III: Trade, Competitiveness, and Investment 413
We then chose the optimal lag length, using the Akaike criterion. Given the
relatively small size of the sample, we were able to compare models with only
one and two lags.16 The Akaike criterion suggested a lag of two years for
Egypt, and one year for Jordan and Tunisia. Admittedly, lags of this order may
not be sufficient to properly account for the long gestation periods associated
with some types of public investment, but insufficient degrees of freedom pre-
vented us from experimenting with higher-order systems. Nevertheless, it
should be noted that we introduce explicitly the stock of capital in infrastruc-
ture itself, and that our hypothesis is that the (quality-adjusted) stock pro-
duces a proportional flow of services. Simulations related to changes in the
public capital stock are therefore unaffected by the problem of gestation lags.
The extent to which exogenous changes (or innovations) in public infra-
structure, and other variables in the VAR model, affect the behavior of private
capital formation can be gauged by computing the proportion of the variance
of the forecast error for the private investment-to-GDP ratio that can be
attributed to variations in each variable at different forecast horizons. Figures
16.2, 16.3, and 16.4 show these variance decompositions, at a horizon of up to
24 periods. For Egypt, shocks to the private investment ratio account for more
than 80 percent of its variance in the short run, and close to 50 percent in the
long run. The fraction explained by the credit-to-GDP ratio increases from
about 3 percent in the short term to 17 percent in the long term. So does the
share of public capital spending on infrastructure, the share of which rises
from 3 percent to almost 11 percent. The share of public capital also grows
over time, to about 13 percent.
For Jordan, the credit-to-GDP ratio and the ratio of public capital to GDP
play an important role in explaining the variability of the private investment
ratio, in both the short and the long term. By contrast, all the other variables
have a negligible role. By contrast, for Tunisia, the credit-to-GDP ratio
appears to play a negligible role; in both the short and the long term, it
explains barely 3 percent of variations in the private investment ratio. Public
expenditure on infrastructure and public capital, instead, explain about a
third of these fluctuations in the long term. Shocks to private investment itself
explain a large fraction of the variance of that variable, even in the long term.
Overall, therefore, the variance decompositions suggest that public capital in
infrastructure matters more than public spending on infrastructure (particu-
larly so for Jordan), although other shocks (such as credit for Jordan), as well
as “own” innovations (for Egypt and Tunisia), also appear to have mattered to
a considerable degree.
Impulse Response Analysis
We now examine the impulse response functions associated with shocks to
public spending on infrastructure and our quality-adjusted measure of pub-

Breaking the Barriers to Higher Economic Growth414
lic capital. As noted earlier, this analysis is important because it allows us to
assess to what extent flows and stocks of public infrastructure affect private
investment, taking into account crowding-out effects and the possibility that
indirect effects may occur through changes in the growth rate of output and
the real exchange rate.
Shock to Public Spending on Infrastructure
The left-hand side of figure 16.5 shows the response over a 10-year horizon of
the private investment rate to a one-standard-deviation innovation in the
ratio of public capital expenditure on infrastructure in GDP. The solid lines in
the figure represent the impulse response functions themselves, whereas the
dotted lines are the associated 95 percent upper and lower confidence bands.17
The results indicate that the shock to the public capital expenditure ratio
has a statistically significant effect on the private investment rate in none of
the three countries. Moreover, in the case of Egypt and Jordan, the shock has
no effect on the growth rate of output, the real exchange rate, or the credit
ratio. By contrast, in the case of Tunisia, the real exchange rate appreciates,
and output falls significantly, in the second period. Thus, the lack of signifi-
cance of a public spending shock on private investment in Tunisia may result
from the fact that the resulting increase in aggregate demand (which tends to
raise domestic prices) is offset by an adverse real wealth effect on private con-
sumption expenditure, at the same time as the real appreciation leads to a
contraction in output. In turn, this contraction in activity may offset the ini-
tial positive impact of public expenditure on private investment through a
“reverse” accelerator effect.
Shock to Public Capital in Infrastructure
The right-hand side of figure 16.5 displays the response of the private invest-
ment ratio to a one-standard-deviation innovation in the ratio of public
infrastructure capital to GDP. The results show that the shock has a positive
and statistically significant effect on private capital formation in the first two
periods for Tunisia, and the first three periods for Jordan. In the case of Egypt,
there is no significant effect. Moreover, in the case of Jordan, the growth rate
of output increases significantly in the first three periods, which seems to
occur without any significant pressure on domestic prices (and thus no ten-
dency for the real exchange rate to appreciate); thus, the rise in the investment
ratio may also reflect an indirect accelerator effect.
Overall, therefore, our results indicate that there are significant—albeit rel-
atively small, in absolute terms, and short-lived—“stock” effects only of pub-
lic infrastructure; there is no evidence of a “flow” effect in none of the coun-
tries in our sample. To assess whether the increase in the quality-adjusted
infrastructure public capital stock reflects an increase in the stock itself (that
is, the “raw” quantity of capital) or a change in quality, we re-estimated the

Part III: Trade, Competitiveness, and Investment 415
Figure 16.5. MENA Countries: Response of Private Investment in Percent of GDP to One Standard Deviation
Innovation in Public Spending on Infrastructure or in Public Capital Stock in Infrastructure, Both in Percent of GDP
(+/–2 standard errors)
⫺2.0
⫺1.5
⫺1.0
⫺0.5
0
0.5
1.0
1.5
2.0
2.5
3.0
12345678910
⫺2.5
⫺2.0
⫺1.5
⫺1.0
⫺0.5
0
0.5
1.0
1.5
12345678910
⫺4
⫺3
⫺2
⫺1
0
1
2
3
4
12 345678910
⫺4
⫺2
0
2
4
6
8
12345 678910
⫺2.0
⫺1.5
⫺1.0
⫺0.5
0
0.5
1.0
1.5
12345678910
⫺1
⫺0.5
0
0.5
1.0
1.5
2.0
2.5
12345678910
Innovation in Public Spending
on Infrastructure
(in % of GDP)
Egypt
Jordan
Tunisia
Innovation in Public Capital Stock
in Infrastructure
(in % of GDP)

Breaking the Barriers to Higher Economic Growth416
VAR model for each country with the unadjusted capital stock. The results
showed no significant effects associated with an innovation in the capital
stock in Jordan and Tunisia. Moreover, in the case of Jordan, the private
investment ratio actually fell slightly on impact. Our interpretation of these
results is thus that quality matters for both countries. Increasing the quantity
of infrastructure, by itself, does not have a significant effect on the ratio of pri-
vate investment to output.
Policy Implications
Our empirical results have potential useful policy implications not only for
the three countries in our sample but possibly also for the MENA region in
general. As concluded in a recent report by the World Bank (2004), stimulat-
ing growth and job creation in MENA will require comprehensive policy
reforms. A key issue in this context, as noted earlier, is the role of the public
sector.Although our study is subject to some limitations (owing in part to the
paucity of data and the “black box” nature of VARs) we tentatively view it as
bringing to the fore two main policy messages: the first is that the quality of
public investment, specifically in infrastructure, matters. The second is that
the weak effect of public capital on private investment may reflect the fact that
the complementarity effect, while potentially important, may not “kick in,”
because of an unfavorable environment for private sector activity.
Regarding the first message, an implication of our results is indeed that it
may be more important, in some countries, to improve the quality of the
existing infrastructure than to engage in further investment. Reducing unpro-
ductive public capital expenditure and improving quality must be accompa-
nied by policy reforms aimed at limiting investment to infrastructure capital
that crowds in the private sector and/or corrects for fundamental market fail-
ures.18 To do so requires redefining the role of the public sector as a catalyst
for, rather than a provider of, the majority of infrastructure services. This will
entail privatization and greater involvement of the private sector in infrastruc-
ture investment. Such involvement will allow commercial discipline to be
introduced in the delivery of services, thereby improving efficiency and the
quality and coverage of services, lowering costs, and reducing the burden on
the public budget.
Private sector involvement in the delivery of basic infrastructure has indeed
increased in developing countries. According to a comprehensive report by
the International Finance Corporation (2003), during the 1990s, more than
130 developing countries pursued (through a variety of schemes) private par-
ticipation in that sector. During the period 1990-2001, for developing coun-
tries in general, private participation accounted for 25 percent of total invest-
ment in infrastructure. However, the same report notes that in MENA, invest-
ment in infrastructure projects with private participation fell (in 2001 U.S.

Part III: Trade, Competitiveness, and Investment 417
dollars) from US$3.6 billion in 1993 to US$2.8 billion in 2001. Cumulative
investment in the region for the period 1990–2001 amounted to US$22.8 bil-
lion, which was less than that in Sub-Saharan Africa (US$23.4 billion) and
South Asia (US$39.6 billion), and substantially less compared to East Asia and
the Pacific (US$210.6 billion), Europe and Central Asia (US$97.1 billion), and
Latin America and the Caribbean (US$360.6 billion).19
The second policy message of our study—which is somewhat more specu-
lative in nature, given that we did not estimate a “structural” model—is that
the lack of a strong and persistent effect of public infrastructure capital on
private investment (even after adjusting for quality) may reflect the unfavor-
able environment in which the private sector has operated in the region.
Indeed, a key conclusion of recent research by the World Bank (2003b) is that
the unfavorable investment environment in MENA accounts to a significant
degree for the lack of a strong response from the private sector in the last
decade. While infrastructure (in the form of the provision of critical telecom-
munications, transport, and energy services) is important, other improve-
ments in the environment in which domestic investment is conducted may be
equally important. These include, inter alia, the need to provide financing on
adequate terms, and guarantee a secure and efficient judicial system. Without
renewed effort to tackle these issues, private investment may not achieve its
potential to stimulate growth and job creation in the region.
Notes
1. See World Bank (2004) for a review of employment and growth outcomes in the 1990s and their
links to trends in physical and human capital accumulation, and Agénor et al. (2004). Elbadawi (2004)
emphasized the role of conflict and instability, whereas Hakura (2004) found that excessive govern-
ment intervention and poor institutions were key factors hampering growth.
2. The same observation applies to growth in total factor productivity, which has been either negative
or below international levels since the 1980s (Nabli and Keller 2002). The efficiency of public spend-
ing on education has also been low, limiting the contribution of advances in educational attainment
in the region to growth (see Pritchett 1999 and World Bank 2004).
3. Appendix A of the Working Paper version of this article provides a broad review of empirical stud-
ies linking public investment to private capital formation and growth in developing countries.
4. Greater availability of public capital in infrastructure could in principle also reduce the demand for
private inputs, at a given level of output (net substitution effect). But if inputs are gross complements
(as is the case in general), higher availability of public capital will always increase the marginal pro-
ductivity of private inputs. Moreover, public and private physical capital are likely to have a high
degree of complementarity, that is, a small elasticity of (net) substitution.
5. Note that any component of government expenditure (not only infrastructure investment), as long
as it is financed through domestic borrowing, may lower private investment by driving interest rates
up or increasing the incidence of credit rationing.
6. In a small, open economy with open capital markets facing a fixed world interest rate, crowding-out
effects through a rise in domestic interest rates cannot occur. But for small developing countries, the
supply curve of foreign capital is upward-sloping rather than horizontal. In such conditions, and if the
risk premium faced on world capital markets is positively related to the debt-to-GDP ratio, an increase

Breaking the Barriers to Higher Economic Growth418
in domestic public debt, induced by a rise in public investment in infrastructure, may lead to both
lower credit to the private sector and higher domestic interest rates.
7. Mansouri (2004), in his study on Morocco, uses the real deposit rate as a proxy for the cost of bor-
rowing. However, this is a debatable assumption, given that official nominal interest rates remained
under control during much of his estimation period.
8. In principle, we should use changes in the ratio of the price of imported investment goods to the
domestic price of these goods (or the national accounts deflator of private investment), as for instance
in Mansouri (2004). However, sufficiently long time series were not available for our sample.
9. Let ΩUand ΩCdenote the variance-covariance matrices of the residuals associated with the unre-
stricted and restricted models, respectively, and define the likelihood ratio statistic, λ, as
λ= (T-c) (log|ΩC| - log|ΩU|),
where |ΩC| (respectively |ΩU|) is the determinant of ΩC(respectively ΩU), Tthe number of observa-
tions, and cthe number of parameters (equal to the number of lags times the number of variables, plus
one for the constant term) estimated in each equation of the unrestricted system. This statistic has a
χ2distribution with degrees of freedom equal to the number of restrictions in the system, which is in
turn equal to one times the number of lags.
10. Of course, this ordering may be viewed as somewhat arbitrary. However, in the actual estimation
we conducted sensitivity tests and verified that the results are largely independent of the causal order-
ing described above.
11. See Hulten (1990) for a detailed discussion of the conceptual and measurement problems involved in
constructing capital series. To calculate the public capital stock, we used a uniform depreciation rate of 2.5
percent. By comparison, Nehru and Dhaneswar (1993) used a uniform rate of 4 percent, whereas Larson
et al. (2000) use alternative values of 4 and 6 percent, to estimate aggregate stocks of capital. However, sen-
sitivity analysis showed that our empirical results are not unduly sensitive to our particular choice.
12. According to the World Bank (1994, p. 1), technical inefficiencies in roads, railways, power, and
water in developing countries caused losses equivalent to a quarter of their annual investment in infra-
structure in the early 1990s. See Estache (2004) for a further discussion.
13. Calderón and Servén (2004a, p. 19) found a high degree of correlation between the individual qual-
ity indicators listed above and the related quantities of infrastructure (that is, between power genera-
tion capacity and power losses, or between road density and road quality, the latter measured by the
proportion of paved roads in). In a companion study (Calderón and Servén (2994b, p. 11) they
obtained the same result with their two synthetic indicators of quantity and quality of infrastructure.
Esfahani and Ramírez (2003, p. 446) also noted the existence of a close correlation between stocks of
infrastructure capital and quality in their sample.
14. In their estimation of demand functions for infrastructure services based on panel data, Fay and
Yepes (2003, p. 8) found long-term elasticities of 0.375 for electricity, 0.5 for telephone mainlines, and
0.14 for paved roads. These estimates differ quite significantly from our values of unity. However, the
Fay-Yepes estimates refer to low- and middle-income countries in general, so there is no indication
that they are adequate for MENA countries in particular-or, for that matter,the three countries in our
sample. Moreover, in their regressions, there is no price (or user cost) variable, so their estimated
income elasticities may be biased.
15. See, for instance, Jackson (1991) for a description of principal components analysis. The first step
in this analysis is to put all of the data in standard units. By doing so, all of the transformed variables
have unit variances and the resulting covariance matrix is actually the correlation matrix of the origi-
nal variables.
16. Specifying a maximum lag length that is too short may impose unwarranted zero restrictions. At
the same time, imposing a lag length that is too long may result in inefficient parameter estimates,
because the model is over-parameterized.
17. The impulse responses and their associated confidence intervals are computed using Monte Carlo
simulations employing 1,000 draws. A complete set of results for all the variables in the VARs are avail-
able upon request.

Part III: Trade, Competitiveness, and Investment 419
18. The existence of market failures is not, of course, an automatic justification for government
involvement; such failures only provide a presumption of the need for government intervention.
Moreover, even when it is required or desirable, intervention can take many forms; direct public pro-
vision is only one of them, and not necessarily the best one. In practice, one has to take into account
possible government failures, and compare the costs and benefits of both options.
19. In MENA as elsewhere,a large share of these investments focused on telecommunications (44 per-
cent for all developing countries) and electricity (28 percent). See International Finance Corporation
(2003) Harris (2003), and Kessides (2004), for more details.
References
Agénor, Pierre-Richard. 2004. The Economics of Adjustment and Growth. 2nd ed. Boston, MA: Harvard
University Press.
Agénor, Pierre-Richard, Mustapha K. Nabli, Tarik Yousef, and Henning T. Jensen. “Labor Market
Reforms, Growth, and Unemployment in Labor-Exporting Countries in the Middle East and
North Africa.” 2004. Policy Research Working Paper 3328. The World Bank, Washington, DC.
(June) Journal of Policy Modeling. Forthcoming.
Belloc, Marianna, and Pietro Vertova. 2004.“How Does Public Investment Affect Economic Growth in
HIPC? An Empirical Assessment.” Working Paper 416. University of Siena, Italy. (January)
Calderón, César, and Luis Servén. 2004a. “Trends in Infrastructure in Latin America, 1980-2001.”
Policy Research Working Paper 3401. The World Bank, Washington, DC. (September)
———. 2004b. “The Effects of Infrastructure Development on Growth and Income Distribution,”
Policy Research Working Paper 3400. The World Bank, Washington, DC. (September)
Canning, David. 1998. “A Database of World Infrastructure Stocks, 1950-95.” World Bank Economic
Review 12 (September): 529-47.
Chirinko, Robert S. 1993. “Business Fixed Investment Spending: A Critical Survey of Modelling
Strategies, Empirical Results, and Policy Implications.” Journal of Economic Literature 31
(December):1875-911.
Cohen, Jeffrey P., and Catherine J. M. Paul. 2004. “Public Infrastructure Investment, Interstate Spatial
Spillovers, and Manufacturing Costs.” Review of Economics and Statistics 86 (May): 551-60.
Dhumale, Rahul. 2000. “Public Investment in the Middle East and North Africa: Towards Fiscal
Efficiency.” Development Policy Review, 18 (June): 307-24.
Esfahani, Hadi, and Maria T. Ramírez. 2003. “Institutions, Infrastructure, and Economic Growth.”
Journal of Development Economics 70 (April): 443-77.
Estache, Antonio. 2004. “Emerging Infrastructure Policy Issues in Developing Countries: A Survey of
the Recent Economic Literature.” Policy Research Working Paper 3442. The World Bank,
Washington, DC. (November)
Everhart, Stephen S., and Mariusz A. Sumlinski. 2001. “The Impact on Private Investment of
Corruption and the Quality of Public Investment.” Discussion Paper 44. International Finance
Corporation. (October)
Fay,Marianne, and Tito Yepes. 2003.“Investing in Infrastructure: What is Needed from 2000 to 2010?”
Policy Research Working Paper 3102. The World Bank, Washington, DC. (July)
Ghali, Khalifa A. 1998.“Public Investment and Private Capital Formation in a Vector Error-Correction
Model of Growth.” Applied Economics 30 (June): 837-44.
Hakura, Dalia S. 2004. “Growth in the Middle East and North Africa.” Working Paper 04/56.
International Monetary Fund, Washington, DC. (April)
Harris, Clive. 2003.“Private Participation in Infrastructure in Developing Countries: Trends,Impacts,
and Policy Lessons.” Working Paper 5. World Bank, Washington, DC. (June)

Breaking the Barriers to Higher Economic Growth420
Hulten, Charles R. 1990. “The Measurement of Capital.” In Fifty Years of Economic Measurement, ed.
Ernst R. Berndt and Jack E. Tripplett. Chicago, IL: University of Chicago Press.
_______. 1996. “Infrastructure Capital and Economic Growth: How Well you Use it May Be More
Important Than How Much You Have.”NBER Working Paper 5847.National Bureau of Economic
Research, Cambridge, MA. (December)
International Finance Corporation. 2003. Private Participation in Infrastructure: Trends in Developing
Countries in 1990-2001. Washington, DC: International Finance Corporation.
Jackson, J. Edward. 1991. A User’s Guide to Principal Components. New York: John Wiley and Sons.
Kessides, Ioannis N. 2004. Reforming Infrastructure: Privatization, Regulation, and Competition.
Washington, DC: World Bank,.
Larson, Donald F., Rita Butzer, Yair Mindlak, and Al Crego. 2000. “A Cross-Country Database for
Sector Investment and Capital.” World Bank Economic Review 14 (May): 371-91.
Ligthart, Jenny E. 2000. “Public Capital and Output Growth in Portugal: An Empirical Analysis.”
Working Paper 00/11. International Monetary Fund, Washington, DC. (January)
Mansouri, Brahim. 2004. “Déséquilibres Financiers Publics, Investissement Privé et Croissance au
Maroc.” Unpublished. Marrakesh: Cadi Ayyad University. (March)
Mittnik, Stefan, and Thorsten Neumann. 2001. “Dynamic Effects of Public Investment: Vector
Autoregressive Evidence for Six Industrialized Countries.” Empirical Economics 26 (June): 429-36.
Nabli, Mustapha K., and Jennifer Keller. 2002. “The Macroeconomics of Labor Market Outcomes in
MENA over the 1990s: How Growth has Failed to Keep Pace with a Burgeoning Labor Market.”
Unpublished. World Bank, Washington, DC. (April)
Nehru, Vikram, and Ashok Dhareshwar. 1993. “A New Database on Physical Capital Stock: Sources,
Methodology, and Results.”Revista de Analisis Economico 8 (June): 37-59.
Page, John. 1998. “From Boom to Bust-And Back? The Crisis of Economic Growth in the Middle East
and North Africa.” In Prospects for Middle Eastern and North African Economies: From Boom to Bust
and Back? ed. Nemat Shafik. New York: St. Martin’s Press.
Pritchett, Lant. 1999.“Has Education Had a Growth Payoff in the MENA Region?” Discussion Paper
18, MENA Region. World Bank, Washington, DC. (June)
Richards, A., and J. Waterbury. 1996. A Political Economy of the Middle East. 2nd ed. Boulder, CO:
Westview Press.
Shafik, Nemat. 1992. “Modeling Private Investment in Egypt.” Journal of Development Economics 39
(June): 263-77.
Turnovsky, Stephen J. 1996. “Fiscal Policy, Adjustment Costs, and Endogenous Growth.” Oxford
Economic Papers 48 (July): 361-81.
Voss, Graham M. 2002. “Public and Private Investment in the United States and Canada.” Economic
Modeling 19 (August): 641-64.
World Bank. 1994. Investing in Infrastructure. World Development Report. New York: Oxford
University Press.
———. 1995. Claiming the Future: Choosing Prosperity in the Middle East and North Africa.
Washington, DC: World Bank.
———. 2003a. Jobs, Growth and Governance in the Middle East and North Africa-Unlocking the
Potential for Prosperity. Washington, DC: World Bank.
———. 2003b. Trade, Investment and Development in the Middle East and North Africa. Washington,
DC: World Bank.
———. 2004. Unlocking the Employment Potential in the Middle East and North Africa: Toward a New
Social Contract. Washington, DC: World Bank.

