Explaining Growth in the Middle East, Volume 278 (Contributions to Economic Analysis) (Contributions to Economic Analysis)

EXPLAINING GROWTH IN THE MIDDLE EAST CONTRIBUTIONS TO ECONOMIC ANALYSIS 278 Honorary Editors: D.W. JORGENSON J. TINB...

Author: Jeffrey B. Nugent | Hashem Pesaran

71 downloads 1184 Views 3MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!

Report copyright / DMCA form

DOWNLOAD PDF

EXPLAINING GROWTH IN THE MIDDLE EAST

CONTRIBUTIONS TO ECONOMIC ANALYSIS 278

Honorary Editors: D.W. JORGENSON J. TINBERGENy Editors: B. BALTAGI E. SADKA D. WILDASIN

Amsterdam – Boston – Heidelberg – London – New York – Oxford – Paris San Diego – San Francisco – Singapore – Sydney – Tokyo

EXPLAINING GROWTH IN THE MIDDLE EAST

Edited by Jeffrey B. Nugent University of Southern California, USA M. Hashem Pesaran University of Cambridge, UK

Amsterdam – Boston – Heidelberg – London – New York – Oxford – Paris San Diego – San Francisco – Singapore – Sydney – Tokyo

Elsevier Radarweg 29, PO Box 211, 1000 AE Amsterdam, The Netherlands The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, UK First edition 2007 Copyright r 2007 Elsevier B.V. All rights reserved No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means electronic, mechanical, photocopying, recording or otherwise without the prior written permission of the publisher Permissions may be sought directly from Elsevier’s Science & Technology Rights Department in Oxford, UK: phone (+44) (0) 1865 843830; fax (+44) (0) 1865 853333; email: [email protected]. Alternatively you can submit your request online by visiting the Elsevier web site at http://elsevier.com/locate/permissions, and selecting Obtaining permission to use Elsevier material Notice No responsibility is assumed by the publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. Because of rapid advances in the medical sciences, in particular, independent veriﬁcation of diagnoses and drug dosages should be made Library of Congress Cataloguing-in-Publication Data A catalog record for this book is available from the Library of Congress British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library ISBN-13: 978-0-444-52240-5 ISBN-10: 0-444-52240-9 ISSN: 0573-8555 For information on all Elsevier publications visit our website at books.elsevier.com

Printed and bound in The Netherlands 07 08 09 10 11

10 9 8 7 6 5 4 3 2 1

INTRODUCTION TO THE SERIES This series consists of a number of hitherto unpublished studies, which are introduced by the editors in the belief that they represent fresh contributions to economic science. The term ‘economic analysis’ as used in the title of the series has been adopted because it covers both the activities of the theoretical economist and the research worker. Although the analytical method used by the various contributors are not the same, they are nevertheless conditioned by the common origin of their studies, namely theoretical problems encountered in practical research. Since for this reason, business cycle research and national accounting, research work on behalf of economic policy, and problems of planning are the main sources of the subjects dealt with, they necessarily determine the manner of approach adopted by the authors. Their methods tend to be ‘practical’ in the sense of not being too far remote from application to actual economic conditions. In addition, they are quantitative. It is the hope of the editors that the publication of these studies will help to stimulate the exchange of scientiﬁc information and to reinforce international cooperation in the ﬁeld of economics. The Editors

This page intentionally left blank

PREFACE

This book is about the Middle East and North Africa (MENA) region. This region is generally perceived as having experienced the most disappointing growth performance over the last couple of decades among all regions in the world with the possible exception of Sub-Sahara, Africa. Despite the region’s immense endowment of natural resources, its per capita income is often viewed as having stagnated. At the same time, most economies of the region have been characterized by extremely high volatility, a condition only partly attributable to the ﬂuctuating price of oil. Another motive for writing about the MENA region is that it has been far less studied and researched than the other regions of developing countries such as Latin America, East Asia, South Asia, and Sub-Sahara, Africa. Much of what has been written about the region is conﬁned to the country level or to a speciﬁc aspect of the economy such as agriculture, ﬁnance, trade, or poverty. Moreover, seldom has any of these themes or aspects of the economy been treated comprehensively. A third motive, and one closely related to the other two, is the prominence of the MENA region in the global energy markets and the world political system. Countries in the MENA region are in the international headlines virtually on a daily basis. The topics of interest are wide ranging and diverse, ranging from dictatorships, civil wars, terrorism, and water shortages, to more economic problems arising from volatile oil prices, barriers to trade and foreign investment, dominance of public enterprises, and low-skill work forces. Many of these are interdependent and are collectively worthy of explanation. For all these reasons, we feel that a comprehensive treatment of the factors responsible for the relatively poor performance of the countries in the MENA region would be needed. This is especially true at this critical juncture of the economic and political development of the region and the important implications thereof for growth and growth policies in other parts of the world. A comprehensive treatment should combine micro-level as well as macro-level analyses. It should also make ample use of comparisons between MENA and the other regions, within the MENA region itself, and the two comparisons over time, especially in cases where policy and institutional orientations have changed over time. Since the MENA region is subject to a rather unique and volatile combination of common factors such as oil, religion, poor governance, and substantial

viii

migration ﬂows, the volume provides the reader with a better understanding of how these factors have interacted with initial conditions and with each other to explain variations in growth and development patterns across the MENA countries and over time. Quite naturally, there is greater focus on these rather unique factors than might appear in similar studies of the other regions. The book is meant for a wide readership and to the extent possible technical details have been deliberately kept to the bare minimum. It ought to be of interest to economists of all persuasions, and for the most part should be accessible to non-economists with special interest in the economics of the MENA region. In virtually every chapter of the book there is considerable emphasis on policies and institutions, and it is, therefore, also hoped that the book will be of interest to policy makers in the MENA and other regions. The raw material and idea for the book originated in a very ambitious project compiled by the Economic Research Forum (ERF), a network of scholars, practitioners, and policy makers interested in the MENA region. ERF has counterparts in other regions, for example, LACEA in Latin America and the Caribbean and AERC in Africa. As with the other networks, the ERF has been supported by numerous agencies, bilateral donors, regional ﬁnancial institutions, and international institutions such as the UNDP and the World Bank. The speciﬁc ERF project, which made this book possible was sponsored by the Global Development Network (GDN) which links the various regional networks mentioned above. Various researchers (both individuals and teams) were selected by the ERF specifically for the GDN-sponsored project. Different researchers were commissioned to write papers on various aspects of the development process in the region, both country-speciﬁc as well as thematic contributions relevant to the region as a whole. Each of the resulting papers was presented and discussed formally at a conference organized by the ERF. All these papers, with a few exceptions, were submitted to us to be considered for publication. In addition to these papers, we also commissioned a paper on Turkey which was not covered by the ERF–GDN initiative. All the submitted papers were then reviewed again by at least two referees and subsequently assessed by us. The authors of those papers considered for publication in this volume were then requested to revise their papers in line with the comments of the referees and editors. The resulting re-submissions were eventually refereed again and in most cases a third time. Inevitably, this meant that some papers could not be included in the volume and that the review and revision process for those included has been rather long and protracted. We wish to express our deep appreciation to all the referees, internal to the group (i.e., authors of the contributed papers) as well as external to it. The external referees included Agnes Benassy, Mongi Boughzala, Nauro Campos, Merih Celasun, Mine Cinar, Christopher Clague, Ali Darrat, Panicos Demetriades, Matthias Doepke, Alpay Filiztekin, Dipak Gupta, Hassan Hakimian, Andrew Horowitz, Hyeok Jeong, Magda Kandil, Taher Kanaan, Massoud Karshenas, Andrew Mason, Hamid Mohtadi, Mokhtar Metwally, Jennifer

ix

Olmsted, Serdar Sayan, Edward Sayre, Paul Schultz, Ajit Singh, John Strauss, T.G. Srinivasan, Erol Taymaz, Jonathan Temple, Insan Tunali, and Tarik Yousef. We are grateful to all the referees for their help on each draft, in many cases having to operate under very severe time constraints. Their comments have contributed substantially to the ﬁnal form and character of the volume. Above all, we wish to thank the authors for their patience and willingness to revise and re-submit their papers despite their own hectic schedules. We also appreciate the authors’ willingness to modify their papers so that they would ﬁt together in a more coherent whole. Finally, we would not have been able to communicate with the authors and referees and keep track of the status of each paper, without the help of the ERF Secretariat, its Directors (ﬁrst, Heba Handoussa and more recently, Samir Radwan), and our assistant Julia Murkin in Cambridge. Jeffrey Nugent and Hashem Pesaran May 2006

This page intentionally left blank

CONTENTS PREFACE

vii

LIST OF CONTRIBUTORS

xix

CHAPTER 1

1 2 3 4 5 6 7

Coverage of the book Limitations and omissions Cross-country heterogeneity in the MENA region Inter-linkages within the MENA region Growth and volatility in MENA compared to other regions Growth and volatility across MENA countries Contributed papers References

PART I:

CHAPTER 2

1 2 3

4 5

INTRODUCTION: EXPLAINING GROWTH IN THE MIDDLE EAST Jeffrey B. Nugent and M. Hashem Pesaran

1

1 4 6 8 14 18 19 26

THE MAIN THEMES IN THE MENA REGION AS A WHOLE DETERMINANTS OF GROWTH IN THE MENA COUNTRIES Samir Makdisi, Zeki Fattah and Imed Limam

Introduction Growth record of the MENA region MENA economic growth in a global context 3.1. Cross-country regression 3.2. Growth accounting in international perspective Explaining intraregional MENA growth performance Concluding remarks Acknowledgements References

31

32 33 37 37 46 53 56 57 57

xii

CHAPTER 3

1 2

3

4

Introduction A framework for analyzing the political economy of growth 2.1. Resources 2.2. Politics 2.3. Institutions 2.4. Dynamics in the political economy of growth MENA’s growth experience: the roles of resources, politics, and institutions 3.1. Natural resource rents 3.2. Heterogeneity and polarization 3.3. Geopolitical rents and threats and the role of military 3.4. Authoritarianism 3.5. Human capital 3.6. Administrative and judicial institutions Conclusion Acknowledgments References Appendix

CHAPTER 4

1 2

3

4

A REEXAMINATION OF THE POLITICAL ECONOMY OF GROWTH IN THE MENA COUNTRIES Hadi Salehi Esfahani

FINANCIAL DEVELOPMENT AND ECONOMIC GROWTH IN THE MENA COUNTRIES Hasan Ersel and Magda Kandil