Part III: Trade, Competitiveness, and Investment 421
Annex 16A. Data Sources and Definitions
This annex provides a brief description of the data used in this study. The
actual dataset is available upon request.
Gross fixed capital formation by the private sector in percent of GDP: For Egypt,
the data source is International Finance Corporation (IFC) for 1982-99, and
World Development Indicators (WDI) for 2000-02. For 1965-81, the shares of
total public and private investment are applied to total fixed capital formation
to construct the private fixed capital formation investment series. For Egypt,
series on total public and private investment for 1965-81 were provided by the
Bank’s desk economist. For Jordan and Tunisia, data source is WDI.
Claims on private sector in percent of GDP: data source is IFS.
Growth rate of the real effective exchange rate (REER): defined as the log differ-
ence of the REER. For 1980-2002, data source is IFS. Before 1980, the REER is
calculated as the ratio of the nominal exchange rate times the unit value of
imports, divided by the consumer price index.
Growth rate of real GDP: defined as the log difference of real GDP. Data source
is WDI.
Gross fixed capital formation by the public sector as percent of GDP: For Egypt,
data source is IFC for 1982-99, and WDI for 2000-02. For 1965-81, the shares
of total public and private investment are applied to total fixed capital forma-
tion to construct the series. Series on total public and private investment for
1965-81 were provided by the Bank’s desk economist. For Jordan and Tunisia,
data source is WDI.
Public capital expenditure infrastructure: Data source is the IMF’s Government
Finance Statistics (GFS). It is obtained by adding fuel and energy; railway, air,
pipeline, and other transportation; water transport; transportation and com-
munication; and other transportation and communication. This series is
deflated by the GDP deflator to calculate a series at constant prices.
Public infrastructure capital stock at constant prices: calculated using the perpet-
ual inventory method, using a uniform depreciation rate of 2.5 percent.
Adjusted public infrastructure stock at constant prices: obtained by multiplying
the raw stock data by the composite quality index for public infrastructure,
normalized at unity in the first period of estimation.

Breaking the Barriers to Higher Economic Growth422
Composite quality index: defined as the weighted average of the two principal
components, using as weights the proportion of total variance explained by
each component. The principal components are calculated over the two sep-
arate indicators (ICOR-based and “excess demand” measures) described in
the text.

423
Governance, Institutions, and
Private Investment
An Application to the
Middle East and North Africa
Ahmet Faruk Aysan*
Mustapha Kamel Nabli
Marie-Ange Véganzonès-Varoudakis†
17
During the 1980s and 1990s, private investment in the Middle East and North
Africa (MENA), on average, showed a decreasing trend. With the liberaliza-
tion of economies and the acceleration of reforms, private investment
increased throughout much of the world in the 1990s. The Middle East and
North Africa countries did not follow this pattern. While private investment
to GDP declined by 2.4 percent in the region, this rate increased by 4.8 per-
cent in Latin America and the Caribbean (LAC), 15.8 percent in Africa (AFR),
23.4 in South Asia and 14 percent in East Asia (EAP), despite the financial
crisis.
This paper addresses the issue of the low level of private investment in the
MENA region, with special emphasis on the role of governance. In fact, the
MENA countries have on average been characterized by a clear deficit of
“good” governance institutions, particularly as regards democratic institu-
tions such as political rights, civil liberties, and freedom of the press. Similarly,
the quality of the administration has also been of some concern. These defi-
*Bog˘aziçi University, Istanbul, Turkey. †Centre d'Etudes et de Recherches sur le Développement
International and Centre National de la Recherche Scientifique, Université d’Auvergne, France.
Published in The Developing Economies XLV-3 (September 2007), 339–77. Used by permission from
Blackwell Publishing Ltd.

Breaking the Barriers to Higher Economic Growth424
ciencies have been reported as being responsible for the slow economic activ-
ity in MENA (see El Badawi 2002; and World Bank 2003).
These results are in line with a growing literature on governance that sug-
gests that successful market-based economies need “good” governance insti-
tutions.1However, the channels investigated to date concern more the impact
of governance on economic growth,2GDP per capita,3and volatility of eco-
nomic activity4; little has been empirically done to analyze the effects of insti-
tutions on private investment. As far as private investment is concerned, the
literature on the role of governance has essentially focused on the effect of the
rule of law—more specifically, on the security of property rights, which is the
best documented effect and the best supported by empirical evidence.5Other
components of governance are not as well documented and the empirical val-
Figure 17.1. Private Investment by Region
(% GDP)
Figure 17.2. Private Investment in MENA Countries
(% GDP)
20
18
16
14
12
10
8
6
4
2
0
Sub-Saharan
Africa
East Asia
and Pacic
Europe and
Central Asia
Latin America
and the
Caribbean
Middle East
and
North Africa
South
Asia
1980s
1990s
20
18
16
14
12
10
8
6
4
2
0
Egypt Iran Morocco Tunisia
1980s
1990s
Source : Authors’calculations.
Source : Authors’calculations.

Part III: Trade, Competitiveness, and Investment 425
idation is not as successful. This is not only the case for corruption and
bureaucratic quality,6but also for democratic participation, research into
which lacks theoretical and empirical precision.7A few studies have, however,
addressed more successfully the links among political instability, policy uncer-
tainties, and firms’ decisions to invest.8
Governance, however, is part of the investment climate of a country.
Investment decisions are mainly driven by profitability motives (Jorgenson
1963). The forward-looking nature of investment underlines the importance
of a stable and secure environment—in particular the security of property
rights. At the same time,“good” governance institutions are viewed as reduc-
ing uncertainty and promoting efficiency (see North 1981). In this respect,
and as reported by World Bank (2003), better governance improves the invest-
ment climate by improving bureaucratic performance and predictability. This
in turn reduces uncertainty, as well as the cost of doing business. Better gov-
ernance also contributes to the effective delivery of public goods that are nec-
essary for productive business. Cross-country correlations using broad prox-
ies for investment climate quality suggest a positive link between the invest-
ment climate and private investment decisions.9decisions.9
As a part of the reflection on the channels through which governance may
affect economic performance, one strand of the literature has recently recon-
sidered the role of economic policies in explaining cross-country economic
achievements. This research has also been important to our interest in the
determinants of private investment. Recent work on the role of both gover-
nance and economic policies has found that governance institutions are the
dominant factor, with little, if any, independent influence by policies.10
However, these results stem from endogeneity and specification problems, as
pointed out by Sachs (2003). In fact, economic policies are likely to affect
cross-country variations in governance quality. There is, in particular, some
evidence that greater openness to trade and stronger competition are con-
ducive to better governance.11 Given these conditions, economic policies may
explain economic performances through their impact on governance.12 In the
case of private investment, we show that both direct and indirect effects can
be brought together if the model chosen is well specified.13 We estimate in
particular a simultaneous model of private investment and of various forms
of governance institutions, where economic policies concurrently explain
both variables.
In this paper, we also address other shortfalls of the empirical literature on
governance and economic performances. We intend, in particular, to catego-
rize what types of governance institutions are more detrimental to entrepre-
neurial investments. To this end, we introduce a large set of governance vari-
ables that are not typically used in the literature on determinants of private
investment. Since these indicators are likely to be correlated, we process a few
aggregated indicators, using the principal component analysis methodology.

Breaking the Barriers to Higher Economic Growth426
Based on the existing literature, we categorize so-called governance institu-
tions in three broad clusters: “Administrative Quality” (QA), “Public
Accountability” (PA), and “Political Stability” (PS). We also generate a global
indicator of governance (GOV) that summarizes these three aspects14. This
method allows a second step to evaluate the contribution of the initial indica-
tors to the private investment decision. This information will be useful in
understanding which factors explain the low investment performance of the
region. In this paper, human development and economic policy variables have
been similarly processed.
Our empirical approach relies also on panel data (cross-section-time series
analysis) that is suitable—contrary to previous studies—to jointly assessing the
impact of economic policies and governance institutions on private invest-
ments. The time series dimension captures the variability of policies through
time, and the cross-section dimension covers the governance variables, which
tend to evolve slowly. This paper also acknowledges the deficiencies of existing
data on governance as subjective and outcome-based, rather than representing
the quality of actual institutions. However, this paper points out that these defi-
ciencies do not constitute a severe problem in analyzing the effects of gover-
nance on private investment.Whether actual or not, what is important for pri-
vate investors is the perceived quality of governance at the time of investment.
This paper also benefits from a newly constructed data set on private invest-
ment. The ultimate purpose of this paper is to determine which factors are cen-
tral in the decision-making process of private entrepreneurs to invest. However,
because of the lack of comparable data on private investment, most of the ear-
lier studies use aggregate investment as a proxy for private investment. Later, the
International Finance Corporation (IFC) of the World Bank launched a project
that addressed the private investments of various developing countries from
1970 to 1999. Studies using this disaggregated data on private investment show
that private and public investment can have very different determinants
(Aizenman and Marion 1999). Therefore, disaggregated data need to be utilized
in order to capture the investment-conducive factors in today’s globalized
economies. Building on the IFC series, our new data set covers 99 countries (60
for high-quality data) over the period 1970-2002 (see Annex 17.1).
Finally,as previously mentioned, our empirical model allows for the simul-
taneous estimation of private investments and of various types of governance
institutions. This model is also justified by the fact that—in addition to eco-
nomic policies—changes in private investment can influence the quality of
governance. Moreover, some hidden factors are likely to affect private invest-
ment and governance institutions in parallel. Our empirical results show that
governance institutions play a significant role in private investment decisions.
This result is particularly true in the case of “Administrative Quality” (QA), in
the form of control of corruption, bureaucratic quality, investment-friendly

Part III: Trade, Competitiveness, and Investment 427
profile of administration, and law and order. Similarly, “Political Stability”
(PS) significantly generates higher private investment. Evidence in favor of
“Public Accountability” is also found.
This paper is organized as follows: The second section introduces our clas-
sification of governance institutions to determine which institutions are detri-
mental to entrepreneurial investments. The third section presents the other
determinants of private investment that will be taken into consideration in
our empirical analysis, and highlights the importance of these factors for the
MENA countries. The fourth section presents the characteristics of the data
used. The fifth section introduces the private investment model tested and the
results of the estimations. The sixth section uses this model to determine
which factors would boost the level of MENA’s private investments in the
future. The last section presents the conclusions.
Governance Institutions: An Attempt at Classification
A first step in our analysis of the link between governance and private invest-
ment has been to differentiate and categorize the numerous different dimen-
sions of governance, to better understand which institutions are investment-
conducive. Existing literature on the classification of governance institutions
provides some alternatives. Various authors have aggregated certain indices to
capture better the common features of the existing data.
Kaufmann, Kraay,and Mastruzzi (2003) categorize governance institutions
in six broad groups.Their measures of governance are based on 194 variables
drawn from 17 different sources in order to measure six different aspects of
governance. “Government Effectiveness” and “Regulatory Quality” summa-
rize the ability of the government to formulate and implement sound policies.
The respect of citizens and the state for the institutions that govern their inter-
actions is categorized as “Rule of Law” and “Control of Corruption.”“Political
Stability and Absence of Violence” measures perceptions of likelihood that the
government in power will not be destabilized, and indicate the continuity of
policies. “Voice and Accountability” captures the process by which citizens of
a country are able to participate in the selection of their government (see
Annex 17.2 for more details on these indicators). The World Bank (2003) has
used two indices on “Public Accountability” and “Administrative Quality” by
aggregating the existing relevant data sets for these features of governance.
Our choice of indicators has been limited by the lack of annual data avail-
able for a large sample of countries over long periods of time. Considering the
existing classifications of governance data, this paper categorizes the gover-
nance variables that are likely to affect individual investors’ decision into three
broad clusters: “Administrative Quality” (QA), “Public Accountability” (PA),
and “Political Stability” (PS).

Breaking the Barriers to Higher Economic Growth428
Quality of Administration
The first set of candidates is intended to provide information on the ability of
government to deal with investors and to provide them with an investment-
friendly and reliable context in which to conduct their investment projects.
Following World Bank (2003), we have defined the first governance variable
as the “Quality of Administration.” This variable incorporates four indicators
from the International Country Risk Guide (ICRG, 1999), namely: (i)
“Control over Corruption;” (ii) “Quality of Bureaucracy;” (iii) “Investment
Profile;”and (iv) “Law and Order”(see definitions of variables in Annex 17.3).
These institutions are part of the investment climate of a country. They pro-
mote investments by reducing the costs and risks of doing business.
Corruption often has adverse effects on economic activities. This fact is well
documented and is often described as one of the major constraints facing
enterprises in the developing world (World Bank 2002). In his cross-country
analysis, Mauro (1995) shows that corruption reduces growth. Gupta,Davooli,
and Alonso-Terme (2002) stress that corruption exacerbates income inequali-
ty and poverty. Mo (2001) documents a causal chain of interest for our work,
linking corruption to low growth through reduced human and physical capi-
tal. In fact, for private investors,corruption increases investment and operation
costs, as well as uncertainties about the timing and effects of the application of
government regulations. Corruption also raises the investment and opera-
tional costs of public enterprises, leading to insufficient and low-quality infra-
structures that are detrimental to private investment. (see Tanzi and Davooli
1997). The same conclusions have been reached about the effects of bureau-
cratic quality on economic activity (see Evans and Rauch 2000).
The “Quality of Bureaucracy” index of ICRG summarizes the ability of the
government to formulate and implement sound policies. Moreover, the
“Quality of Bureaucracy” index indicates that “countries where the bureaucra-
cy has the strength and expertise govern without drastic changes in policy or
interruptions in government services. In these low-risk countries, the bureau-
cracy tends to be somewhat autonomous from political pressure and to have
an established mechanism for recruitment and training.” (ICRG, 1999)
The “Investment Profile” is a measure of the “government’s attitude to
inward investment as determined by the assessment of four subcomponents:
risk to operations, taxation, profit repatriation, and labor costs.” Because
investors are making long-term decisions, risks to operations and other
uncertainties about future policies are detrimental to investment decisions.
Taxation and labor costs have also a first-order implication on costs, and
therefore on decisions to invest. Although government regulations and taxa-
tion are reasonable and warranted in order to protect the general public, and
to generate revenues to finance the delivery of public services and infrastruc-
tures, overregulation and overtaxation deter investments by raising business
startup and operating costs.

Part III: Trade, Competitiveness, and Investment 429
In the “Law and Order” index, the “Law” subcomponent provides an “assess-
ment of the strength and impartiality of the legal system,” while the “Order”
subcomponent concerns the “popular observance of the law.” Although many
aspects of the business environment affect investments, the security of proper-
ty rights is the most important and the better documented issue. Because of the
forward-looking nature of investment, investors need institutions that preserve
the right of private property, ensure equitable and consistent rule of law in pro-
tecting this right, and offer effective incentives to respect and enforce it. A reli-
able judiciary, in particular, reduces transaction costs for businesses and sends
positive signals to investors that rules of law will be equitably and consistently
protected and enforced. On the empirical side, the issue of property rights and
of rule of law has been widely covered by the literature, and the results of cross-
country analysis are robust to various tests and specifications.15
Public Accountability
The second set of indicators measures “Public Accountability.” This index
consists of two indicators from Freedom House (FH): “Civil Liberties” and
“Political Rights.”
Public accountability is part of the investment climate of an economy.
Because fixed capital investments are generally irreversible, private invest-
ment decisions are highly sensitive to the perception of the credibility and
tenacity of the political regime, as well as of policies.16 An open and partici-
patory political system provides stability of social institutions and ensures
broad public support for policies, which are in this case more sustainable in
the long run. Public accountability is a guarantee of transparency and of bet-
ter availability of information, which also help governments build credibili-
ty. Public accountability provides access to policy makers and can hold them
responsible for failures in implementing policies. In particular, freedom of
press, free political parties, and open elections contribute to government’s
legitimacy and give voice to citizens in the decision-making process. On the
empirical side, the literature on democratic participation has focused on the
effects of transparency and accountability on growth, using data on civil lib-
erties, political rights, and freedom of the press from various sources. The
empirical validation has, however, produced mitigated success.17 The work of
Pastor and Sung (1995) is one of the few to show a positive effect of various
indicators of democratic institutions on private investment in the developing
world.
Political Stability
The last set of variables is intended to measure “Political Stability.” Political
instability increases the uncertainty in the economy and deters risk-averse
entrepreneurs from taking action for profitable investment opportunities. The
political stability index includes the following variables from ICRG (1999):

Breaking the Barriers to Higher Economic Growth430
“Government Stability,” “Internal Conflict,” “External Conflict,” and “Ethnic
Tensions.” Various authors, using different indicators of political uncertain-
ties, have brought empirical evidence that institutions associated with politi-
cal instability hamper aggregate investment.18
All the political and governance indicators have been aggregated by using
principal component analysis (PCA) to account for the multicollinearity issue
in using these potentially correlated variables in the same regression equation.
In addition, we have generated a global indicator of governance (GOV), which
summarizes the information contained in the three previous indicators (QA,
PA, and PS). Results of PCA are given in Annex 17.4.
Other Determinants of Private Investment
Although the importance of private investment has been widely discussed in
the literature, there is less evidence on what induces private firms to invest in
developing countries. In fact, developing countries do not operate in a com-
petitive environment, and face constraints that are not accounted for in the
neoclassical model. This partly explains why most economists do not agree on
the subject of the determinants of investment in developing countries.19 This
phenomenon is also the case for MENA economies, for which the empirical
literature is deficient.20 In this paper, we address some of these constraints, in
particular the ones linked to economic policy and to the quality of governance
institutions.
The Neoclassical Accelerator Model
In the macroeconomics literature, the neoclassical flexible accelerator model
is the most widely accepted model of investment. This model is based on the
neoclassical idea of the theory of the firm (Jorgenson 1963), which postulates
that enterprises decide to invest so as to generate more profit in the future.
The investment function is derived from the optimization problem of the
firms, which maximize current and expected profits by equating the produc-
tion prices to their marginal costs. Firms will invest so long as the marginal
benefit of doing so outweighs the additional cost. The net investment is the
gradual adjustment of the actual capital stock to its desired level, which is
derived from maximization of profit. The determinants of investment in the
neoclassical flexible accelerator model include the expected aggregate
demand (the accelerator), the user cost of capital, the wage rate, and the ini-
tial capital stock.
This model postulates, however, that firms operate in competitive markets,
which contradicts the structural and institutional factors prevailing in devel-
oping countries. Even though the empirical tests of the model appear to be
successful for several developed countries, the firms in developing countries
face certain constraints that are not accounted for in the conventional neo-