Introduction Financial development and economic growth: direct contribution 2.1. Resource mobilization function 2.2. Resource allocation function 2.3. Operational efﬁciency 2.4. Economic growth and ﬁnancial development indicators Indirect contribution of ﬁnancial markets to growth: the effectiveness of economic policy 3.1. Monetary policy design 3.2. Time-series results 3.3. Policy design and the effects of monetary growth Summary and policy recommendations Acknowledgments References Appendix

61

62 66 67 68 71 75 78 78 83 86 87 93 94 95 97 97 101

103

104 105 106 109 111 113 116 119 123 125 126 131 131 133

xiii

CHAPTER 5

1 2

3

4 5

Introduction: the growth context Human capital and economic growth 2.1. Human capital deﬁnitions 2.2. Growth theory 2.3. Growth accounting Labor market structure 3.1. Static efﬁciency 3.2. Institutional foundations 3.3. Market ﬂexibility Acquisition of skills Conclusions References

CHAPTER 6

1 2

3

4

MICROECONOMICS OF GROWTH IN MENA: THE ROLE OF HOUSEHOLDS Djavad Salehi-Isfahani

Introduction 1.1. The role of urban households Economic and social environment 2.1. Natural endowments and climate 2.2. Trade shocks 2.3. Institutional environment Household decisions 3.1. Household time allocation 3.2. Human capital 3.3. Savings Conclusions Acknowledgments References

PART II: CHAPTER 7

1

LABOR MARKETS AND ECONOMIC GROWTH IN THE MENA REGION Christopher A. Pissarides and Marie Ange Ve´ganzone`sVaroudakis

137

138 139 139 140 143 147 147 151 152 153 154 156

159

160 161 164 165 165 166 173 173 181 186 188 190 190

COUNTRY-SPECIFIC STUDIES SOURCES OF ECONOMIC GROWTH AND TECHNICAL PROGRESS IN EGYPT: AN AGGREGATE PERSPECTIVE Hanaa Kheir-El-Din and Tarek Abdelfattah Moursi

Introduction

197

198

xiv

2 3

4 5

Total factor productivity estimates (1960–1998) Role of input changes and technical progress in economic growth 3.1. Role of physical inputs 3.2. Role of technical progress Relation between technical progress and selected macroeconomic indicators Conclusion: reﬂection on sources of economic growth in Egypt (1960–1998) Acknowledgments References Appendix A: Review of the methods used to estimate the APF Appendix B: Source of data

CHAPTER 8

1 2

3 4

5

6

Introduction An overview of the Turkish economy from a growth perspective 2.1. Inward-oriented strategy: 1963–1980 period 2.2. The 1980s and the 1990s 2.3. The millennium onward Comparison with the MENA countries Macroeconomic policies and the ﬁnancial crises 4.1. The 1994 currency crisis 4.2. The 2001 ﬁnancial crisis 4.3. Fiscal dynamics 4.4. Liberalization of the ﬁnancial markets Output gap and business cycle characteristics 5.1. Role of short-term capital ﬂows 5.2. Business cycle facts for Turkey Conclusion References Appendix A: The output gap estimation for the Turkish economy Appendix B: Robustness tests with the output gap measure from Dees et al. (2005)

CHAPTER 9

1 2

SOURCES OF GROWTH AND THE OUTPUT GAP FOR THE TURKISH ECONOMY Kıvılcım Metin O¨zcan, U¨mit O¨zlale and C - ag˘rı Sarıkaya

ECONOMIC GROWTH AND ECONOMIC POLICY IN IRAN: 1950–2003 Ahmad R. Jalali-Naini

Introduction Aggregate growth performance and sectoral growth patterns: 1950–2003

200 204 204 211 214 223 225 225 227 235

237

238 239 239 241 242 246 250 250 251 252 253 254 255 259 261 261 264 265

267

268 268

xv

3

4 5

6 7

Stylized facts of growth 3.1. Grand ratios and structural change 3.2. Persistence of factor accumulation and divergence in growth rates 3.3. Growth volatility and shock persistence Sources of growth: factor accumulation or productivity 4.1. HC and TFP Economic policy, strategy, and growth performance during 1950–2003 5.1. High growth years: 1959–1977 5.2. The revolution and war: 1978–1987 5.3. Post-war economic growth and policy shift: 1988–2003 Oil, rent-seeking, and economic growth Summary and conclusion Acknowledgments References

CHAPTER 10

1 2

3

4

5

6

THE POLITICAL ECONOMY OF DEVELOPMENT POLICY IN TUNISIA Mohamed Z. Bechri and Sonia Naccache

Introduction Policy episodes and the political economy approach 2.1. Policy episodes: 1956–2004 2.2. Political economy approach Establishment of a State-led economy (1956–1969) 3.1. Political setup and interest groups 3.2. Representation, coordination, and commitment 3.3. Policy choices 3.4. Rise and fall of collectivism The two-track strategy 4.1. Offshore Company Law and export promotion 4.2. Financial repression and exchange controls 4.3. Role of interest groups Economic populism and the crisis-triggered reforms of the 1980s 5.1. Political and economic setup 5.2. Crisis-triggered reforms of 1986 5.3. Post-adjustment uncertainties (1992–1994) 5.4. The association agreement with the European union Conclusion References

269 272 274 277 279 281 283 284 285 287 297 299 301 301

307

308 308 308 311 314 314 316 317 318 322 322 322 323 326 326 328 330 330 332 333

xvi

CHAPTER 11

1 2

3

4

5

Introduction Overview of Algeria’s economic growth 2.1. 1962–1973: High but unstable economic growth 2.2. 1974–1986: Macroeconomic instability, difﬁcult adjustments, and volatile economic growth 2.3. 1987–1994: Structural adjustment programs and economic recession period 2.4. 1995–2000: A new wave of economic reforms and economic growth Sources of economic growth in Algeria: a growth accounting exercise 3.1. Contribution of labor and capital to economic growth 3.2. TFP and its contribution to economic growth Structural weaknesses in the Algerian economy 4.1. Weak institutions and administrative capacity 4.2. High external debt and debt servicing 4.3. High inﬂation rate 4.4. Low labor productivity and the mismatch between the education and training system 4.5. Underdeveloped banking sector 4.6. Inadequate infrastructure Conclusion and policy options Acknowledgments References

CHAPTER 12

1 2

ALGERIA’S MACROECONOMIC PERFORMANCES FROM 1962 TO 2000 Mohamed Abdelbasset Chemingui and Moataz Mostafa El-Said

EXPLAINING GROWTH IN AN OILDEPENDENT ECONOMY: THE CASE OF THE UNITED ARAB EMIRATES Adam B. Elhiraika and Annas H. Hamed

Introduction Performance of the UAE economy 2.1. Domestic investment 2.2. Domestic savings 2.3. Export growth 2.4. Financial development 2.5. Human capital development 2.6. Macroeconomic stability

335

336 337 338 339 340 341 342 343 345 347 348 349 350 351 353 355 355 356 356

359

360 361 364 365 365 366 366 366

xvii

3

4

5

UAE growth in regional perspectives 3.1. Growth accounting framework 3.2. Econometric model of the determinants of UAE growth Markets and institutions 4.1. Labour market 4.2. Financial markets and institutions 4.3. Investment climate Conclusion and policy implications Acknowledgments References

CHAPTER 13

1 2 3

4

5

INSTITUTIONS, HOUSEHOLD DECISIONS, AND ECONOMIC GROWTH IN EGYPT Ragui Assaad

Introduction Growth implications of the demographic transition Institutional context and investments in educational human capital 3.1. Expenditures on education and school enrollment 3.2. Role of the state as a dominant employer of educated workers 3.3. Effects of the graduate employment guarantee on human capital acquisition and rates of return to schooling 3.4. ‘‘Entrapment’’ of graduates in the public sector Gender norms and barriers to entry for women in the private sector 4.1. Female participation in the Egyptian labor market in the 1990s 4.2. Barriers to private sector employment for women Conclusions Acknowledgments References

367 367 370 375 376 378 379 380 382 382

385

385 387 390 390 393 395 401 402 404 406 407 409 409

NOTES ON THE AUTHORS

413

SUBJECT INDEX

419

This page intentionally left blank

List of Contributors Ragui Assaad

Population Council, West Asia and North Africa Ofﬁce, Cairo, Egypt, and Humphrey Institute of Public Affairs, University of Minnesota, USA

Mohamed Z. Bechri

Professor of Economics, Faculty of Law and Economics, Sousse, Tunisia

Mohamed Abdelbasset Chemingui

United Nations Economic Commission for Africa Trade, Finances and Economic Development Division, P.O. Box 3001, Addis Ababa, Ethiopia

Adam B. Elhiraika

Economic and Social Policy Division, United Nations Economic Commission for Africa, POB 3005 Addis Ababa, Ethiopia

Moataz Mostafa El-Said

Fiscal Affairs Department, International Monetary Fund, 700 19th Street, N.W. Washington, DC 20431, USA

Hasan Ersel

Faculty Arts and Social Sciences, Sabanci University, Orhanli, 81474, Tuzla, Istanbul, Turkey

Hadi Salehi Esfahani

210 David Kinely Hall, 1407 W. Gregory Dr. Urbana, IL 61801, USA

Zeki Fattah

UN ESCWA, Beirut, Lebanon, Riad El Solh SQ., Beirut, Lebanon

Annas H. Hamed

Abu Dhabi Fund for Development, Abu Dhabi, UAE

Ahmad R. Jalali-Naini

The Institute for Management and Planning Studies, Tehran, Iran

Magda Kandil

International Monetary Fund, IMF Institute, HQ2, 4-899, 700 Nineteenth St., Washington, DC 20431, USA

xx

Hanaa Kheir-El-Din

Department of Economics, Faculty of Economics and Political Science, Giza, Egypt

Imed Limam

The Arab Fund for Economic and Social Development, P.O. Box 21923, Safat 13080, Kuwait

Samir Makdisi

American University of Beirut, Nicely 316, Bliss Street, Beirut 1107 2020, Lebanon

Tarek Abdelfattah Moursi

Department of Economics, Faculty of Economics and Political Science, Giza, Egypt

Sonia Naccache

Faculty of Economics and Management, Tunis, Tunisia

Jeffrey B. Nugent

USC Department of Economics, University Park Campus, University of Southern California, Los Angeles, CA 90089-0253, USA