Part III: Trade, Competitiveness, and Investment 431
classical theory.21 Some of these constraints (in particular those of specific
interest for the MENA region) are discussed below.
Structural Reforms
Among the most common constraints faced by developing countries is the
deficit of economic reforms. This is the case in the MENA countries, which
have lagged behind other regions in terms of reforming their economy (Nabli
and Véganzonès-Varoudakis 2007). Structural reforms constitute an impor-
tant determinant of the actual and future profitability of private investment.
We have considered trade policy and financial development as part of our
structural reforms index.
By providing more opportunities and incentives for firms to invest, finan-
cial development is an important part of private investment decisions. A
developed financial system mobilizes and allocates resources to enterprises. A
developed financial system is also expected to be more efficient due to increas-
ing technological specialization, which leads to a better selection of projects
and more advanced diversification of risks. This allows firms to finance more
investment projects, and increases the productivity of new investments (see
Levine 1997 for a synthesis). In addition, given the lack of well-functioning
financial markets, the neoclassical assumption of the flexible accelerator
model regarding the availability of credit supply by the banking sector cannot
be taken for granted in developing countries. This discrepancy also occurs
because of public deficits and public debt, which can lead to financial repres-
sion and to the eviction of private investment. On the empirical side, the
impact of financial development on private investment is now well docu-
mented.22 In his survey of investment functions in developing countries,
Rama (1993) presents the positive effect of financial development on private
investment in 21 of the 31 papers surveyed.
Trade reforms constitute another factor that can stimulate private invest-
ment decisions. Trade openness increases competitiveness and provides access
to enlarged markets (Balassa 1978; Feder 1982). Trade openness can be at the
origin of economies of scale and of productivity gains. In addition, trade open-
ness influences the availability of external credit—considering the general con-
sensus on the role of tradable goods in providing positive externalities in the
form of collateral for external financing (Caballero and Krishnamurthy 2001).
All these factors create favorable conditions for enterprises to invest.
However, as mentioned in the introduction, economic reforms are also
expected to affect private investment through their impact on the quality of
governance institutions. There is, in particular, some evidence that greater
openness to trade and stronger competition are conducive to governance
improvement.23 Opening up markets may help to weaken vested interests and
reduce rents derived from prevailing economic and institutional arrange-
ments. Trade openness may also lead to demands for governance institutions

Breaking the Barriers to Higher Economic Growth432
more suited to an increasingly varied and complex range of transactions (see
IMF 2003).
Human Capital
Human capital is part of the investment climate of an economy and is gener-
ally considered as complementary to physical capital. Here, we have consid-
ered health and education as part of the human capital index. Human capital
stimulates private capital formation by raising the profitability of investment.
Human capital can also be at the origin of positive externalities.24 Because
skilled workers are better at dealing with change, a skilled work force is essen-
tial for firms to adopt new and more productive technologies.25 Besides, new
technologies generally require significant organizational changes, which are
handled better by a skilled workforce.26 Human capital also gives the oppor-
tunity to enterprises to expand or enter new markets.
Moreover, human capital entails better governance institutions. Better edu-
cated people with higher life expectancy become more competent bureaucrats
and—in addition to better monitoring of the functioning of government
officials—demand better quality of bureaucracy (Galor et al. 2005). In addi-
tion, educational attainment reduces political instability by generating more
avenues to reconcile opposing parties. This idea constitutes one of the classi-
cal approaches in the literature to highlight the importance of education in
bringing better governance institutions (Lipset 1959). From the democratic
accountability point of view, a more educated society is more likely to be
enfranchised in terms of civil rights and liberties (Acemoglu and Robinson
2001). These considerations justify that human capital also appears as an
explanatory factor in private investment decisions, through its impact on the
quality of governance institutions.
Although educational attainment has improved in the majority of develop-
ing countries, many firms still rate inadequate skills and education of workers
as severe obstacles to their operations (see World Bank 2002). This is the case
in the MENA region, where progress is still needed in order to catch up with
South East Asia and Latin America.27 To meet this challenge, and as pointed
out by World Bank (2003), MENA countries have to gear up their education-
al systems, both to improve basic education and to equip the labor force with
skills appropriate for enterprises to invest efficiently.
Some Considerations on the Data Used
A New Data Set on Private Investment
Data on the breakdown of investments as either private or public are scarce.
The best available data set on private investment was provided by the
International Finance Corporation of the World Bank.28 However, this data
set covers only the period from 1970 to 1999. It also has some limitations for

Part III: Trade, Competitiveness, and Investment 433
certain countries in terms of the quality of the breakdown of public and pri-
vate investment. This is most often because of the status of state-owned enter-
prises, for which investment data are not always available, and which are in
this case included in the private investment series.
Considering these pitfalls of existing data, we have re-examined the IFC
data set and updated the private investment series for the available years after
1999. More important, we have carefully checked the IFC private investment
series and compared them to the national sources, where available, as well as
to the World Bank and International Monetary Fund series. This has been
done in close collaboration with the country economists of these two institu-
tions. We have thus been able to generate high-quality data for the majority of
our sample countries, as well as for other countries, which can be considered
as relatively accurate. This private investment series, which covers 60 coun-
tries, has been used in the empirical analysis (see in Annex 17.1 the list of
countries). We have also generated a broader set of private investment series,
which includes countries for which the distinction between private and pub-
lic investment is not as satisfactory, but which can be used for robustness
analysis. This data set covers 99 countries.
Nature of Data on Governance
The data usually used in governance are produced by independent, private
firms, which provide consulting services to international investors such as the
International Country Risk Guide, the Heritage Foundation (HF), Freedom
House, or the Fraser Institute (FI). To a certain extent, these indices provide
very similar information on various aspects of governance. These data sets
have certain common features. First of all, they can be considered to be sub-
jective. They measure perceptions of governance quality rather than actual
quality. They also measure outcomes rather than actual rules (see Glaeser et
al. 2004). Finally, given that governance institutions do not change easily in
theory, institutional indices are supposed to be rather persistent, even though
they are relatively volatile in existing data sets.
All these factors appear counter to using the governance indices commonly
used in the literature. In fact, these characteristics are very useful in determin-
ing investor perceptions of the quality of governance at the time of their invest-
ment. Indeed, what we are more concerned about is not actual governance qual-
ity per se, but its perception by the private sector, since our ultimate aim is to
identify the determinants of private investment. This paper strongly shows that,
in addition to the conventional determinants of private investment, governance
institutions—whether perceived or real—are detrimental to investors’ decision
to invest. Hence, we allow the possibility that perceived institutions differ from
actual institutions. It is quite possible that in our framework, even though qual-
ity of actual governance is not high, private investors tend to perceive that their
investment projects are protected by good institutions or vice versa.

Breaking the Barriers to Higher Economic Growth434
To illustrate this idea, it can be noticed that following economic crises, gov-
ernance indices of crises-hit countries can vary enormously. It is hard to
believe that institutions change drastically in a short period of time, but it is
reasonable to argue that perceptions of governance institutions, in the eyes of
beholders, that is, investors, are altered through the crises. Investors modify
their expectations of institutions when new information is revealed in a
crises-hit economy.
One explanation for these drastic upheavals in the governance indices of
crises-hit countries is that investors definitely have incomplete and asymmet-
rical information on the quality of governance institutions in the economy.
During normal times, when business runs as usual, information on the qual-
ity of governance is not widely noticed. However, with the advent of crises or
new information, governance institutions face a real examination. Hence, the
manner in which countries handle new conditions can influence the percep-
tion of governance by private investors. This information also accumulates
over time and provides a basis for long-term perceptions of governance qual-
ity throughout the countries.
The advanced countries of today have built their investor-friendly and per-
sistent governance institutions over long periods of time, after successfully
passing certain historical tests. However, in the short term, investors make
their judgments on the quality of governance based on a numerous factors.
These factors certainly include some historical episodes experienced by the
country that can provide insight into the future potential performance of
existing governance institutions. Debt repudiation or the state’s appropriation
of private property in the past, for example, definitely are taken into account
when entrepreneurs assess the quality of a country’s governance. However, in
addition to this type of backward-looking behavior, entrepreneurs’ perception
of the quality of governance is also shaped by existing conditions and those
anticipated in the future. In this regard, we argue that existing indices meas-
uring the quality of governance capture investors’ concerns about the institu-
tions quite well.
The Econometric Analysis
The Model Tested
The primary purpose of the model tested is to disentangle the effects of gov-
ernance on private investment. More important, in this paper we want to
make a horse race among the different types of governance variables com-
monly used in the literature, to distinguish the ones most vital in accelerating
private investment in our sample of developing countries.
In the empirical model, endogenous variables are the share of private
investment and the various measures of governance, namely: QA, PA, PS, and
GOV. These endogenous variables are simultaneously determined by influ-

Part III: Trade, Competitiveness, and Investment 435
encing each other. In order to account for this reverse causality, we establish a
system of equations to estimate the share of private investment (PI) in GDP
and quality of governance institutions (QI) simultaneously. In the private
investment equation, a lower ranking for quality of governance institutions is
expected to reduce private investment. In the governance equation, private
investment enters on the right side with an expected positive sign. This simul-
taneous system of equations also enables us to take into account other factors
that affect both private investment and governance institutions.
This system of equations is estimated using three-stage least squares (3SLS)
by controlling other determinants of endogenous variables. Three-stage least
square estimation allows us to use the links between endogenous variables
efficiently. Because endogenous variables appear as regressors in other equa-
tions, they have to be instrumented out using exclusion restrictions. Initially,
3SLS regressions are run separately for QA, PS, and PA. However, to complete
the analysis, we have substituted, in this system of equations, the aggregate
GOV, which is calculated as the principal component analysis of all the initial
indicators, and which provides a summary of the three measures of
governance.
The model estimated is the following:
(1)
(2)
where
PIit is the share of private investment in GDP
QIit represents the various indexes of governance (QA,PA,PS, and GOV)
X1i and X2i are the other control variables in private investment (PI) and gov-
ernance (GOV) equations respectively
ε1it and ε2it are the error terms of each equation. iindicates the country and t
represents the time of the variable.
The determinants of private investment in the neoclassical flexible acceler-
ator model include the expected aggregate demand (the accelerator) and the
user cost of capital. Hence, the private investment equation in our specifica-
tion incorporates real interest rate to capture the user cost of capital. It also
accounts for the GDP growth rate in last year to control for the accelerator
effect. These two variables are excluded from the governance equation (QI) in
order to identify the system.
Both of the equations, on the other hand, take into account the GDP per
capita, as well as the variations in structural reform (SR) and human capital
(H). Structural reforms are proxied through trade policy (TP), and financial
development. Financial development is proxied by private credit by banks and
other depository institutions. Trade policy is constructed as commercial
QI PI X
it it i it
=+ + +γβ β ε
01 22 2
PI QI X
it it i it
=+ + +αα α ε
01 211

Breaking the Barriers to Higher Economic Growth436
openness (calculated by aggregating the export and import in total GDP),
from which we have subtracted the exports of oil and mining products, as well
as the “natural trade openness” constructed by Frankel and Romer (1999).
The trade policy and financial development variables form the structural
reform indicators, after implementation of the principal component analysis
(see Annex 17.4 for results of PCA). Structural reform is expected to stimulate
private investment, as well as institutional change for the better.
Human capital (H) is expressed through life expectancy at birth, and aver-
age years of primary, secondary, and higher schooling in the total population
over 15 years old. These variables are also aggregated with principal compo-
nent analysis. Human capital is widely considered to enhance private invest-
ment and to lead to better governance institutions.Therefore, the human cap-
ital variable is expected to have positive coefficients in both of the equations.
GDP per capita is controlled in the investment equation to account for the
neoclassical Solow growth model. Countries with lower GDP per capita are
expected to gradually catch up with their more developed counterparts by hav-
ing more capital investment over time. Moreover, GDP per capita accounts for
possible externalities, such as greater market size, on demand and supply of
good and services, and finally on private investment. GDP per capita in gover-
nance equations represents the idea that more developed countries can afford to
have better governance institutions (Azariadis and Lahiri 2002). Hence, a posi-
tive relationship is expected between GDP per capita and governance quality.
Oil export as a percentage of total merchandise export also enters into both
equations. The typical natural curse hypothesis is taken into account by incor-
porating this variable into the investment equation. When a country relies
more on natural resources extraction in its exports, there can be less incentive
to invest in other products. This result, for example, may stem from the
increase in the cost of labor (Rodriguez and Sachs 1999). This variable also
has an implication for the quality of governance institutions. Countries with
less reliance on natural resources are expected to form better governance insti-
tutions. The natural resource-abundant countries do not need to mobilize the
society to enhance aggregate income. The ruling class can control the econo-
my by collaborating with a small number of people in the society. Therefore,
the production structure of the country does not generate good governance
institutions in favor of society (Ross 2001; Bellin 2001). Under these circum-
stances, the elite are also less inclined to provide better governance by consid-
ering the future effects of today’s enfranchisements (Acemoglu and Robinson
2001) and engage in more rent-seeking activities.29 Hence, the share of oil
exports in merchandise exports is expected to reduce the quality of gover-
nance institutions.
The tenure of the system from Keefer et al. (2001) is excluded from the
investment equation to identify the system of equations. Tenure of the system
reports the number of years that an administrative system—regardless of

Part III: Trade, Competitiveness, and Investment 437
whether autocratic or democratic—lasts in the country. The underlying idea
to include this variable in the governance equation is to account for the fact
that institutions settle over time. The longer time passes with the existing sys-
tem, the better institutions are established. This exclusion restriction is quite
reasonable considering that tenure of the system has a direct impact on the
governance institutions, whereas its influence on private investment is more
likely to be realized through its effect on these institutions.
Finally, a regional dummy for the MENA countries appears as a right-
hand-side variable in both of the equations. One of our primary purposes is
to understand the position of MENA countries among the other countries
and to see whether MENA substantially diverges from the rest of the world in
terms of private investment and of governance performance.
Estimation Results
Equations (1) and (2) have been estimated on an unbalanced panel of 31 devel-
oping countries over 1980-2002, using the 3SLS estimations technique. Four
sets of regressions have been conducted, each one with a different indicator of
governance. Table1 presents the estimation’s results of equations (1) and (2)
when QA, PS, PA, and GOV are taken into consideration, respectively.
Administrative Quality. In Table 17.1—when QA is used as a measure of
governance—estimation results produce quite interesting conclusions. One of
the most interesting outcomes concerns the QA index, which gives a positive
and significant coefficient at the 5 percent level in the investment Equation
(1). This result confirms that a low level of corruption, good-quality bureau-
cracy, clear security of property rights, reasonable risk to operations, sound
taxation and regulation, as well as more effective law and order, are of first
importance for enterprises’ decisions to invest. This result makes a real contri-
bution to the empirical literature on governance by validating, over a relative-
ly long period of time, the role of a large set of governance variables on pri-
vate economic performance.
Our result is unambiguous and robust to the introduction of other
explanatory variables. This is the case of structural reforms and human capi-
tal. The roles of these variables in explaining cross-country economic achieve-
ment have recently been questioned (Easterly and Levine 2003). Our regres-
sion results indicate the significant impact of these variables on private invest-
ment decisions. Hence, our estimations stress that, although the quality of
governance constitutes a major factor in private sector decisions, the role of
economic policies cannot be disregarded. Our result also confirms that firms
in developing countries face constraints that are not accounted for in more
developed economies and that deficiencies in trade policy, financial develop-
ment, and education have a long-term impact on private investment decisions
and growth.

Breaking the Barriers to Higher Economic Growth438
Another conclusion of our model consists in validating the neoclassical
theory of the firm in the case of developing countries. The accelerator variable
has the expected positive sign, which implies that anticipations of economic
growth induce more investment. Similarly, the interest rate appears to exert a
negative and significant effect on private investment, which is consistent with
the user cost of capital theory. Both variables are highly significant, indicating
that at the final stage, supply and demand considerations constitute major fac-
tors for entrepreneurs to undertake a new investment project. However, our
model fails to verify the Solow hypothesis of decreasing return to scale of
physical capital accumulation. The coefficient of the GDP per capita variable,
although negative, is not significant.
Table 17.1. Estimation Results
Endogenous Endogenous Endogenous
variables variables variables
Explanatory Priv Inv QA Priv Inv QA Priv Inv PS
variables (1) (2) (3) (4) (5) (6)
QA 1.99 2.07
(1.98)** (2.06)**
PS 3.51
(1.67)*
PA
GOV
Private 0.1 –0.065
investment (2.16)** (–1.61)
Structural 1.64 0.11 1.64 0.31 1.07 0.5
reforms (4.73)*** (–1.04) (4.75)*** (9.37)*** (–1.45) (4.86)**
Human 0.62 –0.02 0.57 0.04 –0.1 0.25
capital (2.82)*** (–0.46) (2.63)*** (–1.08) (–0.2) (5.21)***
Oil –0.03 0.0003 –0.03 –0.003 –0.05 0
exports (–2.66)*** (–0.11) (–2.91)*** (–1.65)* (–2.86)*** (–0.10)
GDP per 0 0.0001 0 0.0001 0 0.0002
capita (–0.28) (3.54)*** (–0.01) (4.1)*** (–0.42) (4.04)***
MENA –1.2 0.15 –1.1 0.06 –2.3 0.34
dummy (–1.21) (–0.84) (–1.11) (–0.38) (–1.75)* (1.84)*
Rear –0.02 –0.036 –0.05
(–2.06)** (–3.39)** (–2.53)**
Growth 0.22 0.2 0.22
(3.29)*** (2.88)** (3.09)***
Ten syst 0.02 0.017 0.013
(4.36)*** (6.19)*** (3.71)***
Constant 11.8 –1.53 11.8 –0.52 14 –0.2
(16.95)*** (–3.20)*** (16.92)*** (–6.52)*** (7.48)*** (–0.4)
Numb obs 349 349 349 349 349 349
Note: (*) indicates significance at 10%; (**) indicates significance at 5%; (***) indicates significance at 1%. See sources of data in footnote 30.