Kıvılcım Metin O¨zcan

Department of Economics, Bilkent University, Bilkent 06800, Ankara, Turkey

U¨mit O¨zlale

TOBB University of Economics and Technology, Department of Economics, So¨g˘u¨to¨zu¨ 06560, Ankara, Turkey

M. Hashem Pesaran

Faculty of Economics, Sidgwick Avanue, Cambridge CB3 9DD, UK

Christopher A. Pissarides

Department of Economics, London School of Economics, Houghton Street, London WC2A 2AE, UK

Djavad Salehi-Isfahani

Department of Economics, Virginia Polytechnic Institute and State University, Blaksburg, VA 24061, USA

C ¸ agˇrı Sarıkaya

The Central Bank of The Republic of Turkey, Research and Monetary Policy Department, Istiklal Caddesi No 10, Ulus 06100, Ankara, Turkey

Marie Ange Ve´ganzone`sVaroudakis

CERDI, CNRS, Universite´ d’Auvergne, 65 Boulevard Francois Mitterand, 63 000 Clermont Ferrand, France

CHAPTER 1

Introduction: Explaining Growth in the Middle East Jeffrey B. Nugent and M. Hashem Pesaran Keywords: Middle East, North Africa, economic growth, oil, institutions, policy choices JEL classifications: O1, O4, O53, E00 1. Coverage of the book The Middle East and North Africa (MENA) region, in its broadest sense, is deﬁned as the 21 Arabic-speaking countries of the League of Arab States1 plus Iran and Turkey. Despite its geographical location, Israel is normally not included in the region on the grounds that it is a highly developed country and as such is usually included by the World Bank and other international agencies in Europe (along with Cyprus and Malta). While the MENA region has been the subject of a large literature on terrorism, religion, oil, governance, international political relations and economic history, compared to other regions, comprehensive studies of economic growth in the MENA region are very few and tend to focus on individual

Corresponding author. 1

The 22 Arabic-speaking members of the League of Arab States are Algeria, Bahrain, Comoros, Djibouti, Egypt, Iraq, Jordan, Kuwait, Lebanon, Libya, Mauritania, Morocco, Oman, Qatar, Palestinian Territories, Saudi Arabia, Somalia, Sudan, Syria, Tunisia, UAE and Yemen. Eritrea is an observer in the organization and hence is often included in the deﬁnition of the region. Because they are extremely small in size, have little reliable statistics, and in some cases have been subject to so much ﬁghting as to make it hard to describe them as nation states, Comoros, Djibouti, Eritrea, Somalia and to a lesser extent Mauritania are often excluded (as they are for the most part in this volume). Another reason for their exclusion is that most of these countries are often classiﬁed as being in Sub-Sahara Africa.

CONTRIBUTIONS TO ECONOMIC ANALYSIS VOLUME 278 ISSN: 0573-8555 DOI:10.1016/S0573-8555(06)78001-4

r 2007 ELSEVIER B.V. ALL RIGHTS RESERVED

2

Jeffrey B. Nugent and M. Hashem Pesaran

countries rather the region as a whole. The purpose of this book is to ﬁll this gap by providing a relatively comprehensive overview of the growth experience of the MENA region as a whole as well as that of important individual countries in the region. It covers different aspects of the growth process, such as growth of different sectors, population growth and patterns of demographic change, integration of the MENA countries within the global economy and it addresses the different sources of growth, i.e., the accumulation of factors (labor and capital), technological change and institutional factors. Particular attention will also be paid to the cross-country comparisons of growth rates and their volatilities and the various factors that might be responsible for their evolution. The countries in the MENA region share a number of common features, but also differ in a number of important respects. They vary in geographic size, population density, endowment of cultivable land, access to international markets, oil and other natural resources, income and wealth, ethnic, religious and language diversity, vulnerability to civil and international wars and other conﬂicts, governance types and qualities and policy orientations. Despite these differences, the MENA countries, with varying degrees, have been subject to common effects of the Arab–Israeli conﬂicts, the Iran–Iraq war, the Gulf war, the recent invasion of Iraq, the volatility of oil prices, the rise of Islamic fundamentalism and the geopolitical considerations that are largely governed by the desire of the industrialized countries (and more recently, the emerging market economies) to have access to the Middle East oil and gas resources. Many of these factors are addressed in this volume, with different authors discussing the subset of factors deemed to be the most relevant to the economies or issues that they investigate. In addition to the traditional proximate drivers of economic growth, namely accumulation of factor input and technological progress, the papers in this volume also emphasize some of the fundamental determinants of growth such as geography, institutions, ethnicity and cultural heritage as important primary factors in the growth process. The importance of geography, institutions and ethnicity as root causes of observed persistent differences in per capita output across countries have been recently emphasized in the work of Gallup and Saks (2000), Acemoglu et al. (2001), Easterly and Levine (2002) and many others. The individual country studies conducted in this volume complement well the recent evidence on the fundamental determinants of cross-country differences in log per capita output obtained using static cross-section regressions. Such regression results, even if estimated by instrumental variables can be subject to endogeneity problems and conclusions based on them should be treated with care. In studying the MENA countries a number of features speciﬁc to the region need to be borne in mind. Here we focus on three of them: oil, demography and institutions. At the end of 2004, it was estimated that 61.7% of the world’s proven oil reserves and 40.6% of world natural gas reserves were located in the Middle East

Introduction

3

region.2 In the same year the Middle East accounted for 30.7% of world oil production and only 6.7% of world oil consumption. The production share of the Middle East area would have been even much higher if it were not for the conﬂicts and political instabilities that have beset the region over the past century. The large reserves of the Middle East oil combined with the fact that they are relatively cheap to produce means that in a politically neutral world the exploitation of Middle East oil and gas ﬁelds would have been given priority over the less-productive oil and gas ﬁelds elsewhere.3 The relatively large investments committed to the exploitation of oil and gas ﬁelds outside the MENA region largely reﬂect the understandable desire by oil-consuming countries to diversify the sources of their energy supplies in the face of mounting uncertainties in the region. However, the relatively cheap and abundant oil and gas reserves of the region will continue to be a source of disequilibrium and induce economic and political intervention by major oil-consuming countries in the region. This in turn could lead to further conﬂicts and political uncertainties, setting in motion a vicious circle of conﬂicts and uncertainties from which neither side (producers and consumers) could beneﬁt. On the domestic front, revenues from oil and gas exports account for a large part of government revenues and a substantial part of foreign exchange earnings in some of the major MENA countries, notably, Saudi Arabia, Iran and Iraq. Since these revenues accrue wholly to the government, there are domestic political ramiﬁcations emanating from rent seeking, non-accountability of government agencies and sluggish policy responses to a volatile and fast-changing international oil prices.4 These factors tend to create further impediments to the development of competitive market-oriented institutions. Another acute and pervasive problem in the MENA region is the extremely rapid growth of its labor force. This is often referred to as a ‘‘demographic gift’’, made possible by a fairly lengthy period of very high fertility rates combined with declining mortality rates, followed by sharply declining fertility rates. Like oil, this demographic gift can be a blessing as well as a curse. As with oil, rather than beneﬁting from this situation, the MENA countries have instead been struggling with how to provide their rapidly growing labor forces with productive employment. Institutional factors and policy mismanagement seem to have been the root cause of the problem. The economic incentives in many of the countries in the region have tended to accentuate labor market rigidities, with

2

Source: British Petroleum Statistical Review of World Energy, June 2005. The theory of exhaustible resource extraction, pioneered by Hotelling (1931), is concerned with the intertemporal optimization of the rate of resource extraction and does not address the issue of relative production rates of oil ﬁelds with different reserve endowments and different marginal extraction costs that is of concern here. 4 See Pesaran (1982) for a discussion of the interaction between oil revenues, dictatorship and dependent capitalism in the case of Iran under the Shah. 3

4

Jeffrey B. Nugent and M. Hashem Pesaran

government and state enterprises often providing remuneration and beneﬁts that are high relative to those employed in the private sector. Institutional problems have also hampered the development of capital and ﬁnancial markets, and are most likely partly responsible for the low labor force participation rate of women in the region. Once again, these problems are confounded by cultural and institutional factors. While these institutional problems are by no means unique to the MENA region, they tend to be more formidable and persistent in this region than elsewhere. Several of the authors of this volume delve into political economy to explain why countries of the MENA region have tended to lag behind other regions in carrying out the kinds of policy and institutional reforms that are generally believed to mitigate these problems. This is done for the region as a whole, for oil and non-oil countries and for several individual countries (in particular, Tunisia, Turkey, Algeria and Egypt). 2. Limitations and omissions Despite its wide coverage, largely owing to data limitations, the present volume focuses on the development experience of some of the main countries in the region post 1950s.5 The book, being primarily an overview of growth at the macroeconomic level, does not go into individual sectors nor does it attempt to provide a multisectoral account of the calibrated computable general equilibrium (CGE) variety.6 For the same reason, neither does it go into detail about non-oil minerals, agriculture or animal husbandry, even though some agricultural crops such as cotton, tomatoes, qat and fruits or phosphates in the case of minerals are important to some individual countries of the region.7 The related issues of property rights in land and water resources are also largely left out.8 Trade and foreign direct investments are covered, but only in relation to the macroeconomic developments of the region and the individual countries.9 There are also some gaps in terms of the geographic coverage of the book. As noted earlier, the MENA region consists of 21 Arabic-speaking countries of the MENA as well as Iran and Turkey. But for several of the 21 Arabic-speaking countries, the necessary statistical information is largely lacking. Table 1

5

For excellent historical accounts of important parts of the MENA region, the reader is referred to Owen (1981), Issawi (1982) and Owen and Pamuk (1999). 6 For some examples of multisectoral CGE models of the MENA countries see Maskus and Konan (1997) and Konan (2003). 7 For agriculture in general and speciﬁc crops of the region in particular see Lofgren (2003). 8 For some interesting accounts of property rights issues in the MENA region see de Soto (2000); Nugent and Sanchez (1999) for Sudan and Nugent and Sanchez (1993, 1998) for much of the rest of the MENA region. 9 For studies on trade, trade issues and trade policies in the MENA region, see Galal and Hoekman (2003); Hoekman and Zarrouk (2000); Miniesy, et al. (2004).

Introduction

5

Table 1. Selected statistics for the countries in the MENA region Country

Algeria Bahrain Djibouti Egypt Eritrea Iran Iraq Israel Jordan Kuwait Lebanon Libya Mauritania Morocco Oman Palestinian Territories Qatar Saudi Arabia Somalia Sudan Syria Tunisia Turkey UAE Yemen MENA

Population (millions)

Share of Population in MENA (%)

32.4 0.725 0.716 68.7 4.5 66.9 25.3 6.8 5.4 2.5 4.6 5.7 2.9 30.6 2.7 3.5

7.22 0.16 0.16 15.30 1.00 14.90 5.64

0.637 23.2 9.9 34.4 17.8 10 71.7 4.4 19.8 448.978

GDP 2004 (Current US Million Dollars)

Share of GDP in MENA (%)

GDP per Capita 2004 (Current US Dollars)

6.20 0.84 0.05 6.81 0.06 12.68 2.05

1.20 0.56 1.02 1.27 0.65 6.82 0.60 0.78

80,783.09 10,975.12 663.80 88,783.56 793.21 165,200.80 26724.14 123,109.11 10813.60 51,805.23 19,946.00 19,535.86 1238.95 49,814.13 24,465.55 5600.00

0.83 3.98 1.53 1.50 0.10 3.82 1.88 0.43

2497 15,332 852 1222 187 2401 952 18,651 1945 19,876 5634 3403 416 1589 9656 1600

0.14 5.17 2.21 7.66 3.96 2.23 15.97 0.98 4.41 100.00

28,451.37 24,4339.41 2088.01 19,948.94 23,440.20 28,134.48 301,999.46 84,226.68 13,079.57 1,302,851.15

2.18 18.75 0.16 1.53 1.80 2.16 23.18 6.46 1.00 100.00

36,620 10,202 262 562 1261 2815 4182 19,659 643 143,768

Source: World Bank, World Development Indicators (2005) CDROM.

provides some basic data on all the 23 countries, plus Israel for the purpose of comparison. But, because of data limitations, several of these countries, Comoros, Djibouti, Eritrea, Iraq, Palestinian territories, Somalia and sometimes even Libya and Yemen, will seldom be treated in the rest of the volume. Given their small sizes and peripheral locations, for the most part these omissions are not serious. For the Palestinian territories, even though central to the region and the political conﬂicts within the region, since they are of course not yet a country and hence not generally covered in international data sources, their omission is perhaps understandable. The omission of Iraq, however, throughout much of the analysis presented in this volume is a more serious shortcoming, especially since it is the fourth largest economy of the region (in population). Were it not for the three wars that it has fought (and lost) in the last 20 years, Iraq would most likely have been among

6

Jeffrey B. Nugent and M. Hashem Pesaran

the top two or three MENA countries in aggregate GDP in 2004.10 Since virtually all estimates of the trajectory of Iraq’s GDP over time (even if primitive and therefore not usually included by the more reliable sources) indicate that its income per capita has declined quite substantially over time, its exclusion from MENA aggregates probably understates MENA’s growth prior to 1975 (when Iraq was growing rapidly) but overstates MENA’s growth since then. The omission of Iraq also deprives us from taking advantage of some of the political economy lessons of its experience. 3. Cross-country heterogeneity in the MENA region As Table 1 shows there are important differences across the countries in the MENA region. While Turkey, Egypt and Iran all had populations of at least 66 million in 2004, four MENA countries (Bahrain, Comoros, Djibouti, and Qatar) had populations under 1 million. Similarly, Gross National Income (GNI) per capita at dollar purchasing power parity (PPP) prices of 2004 varied widely from a low of $820 for Yemen to about $22,000 for Qatar. It is important to point out to the reader various other ways in which these countries differ. For example, many of these countries are in Africa (including Comoros which is in the Indian Ocean far south of other countries of the region), but others are in Western Asia. Some (Bahrain, Comoros, Djibouti, Qatar and even Lebanon and the Palestinian territories) are geographically rather tiny whereas Saudi Arabia covers a sizable portion of Western Asia and Algeria and Sudan are the largest countries in Africa. Comoros and Bahrain are islands whereas Jordan (except for its port of Aqaba on the Red Sea) is virtually landlocked. While manufacturing is well developed in Algeria (especially heavy industry), Egypt, Iran, Morocco, Tunisia and Turkey, there is very little manufacturing in Comoros, Djibouti, Eritrea, Jordan, Libya, Mauritania, Sudan and Yemen. Agriculture is relatively important in Algeria, Egypt, Iran, Iraq, Lebanon, Morocco, Sudan, Syria, Turkey and Yemen, while it hardly exists in many of the other countries. In several countries of the region, such as Algeria, Eritrea, Iran, Jordan, Mauritania, Morocco, Somalia, Sudan and Yemen, animal husbandry is relatively important. Jordan, Mauritania, Morocco, Oman, Saudi Arabia, Tunisia and Turkey have important mineral deposits that are exploited and exported whereas mining is relatively unimportant in most of the other countries of the region. Algeria, Bahrain (until recently), Egypt, Iran, Iraq, Kuwait, Libya, Qatar, Saudi Arabia, Syria, the United Arab Emirates (UAE) and more recently, Sudan and Yemen are exporters of oil and in many cases also of natural gas. Virtually, all the other countries of the region, however, do not possess such resources and are importers of oil and natural gas.

10

For historical as well as contemporary analyses of Iraq, see Issawi (1992) and Alnasrawi (1994).

Introduction

7

In terms of governmental types, many of the countries are or were until recently monarchies. The ones which are still monarchies are Bahrain, Jordan, Kuwait, Morocco, Oman, Qatar, Saudi Arabia and the UAE. Several of these (Bahrain, Jordan, Kuwait and Morocco) have legislatures that are rather democratically elected. Iran is an Islamic state (with some democratic trimmings). Saudi Arabia is also an Islamic kingdom, but without democratic trimmings. Algeria, Libya, Sudan, Syria, Yemen and, until the overthrow of Saddam Hussein, also Iraq were classiﬁed by Henry and Springborg (2001) as ‘‘bunker states’’ ruled by single dictators backed by the military. Fairly close to these are countries such as Egypt, Tunisia and until recently, the Palestinian territories labeled by Henry and Springborg as ‘‘bully praetorian states’’, in which the leadership was again largely dictatorial. Their distinction between bunker and bully praetorian states is that in the latter type the private sector is allowed to have a larger role in the economy than in the bunker states. At the other end of the spectrum are Lebanon and Turkey that are classiﬁed as secular and ‘‘fragmented democracies’’. While secular and having many characteristics of true democracies, these countries have state institutions that show them fragmented along ethnic, religious and regional lines. The countries in the region also differ by their respective degrees of centralization (vis-a`-vis decentralization). Not surprisingly, the small states tend to have a highly centralized governmental system, while the large ones are somewhat decentralized. But even among the large ones, Algeria, Egypt, Saudi Arabia and Turkey are all quite centralized, whereas Sudan is not. Among the small states, the UAE has a relatively decentralized governmental structure in the sense that it is a federation of seven emirates (Abu Dhabi, Ajman, Al Fujayrah, Dubai, Ra’s-al Khaymeh, Sharjah and Umm al Quwayn) with many of the functions of the government reserved for decision-making at the individual emirate level. The MENA countries differ considerably in their degree of openness to trade, as well. Algeria, Egypt, Lebanon, Morocco, Sudan, Syria and Tunisia all had tariff rates (on a trade-weighted basis) averaging over 13% in 1997, with Syria’s being almost 30%. In several of the Gulf States, however, tariff rates are very low and there are virtually no other barriers to trade (other than on goods from Israel). MENA countries have also differed in their proclivity for establishing free trade arrangements with other countries. Tunisia was the ﬁrst MENA country (in 1996) to sign a free trade agreement (FTA) with the European Union, to be followed by Morocco, Jordan, Egypt, Turkey and Lebanon. Jordan and Morocco have also signed FTAs with the United States, and Turkey is moving toward becoming a full-ﬂedged member of the European Union. But other countries like Algeria, Iran, Libya, Syria and Yemen have signed no such agreements. With respect to intra-MENA trade arrangements, several MENA countries have signed agreements for partial free trade among themselves in the past. Notable examples were Turkey and Iran (with Pakistan as well) as part of the

8

Jeffrey B. Nugent and M. Hashem Pesaran

Regional Cooperation for Development scheme; the Maghreb countries (Algeria, Libya, Mauritania, Morocco and Tunisia) which signed onto the Arab Maghreb Union; Egypt, Syria, Iraq, Jordan (and later Libya, Mauritania and Yemen) which joined the Council of Arab Economic Unity and the Arab Common Market, which this Council fostered. Yet, many of these regional trade agreements have not been fully or consistently implemented. In contrast, the six Arab countries of the Gulf (Bahrain, Kuwait, Oman, Qatar, Saudi Arabia and UAE) have achieved and adhered to rather strictly a partial free trade area among each other as part of the Gulf Cooperation Council. More recently, in 2004 four of the ‘‘Mediterranean’’ MENA countries that signed FTAs with the EU (Egypt, Jordan, Morocco and Tunisia) have signed a protocol, the so-called ‘‘Agadir Agreement’’, calling for free trade among each other in stages on a very short timetable. Another feature with respect to which MENA countries differ signiﬁcantly is the relative importance of private and public enterprises in their economies. Public enterprise accounts for around 50% of GDP in Algeria and Sudan, over 65% of gross domestic investment in Egypt, and over 10% of total bank credit in Algeria, Egypt, Saudi Arabia, Sudan and Tunisia. Yet, countries like Bahrain, Lebanon and to a lesser extent Jordan and Morocco are very strongly privateenterprise oriented, and their private sectors account for over 70% of GDP (World Bank, 2005; Henry and Springborg, 2001). While privatization has been pledged in many countries of the region, only in a few (Egypt, Morocco, Turkey and more recently, Iran and Jordan) has the privatization of public enterprises actually raised signiﬁcant amounts of revenue for the government. MENA countries also vary tremendously in their scores on numerous institutional indexes. Several of these indexes are utilized in the subsequent chapters. Here we conﬁne our attention to those presented in the World Economic Forum’s World Competitiveness Reports. These reports have ranked countries with respect to technological infrastructure, the quality of public institutions and the macroeconomic environment, all of which are deemed to be important for business competitiveness and economic growth. In general, the MENA countries do not compare favorably to other regions in these respects. The larger countries of the region all score below the average of the 102 countries rated in the index. Table 2 provides the rankings on all three components of this index among the 12 MENA countries scored. As can be seen, Qatar, the UAE, Bahrain, Oman and Jordan – all small countries – consistently rank at the top of this list of 12 MENA countries in these respects, with Lebanon and Yemen at the bottom. 4. Inter-linkages within the MENA region Given the extent of heterogeneities across the MENA countries outlined above, one might be tempted to challenge the usefulness of analyzing these countries as a single group. Apart from the obvious common factors, language, religion and