Part III: Trade, Competitiveness, and Investment 439
Endogenous Endogenous Endogenous
variables variables variables
Priv Inv PA Priv Inv GOV Priv Inv GOV
(7) (8) (9) (10) (11) (12)
4.43
(–1.6)
2.25 2.26
(1.99)** (2.00)**
–0.08 0.014
(–1.53) (–0.667)
3.27 –0.05 1.98 0.1 1.98 0.13
(4.86)*** (–.40) (8.06)*** (–1.2) (8.07)*** (–1.46)
–0.05 0.21 0.36 0.124 0.35 0.133
(–0.10) (3.49)*** (–1.4) (3.27)* (–1.36) (4.36)***
–0.05 –0.001 –0.035 0 –0.038 –0.001
(–2.90)*** (–0.19) (–2.97)*** (–0.13) (–3.14)*** (–0.49)
–0.01 0 0 0.0003 0 0.0003
(–1.12) (6.56)*** (–0.84) (8.72)*** (–0.81) (8.84)***
3.72 –1.14 –0.34 –0.28 –0.32 –0.29
(–1.11) (4.91)*** (–0.31) (–1.91)** (–0.29) (–2.09)**
–0.05 –0.037 –0.035
(–3.13)*** (–2.87)*** (–3.21)***
0.29 0.22 0.21
(3.04)*** (3.53)*** (3.58)***
0.11 –0.015 –0.016
(2.54)*** (5.75)*** (6.44)***
11.1 –0.73 12.2 –0.82 12.2 –0.66
(16.34)*** (–1.19) (15.55)*** (–2.14) (15.61)*** (–9.39)
349 349 349 349 349 349
Finally, estimation of Equation (1) confirms the natural curse hypothesis.
The coefficient of the oil export variable as a percentage of total merchandise
export is significant and negative. Identically, the regional dummy for MENA
countries exhibits a negative coefficient. MENA countries seem to be diverg-
ing from the rest of the world in terms of private investment, which is the key
determinant of long-term growth. However, this dummy variable is not sig-
nificant at the conventional levels. This result is likely to stem from the oil
export variable in the system, which significantly reduces private investment.
In the QA equation (table 17.1, column 2), our estimations reveal the pos-
itive impact of several factors on the quality of administration. This is the case
with GDP per capita, which means that more developed countries entail bet-

Breaking the Barriers to Higher Economic Growth440
ter governance institutions. Also, private investment helps improve the
administrative quality significantly at a 5 percent level. This last result justifies
the use of the 3SLS estimation technique in order to address—among other
things—the two-way causality. Tenure of system also predicts better adminis-
trative quality at less than 1 percent significance level.
Our estimations fail, however, to validate the negative impact of the share
of oil exports in merchandise exports. This result contradicts the fact that
countries with less reliance on natural resources form better governance insti-
tutions. More important, structural reforms and human capital do not appear
to immediately improve administrative quality. However, when estimating the
system by eliminating private investment from Equation (2), structural
reforms appear to be positive and highly significant, other results being
unchanged31 (see table 17.1, column 4). This result seems to be due to the fact
that the structural reforms index is correlated with private investment. Hence,
the positive impact of private investment on administrative quality appears to
be mainly due to the structural reforms that stimulate firms’ decisions to
invest. This result confirms that, in addition to the direct link highlighted pre-
viously, economic reforms affect private investment through their impact on
institutional quality. This two-channel causality brings new empirical evi-
dence on the link between institutions and private economic activity.
Political Stability. When PS is taken into consideration (columns 5 and 6, Table
17.1), the first interesting result is that this factor—similar to QA—appears to
have a significant and positive impact on firms’ decisions to invest. This conclu-
sion is in line with the findings of various authors who have been able to show—
using various indicators of political stability—that a sound and stable political
environment provides enterprises with more predictable conditions to invest.
Besides, our new set of estimations validates most of the conclusions drawn
previously for QA with a few exceptions. First, structural reforms and human
capital are still validated as important factors for private investment decisions.
This time, however, they play their roles indirectly only, through improving
political stability. This conclusion constitutes quite interesting empirical evi-
dence. Structural reforms, by leading to better economic performances, lower
the discontent of the population and produce a more stable political environ-
ment. Education goes in the same direction.
Another small difference can be seen in the MENA dummy variable, which
is now significant in both equations, though at the 10 percent level: MENA
countries underperform in private investment while appearing to be better in
reaching more stable political systems. This finding is quite understandable
considering that MENA countries display high government stability, which is
one of the main components in the aggregate political stability indicator.
Other conclusions, such as the neoclassical investment model and the natu-
ral curve hypothesis in Equation (1) of private investment, the positive impact

Part III: Trade, Competitiveness, and Investment 441
of the tenure of the system and the greater political stability of richer countries
in equation (2) of institutional quality, hold in the case of the PS index.
Public Accountability. Columns 7 and 8 of table 17.1 report the regression results
when PA is controlled to gauge the quality of institutions. Results are this time
slightly different. In fact, our estimations fail to find strong evidence that PA is
detrimental to private investment. Although some empirical evidence can been
found in the literature (see in particular Pastor and Sung 1995), this result may
be explained by the unresolved debate on the potential role of democratic insti-
tutions on growth (see Glaeser et al. 2004), as well as by some deficiencies in the
specification of our model. For comparison purposes in this paper,we have esti-
mated the same model for each of our indicators of governance. Hence, our
results are likely to stem from the fact that the underlying mechanisms to shape
PA are different from the mechanisms of QA and PS. It is shown in the next sec-
tion that the specification including PA in the aggregated governance indicator
(Gov) indicates the role of this factor in the private investments decisions.
This set of estimations, however, seems to still validate that structural
reforms encourage private investment decisions.This time, the link appears to
be only direct, and the coefficient of the structural reform indicator in the
institutional quality equation [Equation (2)] appears insignificant. Besides, as
for PS education and health of the population seem to encourage private
investment by participating in the democratization process of the country.An
interesting result also concerns the MENA dummy variable in the PA equa-
tion: Its coefficient is now significant and negative. This finding confirms the
deficit in democratic institutions of the MENA region, as already stressed by
several authors (see, in particular, World Bank 2003). Moreover,following our
estimations, richer countries exhibit better democratic institutions, and natu-
ral resources exporters still show low private investment performances. The
initials results for real interest rate, economic growth and tenure of system
remain unaltered as well.
Governance. Our last set of estimations takes into consideration the aggregate
indicator of governance, which summarizes the information contained in the
three previous indicators. Results of the regressions are reported in columns
(9) to (12) of table 17.1. This last set of estimations confirms most of the
results obtained before. The aggregate indicator of governance appears to have
a positive and significant coefficient that validates the importance of this fac-
tor for the firm’s decisions to invest. Structural reforms are highly significant
in enhancing private investment. The effects of human capital work indirect-
ly by affecting aggregate governance. Overall, MENA countries underperform
in terms of governance institutions. This result indicates the need for institu-
tional reform in the MENA region, especially considering the positive and
persistent role of governance institutions in private investment.

Breaking the Barriers to Higher Economic Growth442
Other results remain the same as well, which confirms the robustness of
our previous results. An interesting point, however, can be seen in the fact that
since PA is included in the aggregate indicator of governance,32 this factor
now actively plays a role in the firm’s decisions to invest. Even though this
result has to be considered with caution, it can be seen as further evidence for
the literature on the positive role of democratic institutions in the economic
performance of the countries.
Governance Institutions: How Much Can They Improve Private Investment in
MENA?
In this section, we use the model estimated previously to determine which fac-
tors would improve investment performance in our MENA countries. We
evaluate, in particular,the contribution of administrative quality,political sta-
bility, and public accountability, which have been revealed to be of primary
importance in firms’ decisions to invest. We also consider the role of structur-
al reforms and human capital. For this purpose, we simulate how much pri-
vate investment the region would have achieved if governance institutions,
structural reforms, and human capital had been improved by one standard
deviation. This simulation has been done for two time periods, the 1980s and
the 1990s, respectively. For the calculations, we use the last set of estimations,
which summarize the effects of the three subcomponents of governance (see
table 17.1, columns 9 and 10).
A first step consisted in calculating the coefficients of the initial variables
that explain the composite indicators of governance (GOV), structural
reforms (SR) and human capital (H). The calculation is based on the estimat-
ed coefficients of these aggregate indicators in the regression (table 17.1, col-
umn 9), as well as on the weights of each principal component in the aggre-
gate indicator combined with the loading of the initial variables in each prin-
cipal component (Annex 17.4).33 In the case of human capital calculations, we
consider the indirect impact on private investment through the improvement
of the quality of governance institutions. We use in this case the coefficient of
the human capital indicator in the governance equation (table 17.1, column
10) combined with the estimated coefficient of the governance indicators in
the investment equation (table 17.1, column 9). Coefficients of the initial vari-
ables are presented in Annex 5and contributions to private investment appear
in tables 17.2 and 17.3. In tables 17.2 and 17.3, the contribution of the QA
index has been calculated by aggregating the contributions of its four sub-
components.34 The same thing has been done for PS, 35 PA, 36 SR, 37 and H. 38
These simulations show quite interesting results (see table 17.2). A first set
of conclusions concerns the potentially significant impact of improved gover-
nance institutions in the region. An amelioration of one standard deviation of
the QA would have increased private investment by 1.4 percent of GDP dur-

Part III: Trade, Competitiveness, and Investment 443
ing the 1980s and the 1990s. This augmentation is of 1.1 to 1.2 percent of GDP
in the case of PA and of 0.8 to 0.9 percent of GDP for an amelioration of the
PS. In total, private investment could have been higher by 3.4 to 3.5 percent of
GDP if governance institutions had been reformed in an appropriate way.
Our calculations also point out that governance deficiencies have not been
the only reasons for low private investment performance in MENA.
Reforming the economy in other dimensions is necessary to boost private
investment in the region. This has been the case over the whole period, but
more importantly during the 1990s, when private investment could have been
increased by 3 percent of GDP if structural reforms had been improved by
one standard deviation (2.3 percent in the 1980s).
On average, private investment could have reached 17.7 and 18 percent of
GDP in the 1980s and the 1990s (compared to 11.6 and 12 percent observed)
if governance institutions and structural deficiencies had been improved at the
same time (see table 17.3). If we add the indirect impact of human capital (0.4
to 0.5 percent of GDP, through the amelioration of governance institutions),
private investment in Egypt could have been stimulated by 48 and 77 percent
during the 1980s and the 1990s, and could have reached 18.7 and 15.7 percent
of GDP (compared to 12.6 and 8.9 observed), respectively. This percentage
increase would have been of 64 and 54 percent in Iran (with a private invest-
ment ratio of 15.7 and 19.5 percent of GDP in this case), 51 and 68 percent in
Morocco (with a ratio of private investment of 18.1 and 16.9 percent of GDP),
and 45 percent in Tunisia (19.7 and 21.7 percent of GDP for the private invest-
ment ratio). These figures are in line with the performances of the East Asian
economies that achieved, on average for our sample of countries, an invest-
ment ratio of 17.5 and 19.9 percent of GDP during the two subperiods.39
Our simulations also give a more precise diagnostic of which specific gov-
ernance institutions would improve private investment performance in
MENA. Contributions are calculated for each indicator of governance, as well
as for structural reforms and human capital. Interesting conclusions relate to
the impact of the different governance institutions. PA appears to be of pri-
mary importance in enhancing the confidence of private investors in MENA.
An improvement of one standard deviation of civil liberties and political
Table 17.2. Private Investment to GDP
(Increase with an improvement in)
Structural Human Total
reformsaGOVaQA PA PS capitalbcontributionsc
1980 2.3 3.5 1.4 1.1 0.9 0.4 6.1
1990 3.0 3.4 1.4 1.2 0.8 0.5 6.8
Source: Authors’calculations.
Note: a. Direct impact on private investment calculated from equation (9), table 17.1.
b. Indirect impact on private investment calculated from equations (9) and (10), table 17.1.
c. Sum of direct and indirect impact.

Breaking the Barriers to Higher Economic Growth444
rights would have respectively increased private investment decisions by
approximately 0.5 and 0.6 percent of GDP during the 1980s and the 1990s
(see table 17.4). This result clearly shows that democratic institutions matter
for the region. This finding can be linked to the significant deficit of demo-
cratic institutions in MENA countries (see World Bank 2003). This aspect
gives to the region a significant scope for improving private investment per-
formances in the future.
On QA, bureaucratic quality and corruption constitute other key factors for
private investment decisions in the region. An improvement in the quality of
the bureaucracy and a reduction of the level of corruption, both by one stan-
dard deviation, would have stimulated firms’ investment by 0.4 to 0.5 percent
of GDP for each factor. These findings confirm some conclusions of the liter-
ature, specifically on the role played by corruption in increasing the cost and
risk of doing business. Besides, our calculations add to the subject by quantify-
ing the importance of these two factors for the countries of our interest.
On the side of PS, attention should be given in the region to the reduction
of internal and external conflicts, as well as of ethnic tensions. Narrowing the
gap with politically more stable developing countries by one standard devia-
tion would have helped private investment decisions, which could have been
higher by 0.2 to 0.3 percent of GDP, depending on the factor. Our findings
corroborate that institutions associated with political instability have a dis-
ruptive effect on aggregate investment. This result makes political stability a
significant factor in reducing uncertainty and creating a friendly business
environment in MENA.
Another striking feature of this set of calculations relates to the critical role
of financial development and trade policy within the MENA region. A more
developed financial system in the region would have helped private firms real-
ize their investment projects—which could have been higher by 1.2 to 1.7 per-
cent of GDP. A more open trade policy would also have stimulated private
Table 17.3. Private Investment to GDP
1980s 1990s
Observed Predicted Observed Predicted
Egypt (% GDP) 12.6 18.7 8.9 15.7
% increase (48) (77)
Iran (% GDP) 9.6 15.7 12.7 19.5
% increase (64) (54)
Morocco (% GDP) 12.1 18.1 10.1 16.9
% increase (51) (68)
Tunisia (% GDP) 13.6 19.7 14.9 21.7
% increase (45) (43)
Average (% GDP) 12.0 18.0 11.6 18.5
Source: Authors’calculations.

Part III: Trade, Competitiveness, and Investment 445
investment by 1 to 1.3 percent of GDP. These results have to be related to the
deficit of the region in these two fields of activity (see Nabli and Véganzonès-
Varoudakis 2007). These findings reveal that structural reforms represent
another important question that MENA governments have to address if the
region wants to catch up with more successful developing economies.
Conclusion
This paper empirically shows, for a panel of 31 developing countries studied
during the 1980s and the 1990s, that governance institutions constitute an
important part of the investment climate of developing economies. This
result strongly holds for the QA and confirms that a low level of corruption,
good-quality bureaucracy, a reliable judiciary, strong security of property
rights, a reasonable risk to operations, and sound taxation and regulation
contribute significantly to firms’ decisions to invest. Our estimations also
verify that PS, by providing a sound and predictable environment to enter-
prises, contributes as well to a friendly business environment. These results
add significantly to the literature on governance by validating the role of a
large set of institutional variables on private economic performances over a
relatively long period of time.
Our findings are unambiguous and robust to the introduction of other
explanatory variables. This is the case for structural reforms—in the form of
trade openness and financial development—and for human capital, which
appear to play significant roles in private investment decisions. This result
shows that firms in developing countries face constraints that are not
accounted for in more developed economies. It also shows that—contrary to
recent works, which make governance institutions the dominant factor, with
little independent influence of economic policies (see Rodrick, Subramanian,
and Trebbi 2002 and Easterly and Levine, 2003)—economic policies and gov-
ernance institutions both contribute to firms’ decisions to invest. We show as
well that structural reforms and human capital contribute to firms’ decisions
to invest by also improving the quality of governance. These conclusions have
been reached by estimating a simultaneous model of private investment and
governance quality, whereby economic policies and human capital concur-
Table 17.4. Private Investment to GDP
(Increase with an improvement in)
trade priv corrup bur inves law pol civ gov int ext ethn
SR pol cred QA tion qual prof ord PA rights lib PS stab confl confl tens
1980 2.2 1.0 1.2 1.4 0.45 0.53 0.19 0.28 1.1 0.53 0.57 0.9 0.10 0.27 0.30 0.27
1990 3.0 1.3 1.7 1.4 0.42 0.44 0.25 0.26 1.2 0.62 0.56 0.8 0.12 0.25 0.20 0.22
Source: Authors’calculations

Breaking the Barriers to Higher Economic Growth446
rently explain both variables. These conclusions can be considered an impor-
tant contribution to the empirical literature on governance.
Our estimations find, in addition, some evidence that PA is detrimental to
private investment. Although our results have to be considered with caution,
they can be regarded as contributing some empirical evidence to the unre-
solved debate on the potential role of democratic institutions on growth (see
Glaeser et al. 2004 and Pastor and Sung 1995 for more conclusive studies).
In MENA, improved governance institutions would greatly stimulate pri-
vate investment. This is the case for all components of governance, with spe-
cial attention to civil liberties and political rights, corruption and bureaucrat-
ic quality, and conflicts and ethnic tensions. By reforming substantially their
governance institutions during the 1980s and the 1990s, (that is, by increasing
by one standard deviation all components of governance), MENA countries
could have boosted private investment by 3.4 to 3.5 percent of GDP. This
result makes governance a key variable for improving the investment climate
in the region.
Governance deficiencies, however, are not the only issues that MENA could
address to encourage private investment in the region. Reforming the econo-
my constitutes another powerful instrument that would also stimulate firms’
investment decisions. A more developed financial system would have permit-
ted the private sector to implement more investment projects. One standard
deviation increase during the 1980s and the 1990s would have raised the pri-
vate investment ratio by 1.2 to 1.7 percent of GDP. Similarly, a more open
trade policy would have stimulated private investment decisions by 1 to 1.3
percent of GDP during the same period. This makes structural reforms an
important issue that MENA governments also have to address if the region
wants to catch up with more successful developing economies.
Notes
1. See in particular Rodrik (1999) and Frankel (2002).
2. See, for example, Knack and Keefer (1995); Acemoglu, Johnson, and Robinson (2001); Rodrik,
Subramanian, and Trebbi (2002).
3. See Hall and Jones (1999; Acemoglu, Johnson, and Robinson (2001); Easterly and Levine (2003);
and Rodrik, Subramanian and Trebbi (2002).
4. See, for example, Acemoglu, Johnson, Robinson, and Taicharoen (2003).
5. See North (1990); Knack and Keefer (1995); Calderon and Chong (2000); Easterly and Levine
(2003); Rodrik, Subramanian, and Trebbi (2002); and Saleh (2004).
6. See, in particular, Keefer (2002).
7. This shortcoming is a more general concern in the literature on democracy and development (see
De Haan and Siermann[ 1996]; Prszeworski and others[ 2000]). The work of Pastor and Sung (1995)
is, however, one of the few to show a positive effect of various indicators of democratic institutions on
private investment in the developing world.
8. See, in particular, Rodrik (1991); Alesina and Perotti (1996); Le (2004); and Brunetti and Weder
(1994).

Part III: Trade, Competitiveness, and Investment 447
9. World Bank (2003) has investigated the correlation between private investment and the ICRG
(1999) index of “investment profile.” This index is based on measures of contract enforceability,expro-
priation, profit repatriation, risk of operation, taxation, and payment delays.
10 See, in particular, Rodrik, Subramanina, and Trebbi (2002) and Easterly and Levine (2003).
11. For the positive spillover from trade openness on institutions, see Berg and Krueger (2003); Islam
and Montenegro (2002); and Wei (2000). For the role of domestic competition, see Ades and Di Tella
(1999); Djankov et al. (2001); and the World Bank (2002).
12 This impact might also be explained by the fact that the measure of institutional quality is most
often subjective and an amalgam of policy and institutional factors.
13. In trying to gauge the exogenous contribution of institutions, recent research has given particular
attention to the possible role played by geographical and historical influences on institutional forma-
tion (see Acemoglu, Johnson, and Robison[2001] and Engelman and Sokoloff 2002).
14. Various methods of aggregation have been used to categorize different types of institutions. See, in
particular, Acemoglu, Johnson, Robinson, and Taicharoen (2003); Kaufmann et al. (2003); and the
World Bank (2003).
15. See Calderon and Chong (2000) and Acemoglu, Johnson, and Robinson (2001) in the context of
growth; see North (1981); Knack and Keefer (1995); Calderon and Chong (2000); Easterly and Levine
(2003); Rodrik, Subramanian, and Trebbi (2002); and Saleh (2004) in the context of investment.
16. See, in particular, Rodrik (1991) and Serven and Solimano (1993).
17. See De Haan and Siermann (1996); Prszeworski and Limongi (1993); and Prszeworski et al. (2000).
18. See, in particular, Rodrik (1991); Alesina and Perotti (1996); Le (2004); and Brunetti and Weder
(1994). In the growth context, see also Alesina et al. (1996); Svensson (1998); and Olson et al. (2000).
19. See, for example, Greene and Villanueva (1991); Blejer and Khan (1984); and Serven (1997).
20. See Shafik (1992) on Egypt; Schmidt and Muller (1992) on Morocco, as well as Bisat, El-Erian, El-
Gamal, and Mongelli (1996); and Aysan, Sang, and Véganzonès-Varoudakis (forthcoming) on MENA.
21. See, in particular, Shafik (1992) and Agenor and Montiel (1999) for a discussion and additional ref-
erences.
22. See, for example, McKinnon (1973) and Shaw (1973).
23. For the positive spillover from trade openness on governance quality,see Berg and Krueger (2003);
Islam and Montenegro (2002);and Wei (2000). For the role of domestic competition, see Ades and Di
Tella (1999); Djankov et al. (2001); and World Bank (2002).
24. See Lucas (1988), Psacharopoulos (1988) and Mankiw, Romer, and Weil (1992).
25. See, in particular, Acemoglu and Shimer (1999).
26. See Bresnahan, Brynjolfsson and Hitt (2002).
27. See Nabli and Véganzonès-Varoudakis (2007).
28 See Aizenmann and Marion (1999) and Everhart and Sumlinski (2001).
29. Aysan (2006) points out that this variable captures the “rentier effect.” He notes that “it is easier for
the elite to control and capture the rents from ‘point source’ resources. Resource rents are generally
high in oil production. Around 80 percent of oil income is considered to be resource rent (Gylfason
2001), while such rents are much lower for other types of products in industry or in agriculture. A
small work force is required to extract oil resources.Most of the time, oil is extracted by foreign firms
with sophisticated technical skills (Isham et al. 2002).As a result, the ruling elite can exclude the major-
ity of the population in extracting oil reserves. In other words, there exists no incentive on the part of
the elite to incorporate the society into increasing aggregate production. Given the lack of economic
preconditions, the citizens cannot generate pressure for increased literacy and political influence. This
lack of political influence further feeds the vicious cycle by not effectively and peacefully revealing
public interest and preferences.”
30. Sources of data are as follows: The private investment series have been processed from various
national and international sources (International Finance Corporation [IFC], World Development

Breaking the Barriers to Higher Economic Growth448
Indicators [WDI], and Life Data Base [LDB]—see section 4.1 for more details); the “Administrative
Quality” and “Political Stability” indexes use ICRG (1999) data; the components of the “Public
Accountability”indicator come from Freedom House (2002); the “Structural Reforms” index uses data
from WDI, but the oil export series entering the trade policy indicator comes from the United Nations;
in the “Human Capital’ indicator, the numbers of years of schooling are from Barro and Lee (1994)
and from Barro (2000a and b), and the life expectancy series is from WDI. All aggregated indicators
have been generated after implementing the PCA methodology (see Annex 17.4 for more details).
Interest rates have been calculated from IFS, and tenure of the system comes from Keefer and others
(2001). All other data are from WDI.
31. Oil export also becomes significant in the “Administrative Quality” equation at a 10 percent level.
32. PA contributes significantly and with the right sign to GOV (see the results of the PCA in Annex
17.4).
33. For more details on the methodology, see Nagaraj, Varoudakis, and Véganzonès (2000).
34. Theses subcomponents are corruption, bureaucracy quality, investment-friendly profile of admin-
istration and law and order (see section 2).
35. Political Stability has been proxied by aggregating the following indicators: government stability,
internal and external conflicts, and ethnic tensions.
36. Public Accountability has been calculated by using civil liberties and political rights.
37. The Structural Reform indicator contains trade policy and financial development.
38. Human Capital is defined from life expectancy, and years of primary, secondary, and tertiary edu-
cation.
39. For comparison purposes, we have benchmarked the effort represented by the improvement of one
standard deviation of each initial indicator entering the governance (GOV), structural reforms (SR)
and human capital (H) indicators to the reforms achieved by the East Asian economies. These calcu-
lations are shown in Table A.6., Annex 17.6. Figures of the table are the level of reforms potentially
reached by our MENA countries divided by the ones achieved by the East Asian economies. For exam-
ple, the value 1 means that an improvement of one standard deviation has increased the reform indi-
cator in MENA to the level of the East Asian economies. As well, the value 1.2 means that the level
reached by MENA is 20 percent higher than the one observed in East Asia. Most of the time, the effort
considered in the simulations gives to our MENA countries an advance in “Administrative Quality,”
“Political Stability” and “Public Accountability” (during the 1990s, in this case). MENA, however,
would not catch up with East Asia in terms of structural reforms (except in the case of Tunisia).
References
Acemoglu, Daron, and Robert Shimer. 1999. “Efficient Unemployment Insurance.” Journal of Political
Economy 107 (5): 893–928.
Acemoglu, Daron, and James A. Robinson. 2001. “A Theory of Political Transitions.” American
Economic Review 91 (4): 938–63.
Acemoglu, Daron, Simon Johnson, and James A. Robinson. 2001. “The Colonial Origins of Com-
parative Development: An Empirical Investigation.” American Economic Review 91 (5): 1369–401.
Acemoglu, Daron, Simon Johnson, James Robinson, and Yunyong Taicharoen. 2003. “Institutional
Causes, Macroeconomic Symptoms,Volatility,Crises and Growth.” Journal of Monetary Economics
89 (September): 49–123.
Ades, Alberto, and Rafael Di Tella. 1999. “Rents, Competition and Corruption.” American Economic
Review 89, no. 4: 982–93.
Aizenman, Joshua,and Nancy P. Marion. 1999. “Volatility and Investment: Interpreting Evidence from
Developing Countries.” Economica 66(262): 157–79.
Agenor, Pierre-Richard, and Peter J. Montiel. 1999. Development Macroeconomics. Princeton, NJ:
Princeton University Press.

Part III: Trade, Competitiveness, and Investment 449
Alesina, Alberto, and Roberto Perotti. 1996. “Income Distribution, Political Instability, and Invest-
ment.” European Economic Review 40(6): 1203–28.
Alesina, Alberto, Sule Ozler, Nouriel Roubini, and Phillip Swagel. 1996. “Political Instability and
Economic Growth.” Journal of Economic Growth 1 (2): 189–211.
Aysan,Ahmet Frank. 2006.“The Role of Efficiency of Redistributive Institutions on Redistribution: An
Empirical Assessment.” Bog˘aziçi University Research Papers, ISS/EC 2006-14. Istanbul: Bog˘aziçi
University.
Aysan, Ahmet F., Gaobo Pang, and Marie-Ange Véganzonès-Varoudakis. Forthcoming. “Uncertainty,
Economic Reforms and Private Investment in the Middle East and North Africa.” Applied Economics.
Azariadis, Costas, and Amartya Lahiri. 2002. “Do Rich Countries Choose Better Governments?”
Contributions to Macroeconomics 2 (1): Article 4.
Balassa, Bela. 1978. “Exports and Economic Growth-Further Evidence.” Journal of Development
Economics 5(2): 181–89.
Barro, Robert J. 2000a. “International Data on Educational Attainment: Updates and Implications.”
CID Working Paper 42. Harvard University, Cambridge, MA.
———. 2000b.“International Measures of Schooling Years and Schooling Quality.” American Economic
Review 86(2): 218–23.
Barro, Robert J., and Jong-Wha Lee. 1994. “Panel Data Set Cross Countries.” National Bureau of
Economic Research, http://www.nber.org/pub/barro.lee/
Bellin, Eva. 2001. “The Politics of Profit in Tunisia: Utility of the Rentier Paradigm?” World
Development 22 (3): 427–36.
Berg, Andrew, and Ann Krueger. 2003. “Trade, Growth, and Poverty—A Selective Survey.” IMF
Working Paper 03/30. International Monetary Fund, Washington, DC.
Bisat, Amer, Mohamed A. El-Erian, Mahmoud El-Gamal, and Francesco P. Mongelli. 1996.
“Investment and Growth in the Middle East and North Africa.” IMF Working Paper WP/96/124.
International Monetary Fund, Washington, DC.
Blejer, Mario, and Mohsin Kahn. 1984. “Government Policy and Private Investment in Developing
Counties.” IMF Staff Papers 31 (2): 79–403. International Monetary Fund, Washington, DC.
Bresnahan, Timothy, F. Erik Brynjolfsson, and Lorin M. Hitt. 2002. “Information, Technology,
Workplace Organization and the Demand of Skilled Labor: Firm-Level Evidence.” Quarterly
Journal of Economics 117 (1): 339–76.
Brunetti, Aymo, and Beatrice Weder. 1994. “Political Credibility and Economic Growth in Less
Developed Countries.” Constitutional Political Economy 5 (1):23–43.
Caballero, Ricardo, and Arvind Krishnamurthy. 2001.“A Dual Liquidity Model for Emerging Markets.”
NBER Working Paper 8758. National Bureau of Economic Research, Cambridge, MA.
Calderon, Cesar, and Alberto Chong. 2000. “Causality and Feedback between Institutional Measures
and Economic Growth.” Economic and Politics 12 (1): 69–81.
De Haan, Jakob, and Clemens L. J. Siermann. 1996. “New Evidence from Historical Accounts and
Contemporary Data.”Land Economics 75 (3): 341–59.
Djankov, Simon, Rafael La Porta, Florencio Lopez-de-Silanes, and Andrei Sheifer. 2001. “The
Regulation of Entry.” CEPR Discussion Paper No. 2953. (September)
Easterly, William, and Ross Levine. 2003. “Tropics, Germs, and Crops: How Endowments Affect
Economic Development.” Journal of Monetary Economics 50 (1): 3–39.
El Badawi, Ibrahim A. 2002.“Reviving Growth in the Arab World.” Working Paper Series 0206. Arab
Planning Institute, Safat, Kuwait.
Engelman, S., and K. Sokoloff. 2002. “Factor Endowments, Inequality, and Paths of Development
among New World Economies.” Economia 3 (1): 41–109.
Evans, Peter, and James Rauch. 2000) “Bureaucratic Structure and Bureaucratic Performance in the
Less Developed Countries.” Journal of Public Economics 75 (1): 49–71.

Breaking the Barriers to Higher Economic Growth450
Everhart, Stephen, and Mariuzs Sumlinski. 2001. “Trends in Private Investment in Developing
Countries, Statistics for 1970–2000, and the Impact on Private Investment of Corruption and the
Quality of Public Investment.” IFC Discussion Paper 44, Washington, DC.
Feder, G. 1982. “On Exports and Economic Growth.” Journal of Development Economics 12 (1–2):
59–73.
Frankel, Jeffrey. 2002. “Promoting Better National Institutions: The Role of the IMF.” Paper presented
at the Third Annual IMF Research Conference, Washington, DC. November 7–8.
Frankel, J.A.,and D. Romer, 1999.“Does Trade Cause Growth?”The American Economic Review 89 (3):
379–99. Fraser Institute, Economic Freedom in the World, Vancouver, Canada.
Freedom House. 2002. Freedom in the World 2001. 2002. http://www.freedomhouse.org. Accessed
2003.
Galor, Oded,Omer Moav, and Dietrich Vollrath. 2005. “Land Inequality and the Emergence of Human
Capital Promoting Institutions.” Development and Comp Systems 0502018. Economic Working
Paper Archive at WUSTL.
Gylfason, T. 2001. “Natural Resources, Education, and Economic Development.” European Economic
Review 45 (No. 6): 847–59.
Glaeser, Edward L., Rafael La Porta, Florencio Lopez-de-Silanes, and Andrei Shleifer. 2004. “Do
Institutions Cause Growth?” NBER Working Paper 10568. National Bureau of Economic Research,
Cambridge, MA.
Greene, Joshua,and Delano Villanueva. 1991.“Private Investment in Developing Countries.” IMF Staff
Papers 38 (March):33–58. International Monetary Fund, Washington, DC.
Gupta, Sandeev, Hamid Davooli, and Rosa Alonso-Terme. 2002. “Does Corruption Affect Income
Inequality and Poverty?” Economics of Governance 3 (1): 23–45.
Hall, Robert, and Charles Jones. 1999.“Why Do Some Countries Produces So Much More Output per
Workers than Others?” Quarterly Journal of Economics 114 (1): 83–116.
Heritage Foundation. 2003. Index of Economic Freedom. Available online at http://www.heritage
.org/research/ features/index (The Wall Street Journal).
ICRG. 1999. Brief Guide to the Rating System. Political Risk Service Group, New York.
International Finance Statistics Database (IFS). The International Monetary Fund, Washington, DC.
International Monetary Fund. 2003. World Economic Outlook. Growth and Institutions.World
Economic and Financial Surveys, Washington, DC.
Isham, J., M. Woolcock, L. Pritchett, and G. Busby. 2005. “The Varieties of Rentier Experience: How
Natural Resource Export Structures Affect the Political Economy of Economic Growth.”
Islam, Roumeen, and Claudio Montenegro. 2002. “What Determines the Quality of Institutions?”
Policy Research Working Paper 2764 (January). World Bank, Washington, DC.
Jorgenson, Dale W. 1963. “Capital Theory and Investment Behavior.” American Economic Review 53
(2): 247–59. (May)
Kaufmann, Dani, Aart Kraay, and Maximo Mastruzzi. 2003. “Governance Matters III, Governance
Indicators for 1996–2002.” Policy Research Working Paper 3106. World Bank, Washington, DC.
Keefer, Philip. 2002. “The Political Economy of Corruption in Indonesia.” World Bank, Washington,
DC.
Keefer, Philip, Thorsten Beck, George Clarke, Alberto Groff, and Patrick Walsh. 2001. “New Tools in
Comparative Political Economy: The Database of Political Institutions.” World Bank Economic
Review 15 (1): 165–76.
Knack, Stephen, and Philip Keefer. 1995. “Institutions and Economic Performance: Cross-Country
Tests Using Alternative Institutional Measures.” Economics and Politics 7 (3): 207–27.
Le, Quan V. 2004. “Political and Economic Determinants of Private Investment.” Journal of
International Development 16(4): 289–604.

Part III: Trade, Competitiveness, and Investment 451
Levine, Ross. 1997. “Financial Development and Economic Growth: Views and Agenda.” Journal of
Economic Literature 35 (2): 688–726.
Life Data Base (LDB). World Bank, Washington, DC.
Lipset, Seymour Martin. 1959. “Some Social Requisites of Democracy: Economic Development and
Political Legitimacy.” American Political Science Review 53 (1): 69–105.
Lucas, Robert. 1988.“On the Mechanics of Development.”Journal of Monetary Economics 22 (1):1–175.
Mankiw, Gregory, David Romer, and David Weil. 1992. “A Contribution to the Empirics of Economic
Growth.” Quarterly Journal of Economics 106 (3): 407–37.
Mauro, Paolo. 1995. “Corruption and Growth.”Quarterly Journal of Economics 110 (3): 681–712.
McKinnon, Ronald I.1973. Money and Capital in Economic Development. Washington, DC: Brookings
Institution.
Mo, Park Hung. 2001.“Corruption and Economic Growth.” Journal of Comparative Economics 29 (1):
66–79.
Nabli, Mustapha, and Marie-Ange Véganzonès-Varoudakis. 2007. “Reform Complementarities and
Economic Growth in the Middle East and North Africa.” Journal of International Development
19(1): 17–54.
Nagaraj, Ratapolu, Aristomène Varoudakis, and Marie-Ange Véganzonès. 2000. “Long-Run Growth
Trends and Convergence across Indian States: The Role of Infrastructures.” Journal of International
Development 12 (1): 45–70.
North, Douglas C. 1981. Structure and Change in Economic History.New York:W.W.Norton.
———. 1990. Institutions, Institutional Change, and Economic Performance. Cambridge: Cambridge
University Press.
Olson, Mancur, Naveen Sarna, and Anand V. Swamy. 2000. “Governance and Growth: A Simple
Hypothesis Explaining Cross-Country Differences in Productivity Growth.” Public Choice 102
(3–4): 341–64.
Pastor, Manuel, and Jae Ho Sung. 1995.“Private Investment and Democracy in the Developing World.”
Journal of Economic Issues 29 (1): 223–321.
Prszeworski, Adam, and Fernando Limongi. 1993.“Political Regimes and Economic Growth.” Journal
of Economic Perspectives 7 (3): 51–69.
Prszeworski, Adam,Michael Alvarez, Jose Antonio Cheibub, and Fernando Limongi. 2000. Democracy
and Development: Political Institutions and Well-Being in the World, 1950–1990. New York:
Cambridge University Press.
Psacharopoulos, George. 1988. “Education and Development: A Review.” World Bank Research
Observer 3 (1): 99–116..
Rama, Martin. 1993. “Empirical Investment Equations in Developing Countries.” In Striving for
Growth after Adjustment, ed. L. Servén and A. Solimano. World Bank: Washington, DC.
Rodriguez, Francisco,and Jeffrey D.Sachs. 1999.“Why Do Resource Abundant Economies Grow More
Slowly?” Journal of Economic Growth 4 (3): 277–303.
Rodrik, Dani. 1991. “Policy Uncertainty and Private Investment in Developing Countries.” Journal of
Development Economics 36 (2): 229–42.
———. 1999. “Institutions for High-Quality Growth: What They Are and How to Acquire Them.”
Paper presented at the IMF Conference on Second Generation Reforms, Washington, DC.
November 8–9.
Rodrik, Dani, Arvind Subramanian, and Francesco Trebbi. 2002. “Institutions Rule: The Primacy of
Institutions over Geography and Integration in Economic Development.” NBER Working Paper
9305. National Bureau of Economic Research, Cambridge, MA.
Ross, Michael. 2001.“Does Oil Hinder Democracy?” World Politics 53 (3): 325–61.

Breaking the Barriers to Higher Economic Growth452
Sachs, Jeffrey. 2003. “Institutions Don’t Rule: A Refutation of Institutional Fundamentalism.” NBER
Working Paper 9490. National Bureau of Economic Research, Cambridge, MA.
Sachs, Jeffrey, and Andrew Warner. 1997. “Natural Resource Abundance and Economic Growth.”
Center for International Development. Harvard University, Cambridge, MA. Available online at
http://www.cid.harvard.edu//ciddata/ciddata.html. Accessed 2003.
Saleh, Jahangir. 2004. “Property Rights Institutions and Investment.” Policy Research Working Paper
Series 3311. World Bank, Washington, DC.
Schmidt-Hebbel, Klaus, and Tobias Muller. 1992. “Private Investment under Macroeconomic
Adjustment in Morocco.” In Reviving Private Investment in Developing Countries, ed. Alay
Chhibber, Mansoor Dailami, and Nemat Shafik. Amsterdam: North Holland.
Serven, Luis. 1997. “Uncertainty, Instability, and Irreversible Investment: Theory, Evidence, and
Lessons for Africa.” Policy Research Working Paper Series 1722. World Bank, Washington DC.
Servén, Luis, and Andrés Solimano. 1993. “Private Investment and Macroeconomic Adjustment: A
Survey.” In Striving for Growth after Adjustment, ed. Luis Servén and Andrés Solimano.
Washington, DC: World Bank.
Shafik, Nemat. 1992. “Modeling Private Investment in Egypt.” Journal of Development Economics 39
(2): 263–77.
Shaw, E. 1973. Financial Deepening in Economic Development. New York: Oxford University Press.
Svensson, Jakob. 1998. “Investment, Property Rights and Political Instability: Theory and Evidence.”
European Economic Review (Elsevier) 42 (7): 1317–41.
Tanzi, Vito, and Hamid Davooli. 1997. “Corruption, Public Investment, and Growth.” IMF Working
Papers 97/139. International Monetary Fund, Washington, DC.
Transparency International. 2002. Global Corruption Report 2003. Berlin, Germany.
Wei, Shang-Jin. 2000. “Natural Openness and Good Government.” NBER Working Paper 7765.
National Bureau of Economic Research, Cambridge, MA.
World Bank. 2002. World Development Report 2002: Building Institutions for Market.New York:Oxford
University Press.
———. 2003. Better Governance for Development in the Middle East and North Africa: Enhancing
Inclusiveness and Accountability. MENA Development Report. World Bank, Washington, DC.
World Development Indicators Database (WDI). 2005. World Bank, Washington, DC.

Part III: Trade, Competitiveness, and Investment 453
Annex 17.1. List of Countries
List of Countries with High Quality Data (60 countries)
Argentina Kenya*
Bangladesh* Lithuania
Barbados* Malawi*
Belize Malaysia*
Benin* Mauritius*
Bolivia* Mexico
Brazil* Moldova
Bulgaria Morocco*
Cambodia Namibia
Chile* Pakistan*
China* Panama
Colombia* Papua New Guinea*
Comoros Paraguay*
Costa Rica* Peru*
Cote d’Ivoire Philippines*
Croatia Poland*
Dominican Rep. Romania
Ecuador* Serbia and Montenegro
Egypt* Seychelles
El Salvador South Africa*
Estonia St Lucia
Ethiopia St. Vincent and the Grenadines
Guatemala* Thailand*
Guinea-Bissau Trinidad & Tobago*
Guyana Tunisia*
Haiti Turkey*
Honduras* Uruguay*
India* Uzbekistan
Indonesia* Venezuela*
Iran Yugoslavia (FR)
Because of the lack of corresponding data for some countries, only counties marked with * are included in the final
regressions.