Introduction

Table 2. Country

Qatar UAE Bahrain Oman Jordan Tunisia Saudi Arabia Morocco Egypt Algeria Lebanon Yemen

9

Selected MENA countries ranked by competitiveness indices Overall Rank

1 2 3 4 5 6 7 8 9 10 11 12

Subcomponents Technology Index

Public Institutions Index

Macroeconomic Environment Index

2 1 3 9 4 5 6 10 7 12 8 11

1 3 4 2 5 6 8 7 10 9 11 12

1 2 4 3 7 6 5 9 10 8 12 11

Source: World Economic Forum, The Arab World Competitiveness Report, 2005, New York and Oxford: Oxford University Press.

cultural heritage, some of the heterogeneities discussed above, especially those with respect to relative factor endowments, do in fact contribute to a certain degree of complementarity across these countries that makes them more interdependent than might be supposed, given their obvious differences. There are four principal ways in which the MENA countries are interlinked. The ﬁrst is through trade in commodities and services. Some countries like Lebanon and Syria have come to specialize in certain agricultural products that are exported to other countries such as the Gulf States. For example, oil booms by increasing incomes in the Gulf States raise the demand and price for agricultural exports and spread growth across the countries in the region. The resulting higher incomes of these countries in contrast increase the demand for fuel and fertilizer products from the Gulf. Turkey and Egypt produce some textiles that are in demand in other countries of the region. Lebanon, Bahrain and more recently, UAE have come to specialize in ﬁnancial services that are demanded by those in other less-ﬁnancially developed countries. To some extent these trade ﬂows have been encouraged by bilateral and multilateral trade arrangements within the region, though the latter have thus far generally not been fully adhered to and enforced. The second, and perhaps the most visible, source of interdependence has been labor ﬂows across the countries of the region. This has been spurred by the differences in labor endowments, with Egypt and several North African and Horn of Africa countries being characterized by large endowments of labor relative to land or other natural resources, whereas countries such as Libya, and the GCC countries being characterized by dearth of population and labor force,

Jeffrey B. Nugent and M. Hashem Pesaran

10

Table 3.

Inward and outward flows of remittances across the MENA countries

Part A: inward remittances as percentage of GDPa Country 1970 s 1980 s Algeria Djibouti Egypt Iran Jordan Lebanon Morocco Oman Palestinian Territories Syria Tunisia Turkey Yemen

3.1 – 6.9 – 12.8 – 6.1 1.0 – – 3.5 3.1 –

1.1 – 9.9 – 19.3 – 7.0 0.8 – – 4.2 2.8 –

1990–1994

1995–1999

2000–2003

2.2 3.5 11.0 2.1 15.0 38.6 7.2 0.3 24.4 6.9 3.8 2.0 26.2

2.3 2.0 4.6 0.7 22.8 9.7 6.0 0.2 19.8 6.1 4.1 2.3 19.7

1.6 – 3.2 0.3 22.4 12.7 8.1 0.1 13.6 3.6 4.8 1.7 13.8

Part B: outward remittances as percentage of total importsb Country 1975 1990 1997

2000

2001

Bahrain Kuwait Oman Qatar Saudi Arabia UAE

15.4 26.8 14.5 18.8 51.2 10.3

8.9 35.1 16.7 20.0 50.3 8.3

6.4 11.6 27.2 13.2 26.4

21.6 19.7 30.5 66.8 46.9 21.9

7.7 17.1 12.0 15.2 53.0 10.2

a

Source: World Bank, World Development Indicators, 2005, CDROM. Askari, Nowshirvani and Jaber (1997, p. 67). The 1997–2001 ﬁgures are calculated from remittance data of Gulf News Online quoted in Alexander’s Gas & Oil Connections and The GCC Economic Bulletin, 2002. Figures for 2001 are calculated from World Development Indicators and Gulf Research Center, 2004. The percentages calculated using remittance ﬁgures from the Gulf Research Center for Saudi Arabia (39.7) and Kuwait (27.0) differ rather signiﬁcantly from those calculated using Gulf News Online and The GCC Economic Bulletin, 2002. For the year 2002. b

especially considering the aforementioned low female labor force participation rates in these countries. To the extent that Arabic language or common cultural values were of special use for employment in the labor-scarce countries of the region, this made it possible for workers from the labor-surplus countries of the region to ﬁnd employment in the labor-scarce countries of the region. From these labor ﬂows come remittances back to the countries of origin of these workers. Table 3 provides data on both inward and outward ﬂows of remittances as shares of the GDP and imports, respectively. Although the availability of such data is limited across countries and time, it is clear that virtually all countries in the region are quite important either as sources or recipients of migrant labor. Although, the data are insufﬁcient to show this in these tables, it is well

Introduction

11

recognized that because of differences in the quality or skills of labor supplied and demanded, the same country (in some cases) can be both an important source country and an important destination country at the same time. For example, Lebanon and Jordan import low-cost unskilled labor from Syria and Egypt, but export skilled labor to the Gulf countries. Note from Panel A of Table 3 that inward remittances have constituted over 20% of the GDP in some years for Jordan, Lebanon, the Palestinian territories and Yemen. Similarly, on the outﬂow side (Part B of the Table), even for the few selected years shown, there has been at least one year for which the remittance outﬂows were at least 20% of the total import bill in each of the six GCC countries. In the case of Saudi Arabia, in all three of the most recent years shown, then were over 50%. Especially, in the earlier years sizable portions of these remittances from GCC countries were from workers from other MENA countries back to their countries of origin (especially, Egypt, Lebanon, Syria and Yemen). The same would have been true for Libya in the past decades, in this case, the workers would have been largely from Egypt. In the case of the Maghreb countries (Algeria, Morocco and Tunisia) and also Iran and Turkey, the remittances received were largely from Europe, especially France and Germany. In each case, the magnitude of these remittances relative to the GDP of the recipient country tended to induce a positive association between the incomes of the source and destination countries. Adams and Page (2003) showed that both the stock of migrants abroad (measured as the percentage of the population of the country of origin living abroad) and remittances received as a share of the GDP have the effect of signiﬁcantly reducing the poverty rates (irrespective of how it is measured) in the country of origin. Hence, this would suggest that these migration ﬂows have been beneﬁcial to the source as well as the destination countries. The size of remittances in the MENA region tended to be larger than those in many other regions with similar migration patterns as in Latin America, particularly Mexico. There are, however, three possibly important qualiﬁcations to the alleged highly beneﬁcial effects of the sizable migration ﬂows in the MENA region. First, to the extent that the out-migrants represent ‘‘brain drain’’, i.e. the loss of skilled workers and entrepreneurial talent to the country of origin, this could represent a social loss to the countries of origin. This would be especially true where the country of origin has invested in nourishing, educating and training the workers, the productive use of whose skills will therefore be lost to the economy.11 Second, it is frequently suggested that the remittances are mainly used for consumption purposes and therefore do not do much for the

11

While this condition would seem to be generally satisﬁed, the out-migration of skilled workers may raise the wage rates of such workers in the country of origin and hence stimulate investments in human capital by numerous others. The migrants may also learn skills in the destination countries that they might subsequently bring back with them to their countries of origin.

12

Jeffrey B. Nugent and M. Hashem Pesaran

development of the country.12 Third, there is a possible political economy effect wherein the remittance income can help balance the current account and keep income at tolerable levels so as to weaken the pressure for welfare and growthincreasing reforms. Another important characteristic of Table 3 is that the entries in some cases reﬂect distinct trends over time as well as considerable variability from one period to another. Remittances and out-migration seem to be declining over time in the case of Iran (where much of the out-migration has been of highly skilled people to Europe and North America) and Turkey. Foreign labour force as the share of total labour supply seems to be declining in several of the GCC countries, and as a result there have been associated declines in outward remittances and employment opportunities for workers in the other MENA countries. The share of Arab workers in the foreign labor force of Kuwait and probably all GCC countries has declined substantially over time (Adams and Page, 2003). Given also the increasing resistance in Europe to migrants from the Maghreb, it is clear that out-migration is less and less reliable and feasible as a means of alleviating high unemployment rates and low remuneration at home. The volatility in the inward remittances relative to GDP is also quite signiﬁcant indicating that remittance income can serve as an important source of income volatility in the recipient country. Particularly vulnerable in this respect have been Palestinian workers to variations in the incidence of ‘‘border and crossing closures’’ by Israel and the changing political relations between Palestinian and host-country governments on the other hand. Some of these ﬂuctuations can also be attributed to the changing demands and requirements of the recipients. Witness the very high percentages of inward remittances during difﬁcult times, e.g. in Algeria in the early 1970s and much of the 1990s, and Lebanon as well as Yemen during their civil wars. The third and considerably less well-known means of linking countries of the MENA region is the capital ﬂows, especially in the form of Foreign Direct Investment (FDI). The extent of FDI is typically measured by the Inward FDI Performance Index (IFPI) deﬁned as the share of a country’s inward FDI ﬂow in its GDP relative to world FDI as a proportion of world GDP.13 Table 4 presents estimates of IFPI. If this ratio is 1 it indicates that it receives a share of FDI exactly in proportion to its share in world GDP. A ﬁgure below 1 indicates that the country is relatively an underachiever in attracting FDI whereas one above 1 may be interpreted as indicating overachievement in the same sense. Unfortunately, while such measures of FDI inﬂows can be obtained from individual OECD countries in many cases, information on ﬂows from other MENA

12

Such effects, however, should be balanced by the fact that the very act of migration represents saving and investment, so that the remittances, even if taking the form of consumption, represent the return on these prior investments. 13 IFPI is reported by UNCTAD in its World Investment Reports.