Breaking the Barriers to Higher Economic Growth454
Annex 17.2. Classification of Governance Institution by Kauffmann, Kraay,
and Mastruzzi (2003)
Kaufmann, Kraay and Mastruzzi (2003) categorize governance institutions in
six broad groups. Their measures of governance are based on several hundred
variables measuring perceptions of governance, drawn from 25 separate data
sources constructed by 18 different organizations. Their aggregate governance
indicators aim to measure six different aspects of governance:
1. Government Effectiveness
2. Regulatory Quality
3. Rule of Law
4. Control of Corruption
5. Political Stability and Absence of Violence
6. Voice and Accountability
“Government Effectiveness” and “Regulatory Quality” clusters summarize
various indicators of the ability of the government to formulate and imple-
ment sound policies. “Government Effectiveness” is composed of a single
grouping of responses on the quality of public service provision, the quality
of bureaucracy, the competence of civil servants, independence of civil service
from political process, and credibility of government’s commitment to poli-
cies. These are considered requisite conditions enabling governments to pro-
duce and implement sound policies and deliver more efficient redistribution.
“Regulatory Quality” measures governance by extracting information from
policy outcomes. It includes measures of market-unfriendly policies such as
excessive regulation of economy, price controls, or inadequate bank
supervision.
The respect of citizens and the state for the institutions which govern their
interactions is categorized as “Rule of Law” and “Control of Corruption.” The
“Rule of Law” cluster includes several indicators which measure the extent to
which agents have confidence in, and abide by, the rules of society, such as
perception of incidence of crime, effectiveness and predictability of the judi-
ciary, and enforceability of contracts. They also provide a cluster for “Control
of Corruption,” measuring perceptions of corruption, conventionally defined
as the exercise of public power for private gain. The presence of corruption is
often a manifestation of lack of respect for the rules which govern the inter-
action between state and citizens, and hence represents inefficiency in govern-
ing redistribution.
The quality of institutions is very much related to how authority figures are
selected and replaced. The “Political Stability and Absence of Violence” cluster
combines several indicators which measure the perceived likelihood that the
government in power will be destabilized or overthrown by unconstitutional
and violent means such as domestic violence and terrorism. This index indi-

Part III: Trade, Competitiveness, and Investment 455
cates the continuity of policies and its effects on the efficiency of
redistribution.
The “Voice and Accountability” cluster is the closest to measures of democ-
racy indicators. Kaufmann, Kraay, and Mastruzzi include this cluster as an
indicator of governance, in an attempt to capture the process by which citi-
zens of a country are able to participate in the selection of governments. This
cluster includes a number of indicators measuring various aspects of the
political process, civil liberties and political rights, as well as various indicators
measuring the independence of the media, in an attempt to quantify the
accountability of those in a position of authority.

Breaking the Barriers to Higher Economic Growth456
Annex 17.3. Our Governance Indicators
Quality of Administration
The QA index is composed of four indicators from ICRG, defined in the fol-
lowing manner:
(i) Control over Corruption “is a measure of corruption within the political
system. Such corruption is a threat to foreign investment for several rea-
sons: it distorts the economic and financial environment; it reduces the
efficiency of government and business by enabling people to assume posi-
tions of power through patronage rather than ability, and, last but not
least, introduces an inherent instability into the political process.”
(ii) Quality of Bureaucracy indicates “countries where the bureaucracy has
the strength and expertise to govern without drastic changes in policy or
interruptions in government services. In these low-risk countries, the
bureaucracy tends to be somewhat autonomous from political pressure
and to have an established mechanism for recruitment and training.
Countries that lack the cushioning effect of a strong bureaucracy receive
low points because a change in government tends to be traumatic in
terms of policy formulation and day-to-day administrative functions.”
(iii) Investment Profile “is a measure of the government’s attitude to inward
investment as determined by an assessment of four sub-components: the
risk to operations, taxation, repatriation and labor costs.”
(iv) Law and Order “are assessed separately. The Law sub-component is an
assessment of the strength and impartiality of the legal system, while the
Order sub-component is an assessment of popular observance of the law.”
Public Accountability
The second set of candidates measures PA.”This index includes two indicators
from Freedom House (FH): “Civil Liberties” and “Political Rights.”
The “Civil Liberties” index mainly addresses the following questions:
• Are there free and independent media, literature, and other forms of cul-
tural expressions?
• Is there open public discussion and free private discussion?
• Is there freedom of assembly and demonstration?
• Is there freedom of political or quasi-political organization?
• Are citizens equal under the law, do they have access to an independent and
nondiscriminatory judiciary, and are they respected by the security forces?
• Is there protection from unjustified imprisonment, exile or torture,
whether by groups that support or oppose the regime?
• Is there freedom from war or insurgency situations?
• Are there free trade unions and peasant organizations or the equivalent,

Part III: Trade, Competitiveness, and Investment 457
and is there effective collective bargaining?
• Are there free professional and other private organizations?
• Are there free businesses or cooperatives?
• Are there free religious institutions, and free private and public religious
expression?
• Is there personal and social freedom, which include aspects such as gender
equality, property rights, freedom of movements, choice of residence, and
choice of marriage and size of family?
• Is there equality of opportunity—which includes freedom from exploita-
tion by or dependency on landlords, employers, union leaders, bureaucrats,
or any other type of denigrating obstacle—to a share of legitimate eco-
nomic gains?
• Is there freedom from extreme government indifference and corruption?
The “Political Rights” index addresses the following questions:
• Is the head of state, head of government, or other chief authority elected
through free and fair elections?
• Are the legislative representatives elected through free and fair elections?
• Are there fair electoral laws?
• Are the voters able to endow their freely elected representatives with real
power?
• Do the people have the right to freely organize into different political par-
ties or other competitive political groupings of their choice, and is the sys-
tem open to the rise and fall of those competing parties or groupings?
• Is there a significant opposition vote, a de facto opposition power, and a
realistic possibility for the opposition to increase its support or gain power
through elections?
• Are the people free from domination by the military, foreign powers, total-
itarian parties, religious hierarchies, economic oligarchies, or any other
powerful groups?
• Do cultural ethnic, religious, and other minority groups have reasonable
self-determination, self-government, autonomy, or participation through
informal consensus in the decision-making process?
• For traditional monarchies which have no parties or electoral process, does
the system provide for consultation with the people to encourage discus-
sion of policy, and to allow the right to petitions the rulers?
Political Stability
The PS index includes the following variables from ICRG:
•Government Stability “is a measure both of the government’s ability to
carry out its declared program(s), and its ability to stay in office. This will
depend on the type of governance, the cohesion of the government and

Breaking the Barriers to Higher Economic Growth458
governing party or parties, the closeness of the next election, the govern-
ment’s command of the legislature, popular approval of government poli-
cies, and so on.”
• Internal Conflict “is an assessment of political violence in the country and
its actual or potential impact on governance. The highest rating is given to
those countries where there is no armed opposition to the government and
the government does not indulge in arbitrary violence, direct or indirect,
against its own people. The lowest rating is given to a country embroiled in
an ongoing civil war.”
• External Conflict “is an assessment both of the risk to the incumbent gov-
ernment and to inward investment. It ranges from trade restrictions and
embargoes, whether imposed by a single country, a group of countries, or
the international community as a whole, through geopolitical disputes,
armed threats, exchanges of fire on borders, border incursions, foreign-
supported insurgency, and full-scale warfare.”
• Ethnic Tensions “measures the degree of tension within a country attrib-
utable to racial, nationality, or language divisions. Lower ratings are given
to countries where racial and nationality tensions are high because oppos-
ing groups are intolerant and unwilling to compromise. Higher ratings are
given to countries where tensions are minimal, even though such differ-
ences may still exist.”

Part III: Trade, Competitiveness, and Investment 459
Annex 17.4. Principal Component Analysis
Table A17.4.1. The Administrative Quality Indicator
Component Eigenvalue Cumulative R2
P1 2.23 0.56
P2 0.83 0.76
P3 0.51 0.89
P4 0.43 1
Loadings P1 P2 P3 P4
Corruption 0.49 ⫺0.57 0.06 0.65
Bureaucracy Quality 0.54 ⫺0.08 0.64 ⫺0.54
Investment Profile 0.41 0.81 0.08 0.40
Law and Order 0.54 ⫺0.02 ⫺0.76 ⫺0.36
QA = P1*(0.5577/0.7640) + P2*(0.2063/0.7640)
Table A17.4.2. The Political Stability Indicator
Component Eigenvalue Cumulative R2
P1 2.24 0.56
P2 0.70 0.74
P3 0.69 0.91
P4 0.36 1.00
Loadings P1 P2 P3 P4
Government Stability 0.45 0.65 0.57 0.22
Internal Conflicts 0.58 ⫺0.06 ⫺0.08 ⫺0.81
External Conflicts 0.48 0.18 ⫺0.75 0.41
Ethnic Tensions 0.47 ⫺0.73 0.33 0.36
PS = P1* (0.5604/0.7356) + P2* (0.1752 /0.7356)
Table A17.4.3. The Public Accountability Indicator
Component Eigenvalue Cumulative R2
P1 1.88 0.94
P2 0.12 1
Loadings P1 P2
Political Rights 0.71 0.71
Civil Liberties 0.71 ⫺0.71
PA = P1

Breaking the Barriers to Higher Economic Growth460
Table A17.4.4. The Governance Indicator
Component Eigenvalue Cumulative R2
P1 3.94 0.39
P2 1.64 0.56
P3 1.2 0.68
P4 0.91 0.77
P5 0.69 0.84
P6 0.47 0.89
P7 0.43 0.93
P8 0.33 0.96
P9 0.26 0.99
P10 0.13 1
Loadings P1 P2 P3 P4 P5 P6 P7 P8 P9 P10
Corruption 0.25 ⫺0.15 0.61 0.19 0.00 0.53 0.46 0.06 ⫺0.05 0.03
Bureaucracy Quality 0.30 ⫺0.13 0.35 0.51 ⫺0.10 ⫺0.68 ⫺0.13 0.12 ⫺0.10 0.03
Investment Profile 0.33 ⫺0.03 ⫺0.43 0.40 0.11 0.02 0.31 ⫺0.63 0.18 ⫺0.08
Law and Order 0.37 ⫺0.31 0.13 ⫺0.12 0.06 0.20 ⫺0.58 ⫺0.05 0.59 ⫺0.05
Political Rights 0.26 0.63 0.08 0.07 0.06 0.09 ⫺0.15 0.11 ⫺0.06 ⫺0.69
Civil Liberties 0.26 0.63 0.04 0.03 0.07 0.08 ⫺0.13 ⫺0.01 0.06 0.71
Government Stability 0.31 ⫺0.20 ⫺0.52 0.24 0.16 0.21 0.01 0.67 ⫺0.16 0.06
Internal Conflicts 0.41 ⫺0.18 ⫺0.03 ⫺0.30 ⫺0.16 0.10 ⫺0.27 ⫺0.31 ⫺0.71 0.03
External Conflicts 0.33 0.07 ⫺0.14 ⫺0.33 ⫺0.73 ⫺0.15 0.34 0.16 0.26 ⫺0.02
Ethnic Tensions 0.31 ⫺0.04 0.08 ⫺0.52 0.61 ⫺0.36 0.34 0.07 0.06 ⫺0.02
GOV = P1*(0.3937/0.7696) + P2*(0.1641/0.7696) + P3*(0.1204/0.7696) + P4*(0.0915/0.7696)
Table A17.4.5. The Structural Reform Indicator
Component Eigenvalue Cumulative R2
P1 1.49 0.75
P2 0.59 1
Loadings P1 P2
Trade Policy 0.71 0.71
Private Credit 0.71 ⫺0.71
SR = P1

Part III: Trade, Competitiveness, and Investment 461
Table A17.4.6. The Human Capital Indicator
Component Eigenvalue Cumulative R2
P1 3.14 0.78
P2 0.38 0.88
P3 0.31 0.96
P4 0.18 1
Loadings P1 P2 P3 P4
Life Expectancy 0.52 ⫺0.33 0.03 ⫺0.79
H1 0.50 ⫺0.41 0.55 0.53
H2 0.50 ⫺0.05 ⫺0.80 0.32
H3 0.48 0.85 0.23 ⫺0.03
H = P1

Breaking the Barriers to Higher Economic Growth462
Annex 17.5. Short-Term Coefficients of the Disaggregated Indicators
Table A17.5.1. Direct Effect on Private Investmenta
Short-term coefficients
Index Variables Standardized variables Level variables
GOV Corruption 0.49 0.45
Bureaucracy Quality 0.54 0.52
Investment Profile 0.32 0.15
Law and Order 0.29 0.23
Political Rights 0.64 0.32
Civil Liberties 0.63 0.39
Government Stability 0.14 0.06
Internal Conflict 0.29 0.11
External Conflict 0.27 0.12
Ethnic Tensions 0.22 0.15
SR Trade Policy 1.40 0.05
Private Credit 1.40 0.07
Source: Authors’calculations.
Note: a. Direct impact is calculated using the estimated coefficient of the aggregated indicators (GOV and SR, see
equation (9), Table (17.1)), as well as the weights of each principal component in the aggregate indicators, com-
bined with the loading of the initial variables in each principal component (see Annex 17. 4).
Table A17.5.2. Indirect Effect on Private Investmenta
Short -term coefficients
Index Variables Standardized Level Level
variablesbvariablesbvariablesc
HLife Expectancy 0.06 0.01 0.01
Primary Education 0.06 0.04 0.09
Secondary Education 0.06 0.09 0.19
Tertiary Education 0.06 0.51 1.14
Source: Authors’calculations.
Note: a. Indirect impact is calculated using the coefficient of the human capital indicator in the governance equa-
tion (column 10, Table 17.1), multiplied by the coefficient of the governance indicator in the investment equation
(column 9, Table 17.1), in addition to the weights of each principal component in the aggregate indicators, com-
bined with the loading of the initial variables in each principal component (see Annex 17.4).
b. Coefficient entering the governance equation (2).
c. Coefficient entering the private investment equation (1).

Part III: Trade, Competitiveness, and Investment 463
Annex 17.6
Table A17.6. Level of Variables with an Increase of One Standard Deviation Compared to the Actual Level in East Asia
SR QA PA PS H
trade priv corrup bur inves law pol civ gov int ext ethn life
pol cred tion qual prof ord rights lib stab confl confl tens exp Iary Iiary IIIary
1980s
Egypt 1.0 1.0 1.0 1.2 1.0 1.2 0.9 1.1 1.3 1.2 1.0 2.0 1.0 0.8 1.6 1.3
Iran 0.4 1.2 1.4 0.8 1.1 0.9 0.7 0.7 1.1 0.8 0.4 1.4 1.1 0.9 1.5 1.0
Morocco 0.9 0.9 1.1 1.4 1.1 1.0 1.2 1.0 1.5 1.1 0.9 1.9 1.1 0.3 0.5 0.7
Tunisia 1.1 1.6 1.4 1.3 1.1 1.0 0.7 1.0 1.2 1.2 0.8 2.8 1.1 0.9 1.4 1.1
1990s
Egypt 0.9 0.9 1.2 1.3 1.3 1.2 1.1 1.1 1.4 1.1 1.1 1.9 1.1 1.0 1.5 1.6
Iran 0.6 0.8 1.5 1.4 1.1 1.3 1.1 0.9 1.4 1.3 1.0 1.8 1.1 1.0 1.5 1.4
Morocco 1.0 1.0 1.3 1.3 1.4 1.5 1.3 1.3 1.5 1.3 1.1 1.7 1.1 0.3 0.4 0.7
Tunisia 1.3 1.3 1.3 1.3 1.4 1.3 1.1 1.2 1.4 1.4 1.1 1.8 1.2 1.0 1.3 1.4
Source: Authors’calculations.
Note: Figures represent the level potentially reached by the MENA countries when increasing the control variables by one standard deviation
divided by the levels achieved by the East Asian economies. For example, the value 1 means that an improvement of one standard deviation has
increased the reform indicator in MENA to the level of the East Asian economies. Similarly, the value 1.2 indicates that the level reached by MENA
is 20 percent higher than the level observed in East Asia.

465
3SLS. See three-stage least square
accountability. See public
accountability
acquis communautaire, 276
ADF. See Augmented-Dickey-Fuller test
administrative quality. See Quality of
Administration
agreements. See trade agreements
agricultural trade liberalization, 281
agriculture and labor demand in analy-
sis of labor market reforms,
218–19
agriculture employment, decline of,
332
agriculture policy and industrial policy,
152b
agriculture volatility, 301–2
aid flow decline, 14
aid inflows, per capita by region, 275f
aid to GDP ratio, 15f
air transport, 321
Akaike criterion, 413
Algeria, 12, 140
human capital and TFP growth,
179–80
unemployment, 187n, 339, 340
Arab region, 27n, 28n, 96
Argentina, 78, 81, 83
exchange rate regime, 79–80
GDP growth, 82f
government spending, 80–81, 80f
lessons learned from, 85, 87
weakening of institutional environ-
ment, 82–87
association agreements between EU
and MED countries, 268
Association of Southeast Asian Nations
(ASEAN), 20
Augmented-Dickey-Fuller (ADF) tests,
37, 66t
ERER model, 384
RER misalignment, 396t
authoritarian regimes and economic
success, 125
Bahrain, 12
balance of payment equations for labor
market analysis simulations, 261
banking crisis, Argentina, 81,83
banking sector reform, 87, 187
banking sector, Iran, 150b
banking system, 32, 74n
as indicator, 59, 59f
government control of, 147–48
Barcelona Framework, 280–82
binding constraint to growth approach,
127–28, 129
Index

466 Index
black market exchange rate, 49n,56f,
69t
Brazilian Real, devaluation, 84
bribery, 304
budget closure rule, 235
Buenos Aires Stock Exchange (BASE),
84
bureaucracy quality, 428, 444
business as competition driver, 299–300
business climate, 121–22, 303–4. See
also investment climate
business, starting up, 140, 320
capital controls, 377
capital deepening, future improvement,
324
capital growth impact on employment,
344
capital inflow increase, 378n, 392n
capital inflows in ERER model, 384
capital markets, 47n
lessons learned in Argentina, 83–84
capital stock estimates, 191–92
cash transfers to industries by region,
161f
Central and Eastern European countries
(CEECs) countries, 276, 353n
economic integration, 271
FDI and EU trade network integra-
tion, 352f
FDI and growth, 351
labor force growth, 290t
migration to EU, 282–83, 289
Chile, exchange rate regime, 79
Choleski decomposition, 405
Civil Liberties, 456
civil society restrictions, 154
clientelism, 156–57
coalitions, inability to form, 96
coefficients’ calculations, 67, 68t
collective action, education example,
143
collective action, restricting, 98
collective bargaining, 215
Commonwealth of Independent States
(CIS) countries, 351
comparative advantages and innova-
tion, 155
competitiveness drivers, 299–300
competitiveness, effects of exchange
rate management on, 318
complementarity effect, 401
component analysis, 47n, 53.See also
principal component analysis
composite reform program simulation
results, 247–48
conflict, 27, 327n, 444, 458
effects on investment climate,
321–22
impact on competitiveness, 302
construction in value added, share of,
184, 185t
consumption in analysis of labor mar-
ket reforms in labor-exporting
countries, 228, 260
contract enforcement, 320–21
Control of Corruption, 454
Control over Corruption, 456
corruption, 141, 428, 444
governance indicator, 454
credit allocation as industrial policy,
140
credit controls, 163t–164t
crowding in of private investment, 232,
235, 400
public sector layoff simulation,
239
crowding out of private investment,
232, 234, 250, 400, 417n–418n
and complementarity, 401–2
in simulations, 249
currency crisis, Argentina, 81–82
currency overvaluation, 300, 318, 390
current account balance, 44, 47n, 58f
current account deficit, 71t
Cyber City, 139
466

467Index
Czech Republic, economic integration,
271
data definitions and sources for assess-
ing public and private invest-
ment, 421–22
debt, 32, 85
external, 57, 57f, 58f, 70t
reduction, 44, 46
renegotiations, 91
services, ERER model, 384
delivery of public goods and gover-
nance, 122
delocalization of production, 336
demand, excess, 410
demand- and supply-side effects on
output, 403
democracy, 103, 130n
and economic growth, 113–14, 114f,
115–18
and income, 110, 114,117
benefits of, 128
definition of, 106
design for resource-rich countries,
126
fostering/hindering growth, 104–5,
115, 124, 127–28
gap between MENA and other
regions, 108f
impacts on governance, 120, 123
indicator, 455
measuring, 107
or just better governance, 124–26
quality of, 106
strive for sustainable, 129
trends, 108f
democracy deficit, 106, 107–10, 128
democratic institutions, 123, 444
and oil-producing countries, 126
design, 106
democratic reforms, 129
democratization results, 125
deregulation, 78
development and employment
impacts, 8
development gaps, 47
development outcomes lagging in
MENA region, 305
dictatorships, 130n
diversification, 21, 334, 336
low due to currency overvaluation,
390
to sustain growth, 391n–392n
doing business, 16–17
domestic credit directed to private sec-
tor, 164t
East Asia, 174, 175
governance indicators, simulation,
463t
growth and education, 204
economic credibility,85
economic domination by governments,
399
economic growth, 29, 35–36,306, 306f,
307f
1960s and 1970s, 177
and democratic institutions, 123
and employment and productivity,
173t
and public sector employment, 206
and stabilization, 38
binding constraint approach,
127–28, 129
climate, 13
contributors to, 45t,69t–73t
summary of, 74
debt reduction influence, 44
deficit, 106, 110–13
democracy impact, 104–5
democracy links, 113–14, 115–18
effects of public sector size, 186
equations, 36, 37t
factors, 46
gap, 183, 183t
human capital impact, 39–41