Introduction

Table 4. Country Algeria Bahrain Djibouti Egypt Eritrea Iran Iraq Israel Jordan Kuwait Lebanon Libya Mauritania Morocco Oman Palestinian Territories Qatar Saudi Arabia Somalia Sudan Syria Tunisia Turkey UAE Yemen MENA

13

Inward FDI performance index (1988–2004) Average 1988–1996

Average 1997–2004

0.11 8.02 n/a 1.82

0.45 2.74 n/a 0.43

0.01

0.05

0.74 0.34 0.14 0.25 0.16

1.01 1.51 0.05 0.57 0.07

1.36 1.05 n/a 1.05 0.25

1.10 0.19 n/a 1.06 0.26

0.18 0.82 2.48 0.51 0.40 4.63 1.28

1.57 0.68 1.09 0.26 0.12 0.52 0.66

Note: Inward FDI Performance Index is calculated as (FDI inﬂows in the country/world FDI) divided by (country’s GDP/world GDP). The index is calculated as the moving average over a 3 yearperiod to overcome high ﬂuctuations. Source: World Investment Report Series, UNCTAD.

countries are unfortunately not available.14 Since such data are quite volatile from year to year, meaningful comparisons must be based on averages for several years. While the overall MENA average reported for 1997–2004 is below 1, it is above 1 if one considers the sample period 1988–1996. But since it is the small countries like Bahrain, Jordan, Qatar and Tunisia that have the larger IFPI measures, and the large countries like Turkey, Egypt (in the second period at least) and Algeria that have the smaller measures, weighted averages of the index for MENA as a whole would be below 1 in both periods. Egypt in the ﬁrst period in which most of its privatization and other reforms were concentrated

14

In part, this is because such investments are smaller and less subject to formal regulation than in the cases of investments from outside the region.

14

Jeffrey B. Nugent and M. Hashem Pesaran

was an exception. The periods in which Sudan and Yemen had IFPI measures>1 were those in which these countries were opening up their oil sectors to foreign companies. Anecdotal and incomplete information is available on FDI ﬂows into MENA countries from the other FDI countries, most often from the GCC countries. Virtually, all the periods of high and rising oil prices (1973–1974, 1979–1981 and 2002–2006) have witnessed surging ﬂows of FDI from oil-exporting countries of the region to the other countries (including the MENA countries). While much of the earlier FDI ﬂows were by ofﬁcial organizations and joint ventures with a large public sector component, increasingly, the sources of these ﬂows are private. Indeed, political trends in the region have tended to favor this as MENA tourists as well as foreign investors feel more comfortable and secure in investing in the other MENA countries. Morocco, Lebanon, Jordan and Tunisia, for example, have long been popular destinations for investments from the GCC countries. Some stock markets in the region, while closed to other foreign investors, are open to investors from the other MENA countries. The fourth type of interlinkages across the MENA countries is through political relations, bilateral as well as multilateral. Among these are organizations, such as the Arab League, the Organization of Islamic States, the regional bodies of the United Nations such as Economic and Social Commission for Western Asia (ESCWA) and the United Nations Economic Commission for Africa, GCC, Maghreb and other subregional groupings as well as through various specialized agencies. When political relations are good between any two or more countries, one or more of these linking mechanisms is likely to be in place and working quite well. When economic complementarity makes such ﬂows desirable, trade, labor and /or capital ﬂows are all likely to be apparent. Yet when political relations deteriorate, any and all of these ﬂows may be vulnerable to abrupt political interference. Examples have included border closures (as between Syria and Jordan, Algeria and Morocco, Jordan and the Palestinian territories, Kuwait and Iraq), abandoning of trade or investment treaties and expulsions of foreign workers by the host country (e.g. of Egyptian workers by Libya, or Palestinians by Kuwait, Yemenis by Saudi Arabia). These sudden changes in political relationships can therefore become an important source of income volatility in the region. 5. Growth and volatility in MENA compared to other regions While it is generally conceded that the MENA region enjoyed relatively rapid growth over the period 1950–1980, a period in which massive pools of oil were being found and exploited in the region, even then it was hardly a match for the East Asia region. Over the last two decades, however, especially during 1980–1994, it is generally believed that the MENA region experienced relative stagnation. For example, Elbadawi (2005) reported MENA’s per capita GDP growth rate to have been about 1% per annum over this period compared to a

Introduction

15

very healthy 5.2% per annum in East Asia, and lower than in many other regions. Another stylized fact with which the MENA countries have become associated is that their growth rates are very volatile. Elbadawi (2005) employs a relative volatility measure, deﬁned by the standard deviation of growth rates divided by the absolute mean growth rate (assuming that it is not 0!), and reports this measure to be 3.9 for the MENA region over the period 1980–1994, compared to 1.8 for Sub-Saharan Africa (SSA) and a mere 0.2 for East Asia. MENA’s growth experience has been especially disappointing considering its potential. First, this is a region with immense natural resources, especially oil and gas, iron, bauxite, copper, potash and several other minerals. Second, beginning in the 1990s at least, the MENA region as a whole began enjoying a ‘‘demographic gift’’ because of declining birth rates, with the previously high share of population below 15 beginning to reach working age. Especially, according to the well-known and respected life cycle theory, such a demographic gift is usually believed to lower dependency rates, allowing households to have more earners per capita and hence generate higher saving and investment rates. Third, the poorest countries of the region (especially Egypt, Yemen, Morocco, the Palestinian territories and Jordan) continued to receive very sizable inﬂows of foreign assistance, and others such as Turkey, Lebanon, Egypt, Algeria, Morocco, Jordan, the Palestinian territories and Yemen continued to receive substantial remittances from their workers abroad. Essentially, despite the demographic gift, ofﬁcial capital inﬂows and remittances and government spending induced by oil revenues, the rise in MENA’s labour force continued to outstrip employment opportunities created in the region. Providing explanations for this seemingly paradoxical and rather disappointing economic performance of the MENA countries is one of the main objectives of this book. The search for explanations quite naturally calls attention to the region’s rather special features: the relative importance of oil, the rather limited prospects for agriculture, its special demographic features, its political institutions, the dominance of Islam among the religions of the region, religious and sometimes ethnic differences within countries, the relatively high share of public enterprises in economic activity, the conspicuous weakness of governance institutions, the relatively important role of both immigration and emigration and the fact that it has generally been considered a laggard in terms of economic reforms relative to the other regions. Many of these issues have been touched upon earlier and will be covered in some detail in the chapters that follow. But before going into the detail of the various explanations and the individual country experiences, a comparative summary account might be helpful. In view of the fact that country and regional coverage used in making such regional comparisons have varied considerably from one study to another and on the basis of different data sources, we deem it important to reexamine the comparisons of MENA’s growth performance with other regions over different periods using a comparable and reasonably reliable data source. To this we make use of the Penn World Tables (PWT 6.1), that is

16

Jeffrey B. Nugent and M. Hashem Pesaran

generally regarded as the most reliable, even if not quite as complete as the other sources, output data base. The PWT source is increasingly accepted as the best for making time series as well as cross-country comparisons because of the care devoted to consistency in estimating the price deﬂators and PPP-adjusted GDP series. While many studies divide countries into different groups on the basis of income levels, such as highly developed, middle-and-lower income countries, such classiﬁcations have the disadvantage of making the resultant country groups dependent on historically observed long-term growth performances, and thus can lead to biased conclusions. This is problematic especially when one is interested in making comparisons across countries and country groupings. Insofar as possible, our comparisons are geographic. We distinguish ﬁve main country groups: Europe, America, Asia and Oceania, MENA and SSA. America is in turn divided into North America (Canada, United States and Mexico) and Central and South America including the Caribbean. The Asia and Oceania region is in turn divided into Oceania plus Japan and the rest of Asia (including South, East and Southeast Asia, but excluding the MENA countries of Western Asia). For reasons of data comparability and availability, the countries of Central and Eastern Europe are excluded from the analysis since comparable data on these countries did not exist before 1990. For each of these regions, the average growth rates in GDP per capita, standard deviation of these growth rates and the ratio of the standard deviation to the absolute mean of the growth rates over the period 1960–2000 and subperiods 1960–1980 and 1981–2000 are calculated based on PWT 6.1. As noted above, the ratio of the standard deviation divided by the mean of growth rates is the commonly used measure of growth rate volatility and will be adopted here. Table 5 presents the results for each of the regional and subregional categories identiﬁed above. The results for the individual MENA countries are given in Table 6. For details of other country groupings and an analysis of output and growth convergence in the various regions, including the MENA region, see Pesaran (2006). Because of the higher standard for data comparability implicit in our use of the PWT 6.1 data source, the country coverage is not as complete in this table as in some other data sources. For example, the calculations for the MENA region are based on only 8 of the 23 countries listed above as MENA countries. Yet, with one exception we have no reason to believe that these eight countries are not broadly representative of the MENA countries as a whole. The one exception concerns the omission of oil-exporting economies of Bahrain, Djibouti, Iraq, Kuwait, Libya, Saudi Arabia and the UAE. Three of these countries – Iraq, Kuwait and Libya experienced signiﬁcant losses owing to war and/or international sanctions. The other three – Bahrain, Saudi Arabia and United Arab Emirates, are countries that in fact have enjoyed extremely rapid GDP growth (averaging well over 8% per year), but more than offset by extremely high fertility rates and exploding numbers of immigrants attracted by

Introduction

17

Table 5. Mean and volatility of output growth rates across different regions Country Groups

1961–1980

Europe America North South Asia and Australia Advanced Other MENA Africa World

1981–2000

1961–2000

Mean

Volatility

Mean

Volatility

Mean

Volatility

3.58 2.62 3.06 2.55 3.12 3.30 3.08 3.47 1.79 2.64

2.94 3.99 2.21 4.26 3.48 3.34 3.51 7.96 7.98 5.45

2.26 0.65 1.62 0.51 3.33 1.93 3.68 1.62 0.55 1.39

2.31 4.19 2.96 4.38 3.07 2.19 3.28 4.30 7.10 6.15

2.91 1.64 2.34 1.53 3.23 2.61 3.38 2.54 1.17 2.02

2.82 4.30 2.72 4.53 3.50 3.02 3.62 6.54 7.90 6.36

Source: Computations are based on Penn World Tables (PWT 6.1) with country groups deﬁned in Table 1 of Pesaran (2006). Output growth volatility is measured as the standard deviation of output growth rates over the speciﬁed period scaled by the absolute mean growth rate over the same period.