468 Index
importance of higher education,
204
in early 2000s, 113
institutional effects, 116–17, 121
job creation, 175, 184
link to reforms, 36
MENA region
1990s, 171, 182
poor performance in, 176–77,
179–84
necessary realignments for sustain-
ability, 90
public sector layoff simulation
results, 239
reasons for lag, 92
reform program simulation results,
248
sources, 323
under authoritarian regimes, 125
variables’ impacts, 38
wage cut simulation results, 238
weakness causing unemployment,
307
weakness due to slow structural
reform, 399
economic indicators for labor-
exporting countries, 214f
economic performance, 77–78
democracy and oil-producing coun-
tries, 126
disparities due to political market,
117
reasons for country variance, 156
economic policy, 3–4, 89
economic realignment, 15–16, 90
economic reform, 30, 36, 89
boosting private investment, 443,
446
cost of, 27
lack of in MENA, 42
linked to political reform, 26
economic stability. See macroeconomic
stability
Economic Stability (ES) index, 274fn
economic system, historical model,
110–13
economic transformation, 25, 355–56
economic transition costs, 99
education, 119, 432
and unemployment, 6f
in MED countries, 278t
change of focus, 203
collective action unlikely, 143
enhancing quality of, 23–24
GDP impact, 40, 40t
higher education, 204, 205
human capital indicators, 61, 62f
indicators, 48n
labor force payoff expectations, 279f
primary, 72t
education reform, 205
education returns, 203, 251n
Egypt, 12
fiscal discipline, 86
human capital accumulation and
TFP, 180–81
manufactured exports growth, 387
new labor entrants, 277
private investment, 411f, 413, 444t
and public investment, 407f
response to increase in public
spending, 415f
unemployment, 278t
elasticities’ calculations, 67, 68t
electoral market imperfections, 117,
156
elites, networks of privilege, 158n
employment, 4, 15–16. See also job
creation
and gender, 24
and growth and productivity, 173t
and output growth, 188n
bridging export gap to increase, 343
decline in agriculture, 332
development impact, 8
effects of reform followed by trade

liberalization, 342b
first-time seekers, 6f
GDP growth effects, 324
manufacturing sector
determinants for developing
countries, 343t
determinants for high-income
countries, 344t
impact of FDI, 330
negative effect of wages, 341
relationship with trade, 329
trade expansion, 340–45, 341f,
345f
weak impact of, 345–49
weak impact of trade and FDI,
347–48, 348t
meeting challenges of, 21
payroll tax reduction simulation
results, 234
public sector, sustainability of, 278,
280
results of composite reform pro-
gram simulation, 248
skilled and unskilled simulation
results
payroll tax reduction, 233f, 234
public sector layoff, 240f, 241f,
242, 243
public sector wage cut, 236f, 237f
reduction in union bargaining
strength, 246f
subsidies to private employment,
244f
unskilled as engine for growth in
simulation results, 243
employment elasticities, 183
employment equations for labor mar-
ket analysis simulations, 255–57
employment growth
compared to labor force growth
(1990s), 171–72, 172t
lack of due to slow structural
reform, 399
to absorb growing population, 331,
333f
employment opportunity decline,
13–14
employment policy realignments, 25
employment subsidies, 252n
simulation, 243, 244f, 245
employment, under-, 213, 215
equilibrium real exchange rate (ERER),
367n, 374, 378n
calculating, 373
measuring RER gap, 382
modeling long-term, 383–85, 385t
ES. See economic stability index
Ethnic Tensions, 458
EU-MED agreements, 99, 268–69,
269t, 294n
evaluating content and limitations,
272–73
limited achievements, 269–71
missing reform agenda, 276t
structure inhibits implementation
of reforms, 274–75
EU-MED Guest Worker Program, 292
EU-MED migration scheme, 291–92
EU-MED Partnership, 267–68, 293
European Union (EU)
absorbing CEECs parts exports,
353n
accession
and integration, 271
and migration, 282–83
partnerships, 276
reforms, 276t
labor market integration with MED
countries, 283–85
migration scheme, MED, 290–93
migration, coping with, 287–90
migration, current and projected
net, 289f
population, shrinking of, 289
trade network integration, 352
and FDI, 352f
469Index

European Union (EU) Association
Agreements, 98–99, 208, 351
event analysis, 116
excess demand, 410
Exchange Arrangements and Exchange
Restrictions: Annual Report, 376
exchange rate classifications, variables,
376–78
exchange rate management, 79–80,
163t, 185–86, 300, 318. See also
fixed exchange rate regimes
adopting flexible, 391
choice of regime, 361–65, 362tn
de facto, 367n
de jure, 376
economic and political models, 361
floating, 364
not always freely, 366n, 376
for private sector development, 356
models and variables, 364–65, 365t
negative impact of mismanagement,
360
predicting regimes, 361–62, 362t
reflection of political economy, 366
rigidity, 357
standard and augmented models,
365t
trade policy impacts, 338
exchange rate misalignment, 356
exchange rate policy, 389
exchange rate volatility, 319t, 386
exports, 17, 149, 186
as percent of GDP, 309f
bridging gap to increase employ-
ment, 343
concentration of, 329
creating new jobs, 324–25
diversification, 334, 391n–392n
Tunisia, 336
employment in Morocco, 339f
exchange rate management, 318
exchange rate mismanagement, 360
intraregional, 20
negative impact of RER misalign-
ment, 382
nonoil potential, 18f
patterns, 335f
percent of merchandise imports,
272f
performance and output growth,
335f
performance of MED countries, 270
share of oil, 363, 363f
exports, manufactured, 60t, 390
and overvaluation, 358–61
and RER management, 387
by country, 387t
impact of RER overvaluation, 389
model of, 387–88
econometric results, 388–91,
389t
terms-of-trade sensitivity, 391
exports, nonhydrocarbon, 346
exports, nonoil, 309–11, 310f
External Conflict, 458
external debt, 57, 57f,58f, 70t
external stability, 43–44, 44t, 45, 46,
70t, 74t
indicators, 31–32, 32f, 39, 57
variables, 53t
factor endowment trade theory, 281–82
factor inputs per laborer, 179f, 180f,
181f
financial depth, 59
financial development, 71t, 431
financial sector, 150b, 151
and industrial policy, 147–48
reform, lessons learned, 83
financial services, 321
financial sources, 13
financial system, importance of devel-
opment, 444, 446
firm theory, 430
firms as competition drivers, 299–300
fiscal deficit reductions, 91
470 Index

fiscal discipline, lack of, 80
fiscal strain, 112
fixed capital growth per laborer, 178t
fixed exchange rate, 84, 300–301, 318
regimes, 79, 356, 366
likely to become overvalued, 357
reasons for preferring, 362–63
fixed-effect methodology,392n
foreign debt, 57
foreign direct investment (FDI),
313–15, 314f, 315f
and economic integration, 271
and EU trade network integration
in CEECs, 352f
and impact of trade on employ-
ment, 330, 349f
and remittances, as percentage of
GDP, 285t
as a percent of GDP, 272f
investment climate an attracting
factor, 351
MENA region not attracting, 349,
350f
security of jobs created by, 353n
weak impact of trade on employ-
ment, 347–48, 348t
foreign exchange, 47n
foreign market exposure, 208
foreign population in selected EU
countries, 288t
formal sector simulation results
public sector layoff, 240f, 241f
public sector wage cut, 236f, 237f
reduction in union bargaining
strength, 246f
subsidies to private employment,
244f
fractionalization, 124
fragmentation of production chains,
350
Freedom House, 456
Political Rights index, 108–9, 109f
public accountability,429
freedom of the press, 96, 154
general equilibrium effects, 250
geopolitical context, 322
Germany, main EU trading partner,
353n
globalization, 17, 19, 21
failure to take advantage of, 308
removing insulation barriers, 208
slow integration of MENA region,
335
governance
agenda to meet employment chal-
lenges, 21
and drive for democracy, 124–26
and private investment, 121–23,
433–34, 442–45
challenges, 97, 123
democracy impacts, 120, 123
democratic vs. nondemocratic, 125
education, efficient use of, 209
educational attainment impact on
quality of, 432
indicators, 165t, 454–55, 456–58
PCA, 459–61, 459t
short-term coefficients, 462t
standard deviation increase sim-
ulation, 463t
investment climate impact, 425
oil exports and political stability,
440
oil-producing countries, 126–27
preventing private sector develop-
ment, 303–4
private investment decisions,
441–42
private investment impact, 434–42
estimation results, 437–40,
438t–439t
private investment influencing qual-
ity of, 426
private sector,impact on, 186
public accountability impact, 441
471Index

rich natural resources impact, 436
suited to homogenous societies, 124
under authoritarian regimes, 125
weaknesses, 96, 154
governance institutions, 123, 424–25
classification of, 426, 427–30, 454
deficit of “good,” 423–24
importance to investment climate,
445
negative impact on private invest-
ment decisions, 433–34
government
credibility, 121
dominating economy, 399
equations for labor market analysis
simulations, 260
expenditure
and volatility, 302
lowering private investment,
417n
feasibility of economic transforma-
tion, 25–26
in labor market reform analysis,
229, 230–31
information, access to, 96
public sector layoff simulation
results, 243
revenue decline, 11
Government Effectiveness, 454
Government Stability,457–58
Greece, GDP per potential worker,272f
gross domestic product (GDP)
payroll tax reduction simulation
results, 234
per laborer growth, 323t
per laborer growth, accumulation
growth, and productivity, 178t
per potential worker, 272f
ratio to aid, 15f
gross domestic product (GDP) growth
and job creation, 14–15, 183t
and TFP growth, 182
Argentina, 82f
conflict effects, 322
effect on employment, 324
gap, 183t
impact of public sector employee,
207
per capita, 11f, 13f, 36, 74t, 323t
1990s, 170
average (1965–2004), 111f
contribution of infrastructure in
MENA Region, 41t
growth rates, 35f
contributor summary, 45t
education impact, 40
human capital and infrastruc-
ture contributions, 45
structural reform contribu-
tion, 43, 44t
MENA region, 40t
per laborer and factor accumulation
and TFP growth, 182t
Gulf Cooperation Council (GCC)
economies, 12
health, 40t, 41
higher education, 204, 205
horizontal industrial policy, 138, 153
justifications for, 141–42
long-term nature of, 143
transition to, 151, 154
hubs, services and industrial, 139
human capital, 72t, 323t
accumulation, 119, 179f, 180f, 181f,
190
increase due to democracy,115
and openness and TFP, 207
decline of, 184
effect on manufactured exports, 390
growth contributions, 46–47, 74t
growth impact, 39–41
growth per laborer, 178t, 179
improving private capital forma-
tion, 432
indicators, 33, 34f, 61, 459t
472 Index

indirect impact to GDP, 443, 443t
leverage of, 205
loss due to unemployment, 170
private investment impacts, 442
trade influence, 207–8
variables, 54t
human development indicators, 119,
119f
Hungary,economic integration, 271
ICORs, 409–10
ICRG. See International Country Risk
Guide
IFC. See International Finance
Corporation
IMF. See International Monetary Fund
IMMPA. See Integrated Macroeco-
nomic Model for Poverty
Analysis
import potential, 312f
import substitution strategy, 49n
import-competing industries, 337
imports, openness to, 311–12
impulse response analysis, 413, 418n
inclusiveness, 21, 96–97, 156
income
and democracy, 105, 110, 111f, 114f
gains due to, 117
lack of correlation, 114
income equations for labor market
analysis simulations, 259–60
index of public accountability (IPA),
110, 111f
India, socialist industrial policy, 140
indicators, elasticities/coefficient calcu-
lations, 67, 68t
indirect output and price effects, 402–4
Indonesia, 325, 336
industrial development banks, 147
industrial export sector, 153
industrial policy, 146, 152b, 157n
and political economy, 142–44
and redistribution, oil wealth, and
subsidies, 148
better, not more, 155
infant industry protection, 139
interventions, 135, 152b
financial sector, 147–48
mainstream view, 135
MENA region, 136–37
failure to change policy, 149–54
forces reshaping, 156
private sector adaptation, 152–53
socialist, 140
subsidies, 151
to support social contracts, 144
traditional approaches, 136
updating, 155
vertical and horizontal, 138
vertical to horizontal transition, 151
vertical, justification for, 139–41
industrial zones, 139
industry, starting up domestically, 140
infant mortality, 61, 72t
inflation, 49n, 55f, 69t
historical rate for analysis, 377
informal sector simulation results
payroll tax reduction, 233f, 234
public sector layoff, 240f, 241f
public sector wage cut, 236f, 237f
reduction in union bargaining
strength, 246f
subsidies to private employment,
244f
informal sector, share of labor force,
213
information externalities, 139–40
information infrastructure, 208
information, access to, 96
infrastructure. See also public
infrastructure
infrastructure improvement for busi-
ness and finance, 17
infrastructure indicator, 34–35, 34f
infrastructure, physical, 41, 41t, 72t
growth contributions, 46–47, 74t
473Index

indicators, 63, 63f
variables, 54t
innovation and comparative advan-
tages, 155
inputs per laborer, 179f, 180f, 181f
inputs, reduced demand, 417n
instability, macroeconomic, 170
institutional deficit, 49n
institutional effect on economic per-
formance, 116–17
institutional environment, weakening
in Argentina, 82–87
institutional quality, 121, 454–55
institutions. See governance
institutions
Integrated Macroeconomic Model for
Poverty Analysis (IMMPA), 251n
integration. See regional integration;
trade integration
integration, international, 305
interest groups and industrial policy,
141
interest rates, 163t
Internal Conflict, 458
International Country Risk Guide
(ICRG), 428, 456, 457
International Finance Corporation
(IFC), 416
International Monetary Fund (IMF),
82, 376
investment. See also private investment;
public infrastructure investment
attracting foreign and domestic, 326
growth contributions, 74t
ICORs, 409–10
impact of conflict, 302
interaction with trade openness,
320f
labor market reforms in labor-
exporting countries, analysis of,
228
public crowding in and out of pri-
vate, 400
public sector wage cut simulation
results, 236f, 237f
investment climate, 12,16–17
and impact of trade on job creation,
349f
and trade expansion, 319–20
classification of institutions, 428
conflict effects, 321–22
critical role of, 319–20
development barriers, 185–86
factor in migration and return, 286
importance of governance, 425,
445
improving to realize liberalization
benefits, 352
inhibiting private sector involve-
ment, 417
potential gains from reform,
322–25, 323t
public accountability,429
summary of issues, 320–21
to attract FDI, 351
Investment Climate Surveys, 16
investment decisions, 275. See also pri-
vate investment decisions
investment decline, 184–85
investment equations for labor market
analysis simulations, 260
investment in education, 119
investment indicators, 65,65f
investment model, 430–31
Investment Profile,428, 456
index, 447n
investment rate and ERER model, 384
investment, payroll tax reduction sim-
ulation results, 233f
investor obstacles, 16
IPA. See index of public accountability
(IPA)
Iran, Islamic Republic of, 150b, 180,
444t
Israel, stabilization plan, 79
Italy, 272f, 283f
474 Index

job creation, 4, 8,90, 184. See also
employment
accelerating pace of, 330, 331, 333f
and investment climate, 349f
challenges to, 169
due to exports, 324–25
economic transformation impacts,
25
end of traditional modes, 91
FDI impact, 347–48
importance of diversification, 21
lack of impact of trade openness,
331
migration, 14
needed to absorb unemployed, 355
public sector, 175
trade expansion, 326, 337f
with old development model, 9–14
job creation rates, 90, 91f
and GDP growth, 183t
job destruction, 338–39
Jordan, 11–12, 16, 139
manufactured exports growth, 387
public and private investment,
407f
response of private investment to
increase in public infrastructure
spending, 415f
unemployment, 278t
variance in private investment, 411f,
413
judicial system, unpredictability of, 17
knowledge gap, 209
knowledge transfer, 311
knowledge-based economies, 209
Korea, industrial policy, 140
Kuwait,human capital accumulation
and TFP, 180–81
labor demand, 212, 219–21
labor force, 213
growth, 8, 8f, 9f, 90, 173–74, 211, 290
and real GDP growth, 174, 174t
compared to employment
growth (1990s), 171–72, 172t
participation of women, 5, 7f, 24
labor market, 6–7, 8, 213
cause of disruptions, 212
characteristics in MED countries,
277
composition and structure of, 216,
217f
indicators for labor-exporting coun-
tries, 214f
interaction with trade and job cre-
ation, 347
regulations, 215
segmentation, 251n
social crisis in MED countries,
277–78, 280
unemployment, 5, 211
labor market outcomes, 183
disappointing, 171–72, 173–76
labor market policies, crowding in and
crowding out effects, 232
labor market reform, 212
and interaction with fiscal policy,
232
piecemeal vs. comprehensive
approach, 250
labor market reform analysis model,
behavioral parameters of, 231
labor market reform analysis
simulations
composite policy package results,
250–51
equations, 255–61
parameters for, 265–66
policy lessons from, 249–51
labor market reforms in labor-
exporting countries, analysis of,
216–29
agriculture and migration, 218–19,
220
composite reform package, 232
475Index

government, 229, 230–31
other model features, 226–29
payroll tax reduction simulation,
232, 233f, 234–35
pension systems, 225–26
simulation experiments, 231–32
urban sector, 219–24
labor mobility. See migration
labor unions. See unions
labor-exporting countries, analysis of
labor market reforms, 216–29
labor-exporting countries, economic
and labor indicators for, 214f
laborer growth, 178t
Latin America, economic reform after
crisis, 151
Law and Order, 429, 456
layoffs in the public sector, simulation
results, 239, 240f, 241f, 242–43
Lebanon, reforms, 12
legal system, unpredictability of, 17
Levy-Yeyati and Sturzenegger (LYS)
exchange rate classifications,
376–77
liberalization. See trade liberalization
lobby power in manufacturing sector,
153, 153t
lobby power, measuring, 158n
lobbying for policy change, 94, 95f
lobbying, by sectors, 143
macroeconomic instability, 170
macroeconomic policy, 55
macroeconomic reform, 46
indicator, 31, 31f
macroeconomic stability, 38, 69t, 92
growth contributions, 74t
indicators, 55
negligible achievement, 91
macroeconomic stabilization reform,
112
macroeconomic volatility, 42
manufactured exports, 60t, 391
and RER management, 387
by country, 387t
effect of human capital, 390
impact of RER overvaluation, 389
model of, 387–88
econometric results, 388–91,
389t
overvaluation, 358–61, 360t
price elasticity, 392n
manufactures and products, labor-
intensive, 19
manufacturing sector, 334, 336
employment
determinants and FDI, 348t
determinants for developing
countries, 343t
determinants for high-income
countries, 344t
negative effect of wages, 341
trade impact, 345f
and investment climate, 349f
weak, 345–49
exchange rate mismanagement
impact, 360
floating exchange rate preferred, 364
job creation, effects of trade expan-
sion, 337f
lobby power, 94–95, 95f, 153, 153t
measuring for analysis, 378
trade expansion impact, 340–45,
341f
manufacturing specialization, 19
market failures, 419n
government intervention, 140
horizontal approaches, 143
industrial policy response, 139
market protection, 335, 336f. See also
trade protection
market-orientation indicators, 92
MED. See Southern Mediterranean
countries
MEDA, 268–69
MEDA I and II, 293n
476 Index

Menem administration, 80–81, 82
merchandise exports. See exports,
nonoil
merchandise trade growth, 333
MERCOSUR, 20
Mexico, 98
economic integration, 271
NAFTA accession, 276
tariff rates, 274
trade and job creation, 325
trade boost from reforms, 336
Middle East and North Africa (MENA)
region, 3–4, 129n, xn
attracting FDI, 349, 350f
better governance or democracy,
124–26
conflicts, 327n
democracy gap, 108f
democracy not always leading to
growth and better governance,
124
democracy trends, 108f
democracy, strive for sustainable,
129
dominant state role, 146
economic growth, 45t, 69t–73t, 190
decomposition (1990s), 182
in early 2000s, 113
performance, 29, 35–36, 45
1960s and 1970s, 177
and structural reform process,
190
poor, 176–77, 179–84
summary of, 74t
economic instability, 170
economic progress, 77–78
economic reform differences and
similarities, 30
exchange rate management experi-
ences, 300–301
failure to change policy, 149–54
fiscal discipline, 86
fixed exchange rates, 84
historical development model,
110–13
human capital growth contribution,
40t
improving education returns, 205
inclusiveness and accountability,
156
increasing private investment
through better governance,
442–45
industrial policy, reduced by exter-
nal mechanisms, 155
institutional deficit, 49n
investment in, 65, 65f
job creation challenges, 169
laborer GDP growth, factor accu-
mulation, and TFP growth, 182t
lack of gains from education, 204–5
lost decade of growth, 170
macoreconomic stability contribu-
tion, 43t
oil export domination, 149
oil revenues, 284f
old development model, 9–14
physical capital per laborer decline,
184
political and social environment not
reform friendly, 275
private investment, 122f, 424f
private sector development, 86
productivity (1960s–1970s), 149
public sector employment, 206, 207
sustainability, 278, 280
social contracts, 144, 145
structural reforms, 33, 33f
incomplete, 171
lagging, 42, 46
trade policy, 146
trade unions, 153
unemployment, 149
rates, 170, 171, 211, 212
vertical industrial policy, 144–46
outcomes of, 148–49
477Index

migrants, pool from MED countries,
291t
migrants, population in selected EU
countries, 288t
migration, 10, 14, 243, 247. See also
return migration
and trade liberalization, 282
current and net to EU, 289f
EU-MED partnership, 293
in analysis of labor market reforms,
218–19, 220, 225
including in Barcelona Framework,
281
international flows, 216
management, 285, 287, 290–93
maximizing for host and sending
countries, 282, 291
motivating factors, 294n
public sector layoff simulation
results, 242
reasons for poor reputation, 282
skills upgrading, 285
Southern Europe rates, 287f
to prevent skills from becoming
obsolete, 284
to support pay-as-you go pensions,
289
training fund, 292
United States and Italy, 283f
wage cut simulation results, 238
minimum wage, 215, 251n
model for analyzing labor market
reforms, 216–29
Morocco, 11–12, 16
exports and employment, 339f
gains due to economic reforms, 325
private investment, 444t
unemployment, 278t
mortality rate, 61t
National Company for Investment
(SNI, Tunisia), 147
nationalizations, 145
natural curse hypothesis, 436, 439
neoclassical accelerator model, 430–31,
435
validating, 438
neoclassical Solow growth model, 436
networks of privilege, elites, 158n
newspapers, 118
nontariff barriers, 18, 150, 317, 318
ranking of, 162t
reduction, 337
North American Free Trade Agreement
(NAFTA), 271, 276, 276t
oil as major source of income, 363
oil exports, 149, 440
per capita, 14f
share of total, 363, 363f
oil income, resource rent, 447n
oil price collapse, 10–11
oil price fluctuations, 301
oil resources, managing, 21
oil revenues, 93f, 284f
decline, 14
delaying reform, 93, 99
measuring public interest, 378
oil wealth, 146, 148
impact on governance and private
investment, 436
Oman, 12
OPEC revenues, 93f
openness. See trade openness
output elasticity, TFP estimates,
193t–201t
output growth and employment,
188n
output growth vs. growth of factor
inputs, per laborer (1960s vs.
1970s), 179f, 180f
output per laborer growth, 175–76
output volatility, 301
output, indirect, and price effects,
402–4
overvaluation, 390, 391
478 Index

and manufactured exports, 358–61,
360t
effect on prices, 360
RER misalignment, 357, 387–88,
390t
Pan-Arab Free Trade Agreement, 20–21
panel data regressions, 47n, 392n
parts and components, 316t
payroll tax reduction effects on
unskilled labor in model analysis,
232, 233f, 234–35
payroll tax simulation, 243, 252n
as part of composite program,
247–48
payroll taxes’effects on labor demand,
212
pension systems, 261
labor market reform analysis,
225–26
migration to support, 289
payroll tax reduction simulation,
235, 252n
public sector layoff simulation
results, 243
wage cut simulation results, 238–39
Perpetual Inventory Method (PIM),
406, 408
physical capital, 193t–201t, 323t
stock estimates, 191–92
physical infrastructure. See
infrastructure
Poland, economic integration, 271
policy change, lobbying for, 94, 95f
policy lessons from labor market
reform simulations, 249–51
policy, macroeconomic, 55
political change, containing, 98
political coalitions, disruption of, 142
political economy of industrial policy,
142–44
political economy of reform, 145
political instability, 115
political market imperfections, 117
political markets, 118, 157
political openness, 114
political participation, 110
political reform, link to economic
reform, 26
political regime universe, 130n
Political Rights index, 108–9, 109f, 457
political stability, 440, 444
index, 429–30, 457–58
PCA, 459t
Political Stability and Absence of Vio-
lence, 454–55
political system effectiveness, factors
for, 144
political systems in oil-producing
countries, 127
Polity index, 107, 108, 114f
population growth, 8, 331, 333f
urban and rural, 278f
population in EU, 289
population, characteristics of, 4
Portugal, 271, 272f
poverty reduction performance, 308,
308t
power sector, 321
price effects and indirect output, 402–4
price elasticity of manufactured goods,
392n
price equations for labor market analy-
sis simulations, 259
prices, 227, 316, 360
primary education, 62f, 72t
principal component analysis (PCA),
53, 418n, 430
assessing public and private invest-
ment, 411–12
political stability, 459t
private business, lobbying for change,
94–95
private capital formation, 413, 432
output and price change impacts,
402–3
479Index

positive effect of shocks, 414, 416
private credit, 32, 59
private investment, 122f, 246f, 424f. See
also crowding in and crowding
out of private investment; invest-
ment climate
and GDP, 443t
1980s and 1990s, 444t
variables, 445t
and governance, 121–23
assessing link with public invest-
ment, 404–6, 407f, 408
data sources and definitions,
421–22
assessment of governance impact,
433–34, 433–42
by region, 424f
decline of, 423
demand- and supply-side effects on
output, 403
determinants of in neoclassical
accelerator model, 435
governance, impact of improving,
122, 442–45, 446
government spending impact, 417n
influencing quality of governance,
426
institutional impact, lack of analy-
sis, 424–25
lack of, 184–85, 400
lack of response to public infra-
structure capital, 417
other determinants of, 430–32
public accountability impacts, 446
public sector layoff simulation
results, 240f, 241f
response to public spending
increase, 415f
short-term coefficients of indica-
tors, 462t
subsidies to private employment
simulation results, 244f
transmission channels, 401
variance decomposition of, 411f,
412f
private investment decisions, 431,
433–34
administrative quality, 437–40,
438t–439t
bureaucratic quality and corruption
impacts, 444
governance and economic policy
affecting, 445–46
governance impact, 441–42
private sector,16–17, 86
and domestic credit, 164t
and industrial policy, 146, 152–53
contribution to GDP by country,
16f
development indicators, 92
exchange rate management, 356
government support of, 184
increase in delivery of basic infra-
structure, 416–17
job creation, lack of, 280
labor demand in model analysis,
220
lack of response to productivity
growth, 302
old development model, 10
participation for successful reform,
273
payroll tax reduction simulation
results, 233f
privatization, accelerating, 17
production chain fragmentation, 350
production chain participation, 315,
316t
production equations for labor market
analysis simulations, 255
production, delocalization of, 336
production-sharing networks, 347
productivity, 171, 302. See also total
factor productivity
and GDP per laborer and accumula-
tion growth, 178t
480 Index

and growth and employment, 173t
and increasing output, 174–75
gains and increase in GDP, 324
low compared to investment, 400
MENA region (1960s–1970s), 149
negative growth, 181
positive effect of public capital, 401
productivity, worker, 172, 174, 175–76
profits and income, 228
property rights, 120, 121, 130n
public accountability,22–23, 96–97,
156, 456, 459t
and investor confidence, 443
boosting growth, 25
governance indicators, 165t
governance weaknesses, 154
impact on governance and private
investment decisions, 441
index, 110, 111f, 429
lack of demand for in oil-producing
countries, 127
private investment impacts, 446
public administration, impacts of
improvement on growth, 25
public administration, quality of in
MENA, 165t
public capital, 401, 408
public debt, 80–81
public deficit, 56f, 70t
public goods delivery and governance,
122
public infrastructure, 417n, 418n
demand exceeding supply, 410
improve quality or invest more,416
lack of effect on private investment,
417
quality indicators, 409–12
public infrastructure investment, 401,
402
assessing link with private invest-
ment, 404–6
data, 406, 407f, 408, 421–22
flows and stocks, 401, 405, 408–9
response of private investment to
increase, 415f
shocks, 413–14, 416
public sector employment, 347
absorbing workers and resources,
279f
and economic growth, 206–7
expansion of and need to scale back,
332
patterns in and pay, 334f
reduction as part of composite
reform program simulation,
247–48
public sector layoffs, simulation, 239,
240f, 241f, 242–43
public sector size, 86–87, 186
public sector wage bill, 112, 332,333f
public sector wage cut simulation, 235,
236f, 237f, 238–39
public sector wages, model for formal
sector, 347
quality indicators for public infrastruc-
ture, 409–12
Quality of Administration, 428–29,
444
governance estimation results,
437–40, 438t–439t
indicators of, 456
PCA, 459t
Quality of Bureaucracy, 428, 456
quota markets, 19
real exchange rate (RER), 367n, 371,
372
deviations from ERER, 374
overvaluation of, 373–74, 386, 397
impact on manufactured
exports, 389
positively affected by trade restric-
tions, 384
short-term dynamics, 397
volatility effects on growth, 382
481Index

real exchange rate (RER) management,
356, 381, 387
real exchange rate (RER) misalign-
ment, 367n, 378n
and manufactured exports, 388–91,
389t
and volatility, 358–59, 358t, 386t
estimation results, 359t
calculating, 371–74, 392n
costs suffered, 387–88, 390t
damaging to economic perform-
ance, 381
error correction model estimates,
398t
estimation results, 373t
measuring, 382, 385–86
redistribution, 115, 148
reform, 100, 128, 149–50. See also
structural reform
agenda, 97–98
banking sector, 87
challenges and responsibilities, 27
complementarities, 36
disrupting political coalitions, 142
economic, 89, 443, 446
failure to sustain, 92–94
financial sector, 83, 151
followed by trade liberalization,
342b
implementation capacity, 100
implementation, Barcelona frame-
work, 280–81
indicators, 31–35, 31f
coefficients/elasticities, 40t, 68t
initial efforts, 11–12
intensifying factors for, 156
lack of growth, 176
lagging implementation, 275
lessons learned, 83, 144
link between political and eco-
nomic, 26
macroeconomic and structural, 46
political economy of, 145
private sector response, 273
resistance to, 94–96
resulting from crisis, 93, 151
to encourage new business and job
creation, 353
weakness of, 92
reform program simulation results,
247–48
reform programs, managing, 98–100
regional integration, 269, 272–77
and EU accession, 271
limited results of EU-MED agree-
ments, 270, 270t
regional partnerships, 267, 271
regional trade agreements, 20
regressions, 47n
regulations, excessive, 304, 319
Regulatory Quality, 454
remittances, 14, 15f, 285, 285t
rentier effect, 447n
replacement migration, 289
representation, lobbying for change, 96
RER. See real exchange rate
resource-poor countries, 11–12
return migration, 284, 285, 286
training fund, 292
reverse accelerator effect, 403
right to manage approach, 222
rights necessary to mobilize for change,
96
road networks, 41t, 48n,63, 73t
Rule of Law, 454
Saudi Arabia, 12, 179, 182
savings, labor market analysis simula-
tions, 228, 260
secondary education, 62f
service delivery, weakness of,400
services liberalization, 281, 352–53
services trade, 312–13, 313f
services, financial, 321
shocks, 414, 416
skills acquisition, labor market reform
482 Index

model analysis, 224
skills, determining if needed, 208
skills, factors that inhibit demand, 206
skills, lack of, 23
SNI. See National Company for
Investment
social accounting matrix (SAM),
230–31
social cleavages, 124
social contract, 26, 136–37
acceptance of, 145
industrial policy, 144
influence on reform, 94
social crisis in MED countries, 277–78,
280
social polarization, 118, 124
social security contributions, 215
social security systems, migration to
support, 289
Solow hypothesis, not validated, 438
Southern Mediterranean (MED) coun-
tries, 267. See also EU-MED
Partnership
EU migration scheme, 290–93
export performance, 270
improving credibility of reform pro-
grams, 293
labor market integration with EU,
283–85
lack of integration and growth, 270,
270t
lag in reform implementation,
274–75
pool of potential migrants from,
291t
population growth, 277
slow transition to liberalized market
effects, 273
structural reforms lagging, 273, 274f
tariffs, 274
unemployment
and labor market characteristics,
277
by age and education level, 278t
wage bill, 279f
weak economy poses political and
social threats, 277–78, 280
weak growth rates, 271
Spain, 271, 272f
specialization, 19, 311
spillovers, 311
SR. See structural reform index
stability, 42, 46, 69t
economic, 43–44, 45
macroeconomic, 74t, 92
contribution to growth, 43t
indicators, 55
negligible achievement, 91
variables, 53t
stability, external, 70t, 74t
stabilization programs, 38
stabilization reform, 112
starting businesses, 320
structural reform, 46, 71t, 445
agenda, 13
and growth performance, 190
complementarities with stability, 38
contribution to GDP growth, 43,
44t
efficiency of, 36
growth contributions, 74t
incomplete, 171
indicators, 32–33, 33f, 59
PCA, 459t
lack of external stability correlation,
39
lagging, 42, 399, 431
MED countries, 273, 274f
private investment, 443t
variables, 54t
Structural Reform (SR) index, 274fn
subsidies, 148, 151
to private employment, 243, 244f,
245
unintended consequences, 252n
supply and demand, 226–27, 258
483Index

supply-side effects on output, 403
Syria, Arab Republic of, 12, 152b, 182
tariffs, 18, 150, 335, 336f
average by region, 161f
current average and standard devia-
tions by country, 162t
industrial policy, 143
MED countries, 274
reduction, 337
taxation, 115
technical progress. See total factor
productivity
telecommunications, 321
telephone lines, 41, 41t, 48n, 49n, 73t
telephone network, 63, 64f
terms of trade, 384, 391
tertiary education, 62f
theory of the firm, 430
three-stage least square (3SLS) estima-
tion, 435
time series analysis, 47n, 392n, 426
total factor productivity (TFP),
176–77, 323t
and openness and human capital,
207
growth, 178t, 179f, 180f, 181f, 190
1980s, 180, 181f
and GDP growth, 182, 182t
estimates, 191–92, 193t–201t
increasing to promote growth, 324
tourism, 143, 312–13
trade
and FDI, 314
concentration, 18
conflict effects, 322
equations for labor market analysis
simulations, 258
expanding intraregional, 20–21
growth in nonoil, 334
labor market reforms in labor-
exporting countries, analysis of,
227
merchandise, 333
outcomes, 308
patterns, 335f
performance, 309f
comparing across regions, 310,
310f
weakness of, 334
potential gains from reform,
322–25, 323t
protection, 316–18, 335, 336f
indicators, 317t
tariffs, 150
quality gap, 331, 346,350
regulation, 145
relationship with manufacturing
employment, 329
trade agreements, 19, 20, 99
trade balance, 230, 315
trade environment, enabling factors, 20
trade expansion, 334
and investment climate, 319–20
employment effects, 336–37
FDI reason for weak impact on
employment, 347–48
following domestic reforms, 337
indicators of, 341
job creation, 325, 330,337f
manufacturing employment,
340–45, 341f, 345f
and investment climate, 349f
effects different in developed
countries, 345
weak impact of, 345–49
measuring, 342, 342b
positive effect on employment in
developing countries, 341
potential, 326
trade integration, 17, 21, 92
barriers, 18–19
trade liberalization, 78, 207, 351
adjustment costs, 338–39, 339f, 340
agricultural trade and services, 281
and migration, 282
484 Index

anti-export bias reduction, 337
EU-MED association agreements,
268
fears of, 337–38
financial sector, 151
importance of investment climate,
352
lessons learned from Argentina, 87
measuring, 342b
services, 352–53
temporary increase in unemploy-
ment, 338
trade openness, 59, 74n, 311–12,378n
and wages, 339f
developing countries, 343t
interaction with foreign investment,
320f
lack of impact on job creation, 331
relationship with human capital and
TFP, 207
to increase private investment, 431
trade policy, 60f, 71t, 75n, 146
impact of exchange rate manage-
ment, 338
restrictions of, 87
restrictive, 208
stimulating private investment, 444,
446
trade reform, 12, 19, 186–87
impact on private investment deci-
sions, 431
Indonesia, 336
trade relations, intensifying, 281–82
trade theory, factor endowment,
281–82
trade-related services and production
chains, 350
tragedy of the commons, 99
Tunisia, 11–12, 147
export diversification, 336
manufactured exports growth, 387
private investment, 412f, 413, 444t
and public investment, 407f
response to increase in public
spending, 415f
unemployment, 278t
underemployment, 213, 215
unemployment, 5, 6f, 8, 11, 158n
“queuing” or “wait,” 213, 215
absorption of, 90
and wage growth decline, 339
due to slow economic growth, 307
effects of, 331
eliminate, 169
female workers, 24, 332, 333f
gender gap, 5–6
high for educated, 206
MED countries by age and educa-
tion level, 278t
MENA region, 149
of skilled and educated workers,
332, 333f
political and social costs of, 280
trade liberalization increase, 338
youth, 251, 278t, 280
unemployment rates, 5f
by gender, 7f
due to trade liberalization, 339f, 340
MED countries, 277
MENA region, 170, 171, 211–12,
307f
simulation results
reduction in union bargaining
strength, 246f, 247
subsidies to private employment,
244f
unskilled public sector layoff,
240f, 241f, 242
wage cut, 238
world comparison, 172f
union movement, 215
unions, 95, 153
reduction in bargaining strength
simulation, 245, 246f, 247–48
United Arab Emirates, 12
485Index

United States, growth rates and migra-
tion, 283f
VAR model, 404–6, 412–13
variable names and definitions, 262–64
vertical industrial policy, 138–41
1950s–1970s, 144–46
breeding dependency, 155
economic outcomes, 148–49
lack of success, 140
rent seeking and corruption, 141
transition to horizontal, 151
violence, 27
Voice and Accountability, 455
voice, lobbying for change, 96
volatility, 301–2
and RER misalignment, 358–59,
358t
estimation results, 359t
macroeconomic, 42
of exchange rate, 319t, 386, 386t
voters, informed, 118
wage bill, 279f, 332, 333f
wage cut simulation, public sector, 235,
236f, 237f, 238–39
wage decline, 251n, 339
wage determination, 215, 222–23
wages, 6–7
and trade openness, 339f
in government and manufacturing,
333f
negative effect on manufacturing
employment, 341
public sector as model for formal
sector, 347
simulation results
payroll tax reduction, 233f
public sector layoff, 240f, 241f
reduction in union bargaining
strength, 246f, 247
subsidies to private employment,
244f
women, labor force participation, 5–6,
7f, 24
worker productivity, 172, 179f, 180f,
181f
and wages, 188n
output per laborer, 175–76
workers, highly skilled, 204
World Trade Organization (WTO), 20
Yemen, reforms, 12
youth unemployment, 251n, 278t, 280
486 Index

SKU 17415
ISBN 978-0-8213-7415-3
The world’s attention to the countries of the
Middle East and North Africa (MENA) region
has often been dominated by headline issues:
con ict, sanctions, political turmoil, and rising oil
prices. Little of this international attention has
considered the broad range of development
challenges facing this diverse group of countries.
Breaking the Barriers re ects the collected
thinking of the World Bank’s O ce of the
Chief Economist for the MENA Region on
the long-term development challenges facing
the region and the reform priorities and strate-
gies for e ectively meeting these challenges.
It is a comprehensive reform agenda to “break
the barriers” to higher economic growth, to
ensure su cient jobs can be created for the
region’s rapidly growing labor force. At its core,
it requires the region’s public sector-dominated
economies to move to private sector-driven
economies, from closed economies to more
open economies, and from oil-dominated and
volatile economies to more stable and diversi ed
economies. This book examines some of these
reforms and the complex issues surrounding
their successful implementation. In order for
the countries of the MENA region to successfully
implement the reforms needed for higher
growth and job creation, they will also need
to address the fundamental weaknesses in
governance throughout the region.