Table 6. Country

Algeria Egypt Iran Jordan Morocco Syria Tunisia Turkey

Mean output growth rates and measures of growth volatility for selected MENA countries 1961–1980

1981–2000

Mean

SD

3.407 2.673 2.536 3.236 4.248 4.954 4.297 2.439

10.768 6.099 9.655 9.123 4.882 15.210 4.522 3.440

Mean 0.202 2.785 2.238 0.072 1.280 1.834 2.269 2.386

1961–2000 SD

Mean

SD

3.148 1.241 6.126 5.048 5.827 6.617 2.569 3.797

1.804 2.729 2.387 1.582 2.764 3.394 3.257 2.412

7.997 4.344 7.983 7.468 5.515 11.685 3.747 3.577

Source: Computations are based on Penn World Tables. PWT 6.1. Also see the note to Table 5.

jobs at relatively high wage rates. On balance, we do not feel that the overall growth picture of the MENA region captured in Table 5 will be unduly affected by these omissions. Tables 1 and 3 of Chapter 2 in this volume by Makdisi, Fattah and Limam provide a more comprehensive country coverage (including 20 of the 23 MENA countries) based on the World Bank’s World Development Indicators (WDI). The results in Table 5 and those reported in Chapter 2 should be viewed as complementary rather than as substitutes. The most obvious conclusion to emerge from Table 5 is that the MENA countries’ growth has not been as bad as generally perceived, once a comparable output series is used across well-deﬁned geographical regions which are likely to be free of the sample selection bias encountered when country groupings by

18

Jeffrey B. Nugent and M. Hashem Pesaran

income levels are used. Indeed, for the 1960–80 period, MENA’s GDP per capita growth rate was second highest of all the regions, second only to Europe. While MENA’s growth rate of GDP per capita dropped off rather sharply after 1980, so did those of most regions. As a result, even during this rather dismal period in MENA’s growth experience, MENA’s GDP per capita grew faster than Latin America and SSA. For the whole period, 1960–2000, MENA’s GDP per capita grew at 2.54% per annum, which was well above the world average, slightly above the averages for both North and South America, more than double the growth rate of SSA. What also stems from Table 5 is that MENA’s growth has indeed been volatile. MENA’s growth rate volatility, however, fell slightly in absolute as well as relative terms between the 1960–80 and 1980–2000 periods. In the ﬁrst period its growth rate volatility was only slightly below that of SSA and much higher than that of any other region, whereas in the second period it was well below that of SSA and even slightly below that of Central and South America. 6. Growth and volatility across MENA countries Another distinctive fact of the growth and development experience in the MENA region is its diversity across countries and over time. See Table 6 in this chapter, and Tables 1 and 3 of Makdisi, Fattah and Limam (Chapter 2). For example, the estimates in Table 6 show that Tunisia and Syria grew considerably faster than the other MENA countries over the whole period. Only Egypt enjoyed more rapid growth in the 1980–2000 period than in the 1960–80 period. In the 1980–2000 period Algeria, Jordan and Syria experienced growth rates in GDP per capita that were at least 3% per annum lower than those over the 1960–1980 period. With respect to the volatility of growth rates, Syria’s was by far the highest over the whole period (11.69) relative to Turkey’s (3.58) which in itself is relatively high by international standards. Most of the MENA countries had lower growth rate volatility in the 1980–2000 period as compared to the 1960–80 period. Turning to WDI-based growth rates for the larger number of MENA countries in Table 3 of Makdidsi, Fattah and Limam (Chapter 2), it can again be seen that for the period 1970–2000, some countries (Tunisia, Egypt, Oman, Syria, Jordan and Turkey) have done relatively well, whereas others such as Djibouti, Iraq, Kuwait, Libya, and the UAE have done especially badly, at least in terms of GDP per capita growth. Moreover, despite its long-standing civil war and the tragic unrest in Darfur, Sudan has also performed better than many other countries of the region since 1985. The fact that some oil countries have done well, but others rather poorly, seems to suggest that a simple thesis like the ‘‘Dutch disease’’ characterization of oil-exporting economies will not explain the entire demise of growth in the MENA countries over the more recent past. Similarly, since many Moslem countries have done well (Indonesia, Malaysia

Introduction

19

and several of the Islamic countries of the MENA region) but others have done rather poorly, a simpleminded theory that considers Islam as the main impediment to growth in the MENA region might be wanting in view of the evidence. Moreover, many of the MENA Islamic countries were once among the most advanced countries in the world irrespective of the fact that they were Islamic. Similar differences can be seen within groups even when one distinguishes countries of the MENA region by their demographic characteristics. Hence, these rather special characteristics of the MENA countries can hardly be counted on to explain the demise of MENA’s growth as a whole. What the diversity of MENA’s growth experience does, however, is to beg explanation for these intraregional differences. Why, for example, has Tunisia been so successful relative to most of the other countries? Moreover, as noted above, the comparatively high volatility of MENA’s growth poses another challenge. What explains the sharp reversals in growth rates observed in individual MENA countries, not simply from one year to the next, but from one decade to another? The search for answers to these questions brings us to venture beyond the conﬁnes of economic analysis to political economy. What explains speciﬁc changes in policy regimes in some countries and why are they in some countries quite frequent, but in other countries virtually non-existent? What explains whether the changes are in a liberalizing direction or a tightening direction? What explains which things are liberalized and which are tightened? What effects do these policies, institutional and other changes have on subsequent performance in individual countries? To what extent can one generalize? It is hoped that the following chapters will be useful in shedding light on these and other related questions. 7. Contributed papers The contributed papers are organized in two parts. The ﬁrst part provides thematic, overall views of a number of areas in the whole of the MENA region, while the papers in the second part focus on speciﬁc issues in speciﬁc countries. A brief summary of the chapters in the order of their inclusion is given below. The remainder of the volume begins with a chapter by Samir Makdisi, Zeki Fattah and Imed Limam that investigates the overall growth performance of the MENA region over the period 1960–2000 and discusses its various determinants. The authors argue that the growth performance in the region has been mixed, and characterized by a higher degree of volatility compared with the other regions in the world. In comparing the growth pattern of the MENA region within an international perspective, they ﬁnd that capital is less efﬁcient; trade openness less beneﬁcial to growth and the impact of adverse external shocks more pronounced. In addition, Total Factor Productivity Growth (TFPG) in the MENA region, was not found to be an important source of growth in comparison to other regions, in part owing to lower quality of the institutions in the MENA region, modest stock of human capital and its educational system

20

Jeffrey B. Nugent and M. Hashem Pesaran

that focused on preparing students for public sector employment rather than private enterprises. These authors also ﬁnd that within the MENA region, nonoil and diversiﬁed economies have fared much better than oil-exporting countries both in terms of output growth and TFPG. They also note that the degree of exposure to internal and external shocks, the extent of economic diversiﬁcation and international competitiveness have been important in explaining variations in growth performance within the MENA region. Chapter 3 by Hadi Salehi Esfahani focuses on the political economy of the growth process in the region. The author argues that since the 1980s, economic growth in the MENA region has been slow and volatile. He attributes the problem to poor economic policies, such as excessive government interventions, large public sectors, restrictive trade policies, inefﬁcient mass subsidies and badly managed ﬁscal policies, and examines the political economy factors that have shaped those policies and their outcomes. This chapter offers a framework, based on the recent political economy literature, to explore the deeper forces behind the puzzles of growth variations in the region, going beyond the claims that powerful interest groups have been blocking reform. It focuses on the reasons why interest groups and politicians may not have been able to adopt better policies and generate larger surpluses for them to reap. The usefulness of the framework is illustrated by applications to key political economy issues in MENA. In particular this chapter examines the mechanisms through which resource rents have affected government policies and economic growth in the region and argues that the impact has been varied, depending on a number of other prevailing factors. The analysis includes the roles of authoritarianism, social heterogeneity, geopolitical rents and threats, human capital and administrative and judicial institutions. Chapter 4 by Hasan Ersel and Magda Kandil consider the interaction of ﬁnancial development and economic growth in the region. They identify two important channels: the direct channel that operates through mobilizing and allocating ﬁnancial resources and the indirect channel that acts through creating the appropriate environment for monetary policy to be effective. The authors analyze the performance of the MENA countries in mobilizing and allocating ﬁnancial resources using aggregative indicators. Their ﬁndings indicate that the MENA countries were quite successful in mobilizing ﬁnancial resources, but relatively less efﬁcient in allocating these resources. However, on average, the direct contribution of ﬁnancial development to economic growth in the MENA countries is not signiﬁcantly different from other developing economies. Further, the chapter analyzes the effect of ﬁnancial development on the effectiveness of monetary policy in a subset of MENA countries. The ﬁndings indicate that monetary policy in these countries is capable of inducing long-lasting nonneutral effects that differentiate trend output growth across countries over time. Monetary policy appears, therefore, to be an important determinant of economic growth in the MENA countries. The authors conclude by identifying areas for improvements, including the promotion of competition in the ﬁnancial

Introduction

21

sector, strengthening supervision and increasing the independence of central banks in the region. They also argue that further steps need to be taken to create a more appropriate legal environment (e.g. improve the quality of legislation and the judiciary system). Chapter 5 by Christopher Pissarides and Marie-Ange Veganzones-Varoudakis examines the relationship of the labor market and economic growth in the region. They point out the role that labor market plays in economic development through its impact on the acquisition and deployment of skills, and argue that countries in the MENA region have failed to deploy human capital efﬁciently despite high levels of education. This has been largely due to the presence of large public sector which has distorted incentives and because of excessive regulation in the private sector. Pissarides and Veganzones-Varoudakis, like Makdisi, Fattah and Limam in Chapter 2, argue that the education system is geared to the needs of the public sector, so the acquired skills are inappropriate for growth-enhancing activities. Excessive regulation of the private sector further removes the incentives for employers to recruit and train good workers. As a result, the MENA countries found it difﬁcult to adapt to new conditions in the 1990s and their rate of productivity growth fell to very low levels. The authors conclude that the MENA countries as a whole failed to keep up with the countries such as East and South-East Asia that used to be at a comparable level of development. Chapter 6 by Djavad Salehi-Isfahani considers the role of households in the growth process and argues that households in the MENA region as a whole have not been able to play a pro-growth role because of delay in fertility decline and low productivity of investment in education. The author attributes the high fertility and low productivity of education to certain aspects of the environment in which MENA households operate that condition their behavior and distort their decisions to supply labor and to accumulate productive human capital. He emphasizes two aspects of this environment: the large role of the state in the economy, which distorts the incentives in the education and labor markets and social gender norms, which inﬂuence the division of labor at home and in the economy. Salehi-Isfahani argues that the patriarchal gender norms limit women’s participation in work outside the home, resulting in higher fertility and lower labor force participation for the MENA women compared to countries with similar incomes. The strong role of the state affects incentives in the markets for education and labor. The prevalence of public sector employment and the regulation of private employment have increased job security, which has promoted formal schooling at the expense of acquisition of productive skills. As a result, households have invested in an inefﬁcient portfolio of human capital, which has limited their role on long-run growth. The remaining chapters in the second part of the volume address the growth experience of the individual economies. Chapter 7 by Hanaa Kheir-El-Din and Tarek Abdelfattah Moursi covers the growth experience of Egypt. It investigates the sources of the ﬂuctuations in

22

Jeffrey B. Nugent and M. Hashem Pesaran

aggregate economic activity in Egypt from 1960 until 1998, and highlights the effects of labor and capital productivity and of total factor productivity (TFP) on that country’s economic performance. Using the estimates of TFP provided in the chapter, the authors provide a detailed chronological characterization of the pattern of economic growth during a number of subperiods. Furthermore, the study examines the correlation between technological change and a number of macroeconomic variables. It ﬁnds that capital accumulation was the main driving force behind economic growth in Egypt from 1960 till the 1980s. It also identiﬁes two reasons behind Egypt’s decision to opt for capital-intensive modes of production despite its relative abundant labor force and chronically low savings rates. First, despite low domestic savings, foreign funds were available to ﬁnance major national projects. Second, the shortage of qualiﬁed workers and the country’s restrictive labor laws made it almost impossible to ﬁre unproductive workers, which caused producers to invest in labor-substituting industries. Although the adoption of capital-intensive methods of production led to average growth rates of above 6% per annum during 1960–89, capital accumulation alone was not sufﬁcient to sustain the rates of technical progress that are also required for growth. Combined with low labor productivity, this caused a decline in the contribution of physical capital to growth after 1991 and an increase in the contribution of TFP. The study suggests that the increase in the share of TFP in growth after 1991 may be attributed more to the fall in the rate of capital accumulation, than to improvements in the system of management and organization of production. Indeed, because TFP growth is also restrained by investment and physical capital shortages, this rise in TFP growth was not sufﬁcient for stimulating growth. The chapter concludes that the need to develop suitable educational and training programs to improve the quality of labor cannot be overemphasized. In the short run, the chapter argues in favor of measures that stimulate TFP growth by strengthening on-the-job training programs, by upgrading and increasing managerial training courses and by increasing investment in market-oriented research and development. In Chapter 8 Kivilcim Metin O¨zcan, U¨mit O¨zlale and C - ag˘ri Sarikaya investigate the growth experience of the Turkish economy over the past four decades. They identify three phases: 1963–1979, 1980–1999 and with the last phase starting with the exchange rate-based stabilization program in December 1999. The authors begin by providing a comparison of the Turkish economy with some of the MENA countries, and show that Turkey’s macroeconomic environment was considerably more stable during the 1960s and the 1970s than from mid-1977 onward where both economic and political conditions worsened signiﬁcantly. The second phase witnessed volatile output performance with rapidly increasing inﬂation, largely owing to lack of ﬁscal discipline. Over the period of analysis inﬂation rate was well above the average of the MENA countries. But in terms of output performance, although excessive ﬂuctuations are observed, the Turkish economy has done remarkably well, with increases in TFP accounting for a signiﬁcant portion of the growth. The analysis also suggests an

Introduction

23

important change in the sources of economic growth in the Turkish economy. While the output growth was largely driven by the increases in factor inputs (labor and physical capital) between 1960 and 1990, the growth since then has been primarily due to increases in TFP. The last phase covers the ﬁnancial crisis of February 2001 and the subsequent stabilization attempts. By utilizing an output gap measure, the authors ﬁnd that short-term capital ﬂows are likely to be the main driving force in generating excessive output ﬂuctuations in Turkey. Based on these ﬁndings, it is argued that the high economic growth rate that the Turkish economy has achieved over the recent years owes much to massive short-term capital inﬂows. Given the long history of volatile economic performance in Turkey, the authors conclude that it is too early to evaluate the effects of recent structural reforms on the long-term output dynamics. Chapter 9 by Ahmad Jalali-Naini provides an analysis of trends and developments of the Iranian economy over the period 1950–2003. The author argues that the Iranian experience reveals, in a number of ways, the potentials, problems and challenges that have been confronted by the other populated oil-exporting countries – a tale of high expectations turning into mediocre performance. From the 1960s up to the ﬁrst oil boom (1973–1974) the economy exhibited high GDP growth and low inﬂation rates. The oil boom brought in large ﬁnancial resources, but also induced soft-budget constraints and expanded the public sector. The huge increase in oil-ﬁnanced government spending raised growth temporarily above its trend; however, by creating inﬂation and opportunities for rent seeking, it resulted in economic and social instability. The positive wealth and saving effects of the oil boom did not have a long-lasting effect on either economic growth rates or per capita income levels. Since 1974 economic growth has been volatile and factor-intensive and TFP did not make a signiﬁcant contribution to the economic growth. The evidence also does not support that human capital accumulation has strongly contributed to economic growth. While ﬁxed investment/GDP ratios in Iran compare favorably with high-growth economies, the growth performance does not; too many resources have been spent to produce only an average growth performance. It is argued that Iran’s relatively rich resources, high rates of accumulation of physical and human capital, but a mundane growth performance and continued dependence on oil reﬂect inadequate economic institutions, price/incentive distortions, policy coordination and rent-seeking problems and the effect of the Iran–Iraq war. Chapter 10 by Mohamed Bechri and Sonia Naccache discusses the political economy of Tunisia. The chapter examines the development of the economy over the past ﬁve decades and identiﬁes the following main features: (1) an early and sustained focus on social modernization and population control, (2) a heavy involvement of the State in the economy, even after the abandonment of collectivism in 1969 and (3) a robust long-run GDP growth of about 5% per year, against only 3.5% for the MENA region as a whole. The chapter adopts a political economy approach to explain independent Tunisia’s development

24

Jeffrey B. Nugent and M. Hashem Pesaran

policy choices. Since the country’s independence from France in 1956, these policies shifted from State intervention (1956–1960) to socialism (1961–1969), to export promotion (1970–1985). Finally, following a severe balance of payment crisis in 1986, Tunisia adopted a structural adjustment program under IMF guidance. The liberal policies implemented under this framework were further boosted with the signing of an FTA with the European Union in 1995. In conclusion, the chapter draws on some lessons from Tunisia for the other MENA countries. It stresses the importance of women rights, population control and modern education, and refers to recent examples of Morocco and Kuwait regarding women status reform to argue that action is still possible. Chapter 11 by Mohamed Abdelbasset Chemingui and Moataz Mostafa ElSaid reviews the macroeconomic performance of Algeria over the period 1962–2000. The chapter considers prevailing economic policies and changes introduced into the economic environment over this period. It also conducts a growth accounting exercise to disaggregate output growth in Algeria into growth in capital, labor and TFP. Over the past four decades, Algeria’s growth performance has been characterized by high ﬂuctuations and generally slower growth than in the neighboring developing countries. The poor economic growth performance that Algeria has experienced is shown to be the result of a slow growth in TFP which can be attributed to the following factors: (i) the inefﬁcient management of the country’s natural resources; (ii) inappropriate use of macroeconomic policies, speciﬁcally price deregulation and economic reforms aimed at increasing investment; (iii) structural impediments limiting the private sector’s contribution to the country’s economic activities and (iv) an overall economic environment dominated by the public sector, a weak ﬁnancial sector and a slow and non-transparent privatization process. The authors argue in favor of policy measures that focus on boosting the economy’s potential growth capacity. These include measures to improve the competitiveness of existing activities, to attract investment in sectors other than the energy-related ones so as to diversify the economy, to accelerate the trade liberalization process, to encourage domestic and foreign investment in export-oriented industries, and to adopt complementary supply-side policies such as promoting education and training and increasing spending in the health sector. In addition, Algeria should aggressively focus its efforts on the improvement of its investment climate. Chapter 12 by Adam B. Elhiraika and Annas H. Hamed explores the economic performance of the UAE as an example of an oil-dependent economy. It attempts to explain the determinants of economic growth in the UAE using growth-accounting as well as regression techniques. The chapter shows that overall the performance of the economy ﬂuctuated widely over the period 1975–2002 under the inﬂuence of volatile oil prices. The period up to the early 1980s saw sustained high real growth rates owing to rising oil prices and huge government investment in physical and social infrastructure. As from around the mid-1980s, GDP growth decelerated sharply with generally declining oil prices. Subsequent government austerity measures were directed largely toward capital

Introduction

25

expenditure since most of the basic infrastructure projects had been completed by then, while most current expenditure categories have become long-term commitments. The chapter also makes a number of recommendations. To reduce exposure to exogenous shocks (in particular the oil price shocks), the economy needs to be diversiﬁed by encouraging private sector participation in production of goods and by expanding markets for the UAE’s products regionally as well as internationally. The authors also argue for a more developed ﬁnancial sector, with a well-functioning equity market, expansion of non-oil government revenues through privatization of public enterprises, rationalization of government expenditures and a reassessment of subsidies and incentive programs. Chapter 13 ends the volume with an essay by Ragui Assaad on the importance of institutions and households’ decisions for the growth process in Egypt. The chapter begins with the observation that macroeconomic analyses of growth in Egypt indicate that labor contributes very little to growth, and that human capital, and in particular female human capital, contributes little to increases in TFP (see, for example, Chapter 7 by Kheir-Eldon and Morsi in this volume). These puzzling results come at a time when the role of human capital in generating growth is being increasingly stressed in the international literature and when investment in female human capital, internationally, is deemed to be one of the activities with the highest social returns. One of the main objectives of this chapter is to shed light on these seemingly puzzling results by analyzing the microeconomics of the production and deployment of human capital in the Egyptian economy, with a particular emphasis on the institutional contexts in which these decisions are made. The chapter argues that the long-standing policy of the Egyptian government to guarantee employment in the government sector for upper secondary and university graduates has distorted household decisions as to the level and type of human capital to acquire and has resulted in the entrapment of signiﬁcant portions of the human capital in unproductive government employment. With the distorted signals they receive from the labor market, the educational decisions of households are strongly shaped by the desire to ensure access to lifetime employment in the government, an objective that might well be at odds with realizing the productivity beneﬁts of education for the economy as a whole. In the case of women the situation is further compounded by social norms deﬁning what constitutes appropriate female employment. The author argues that such norms tend to restrict women’s employment to a limited range of occupations, types of workplaces and locations, which in turn leads to overcrowding in these segments of the labor market and depressed wages. The chapters that follow further our understanding of the economies of the MENA countries and highlight the common issues that are important to the region as a whole. It is hoped that a better understanding of the economic issues and problems of the region will help foster a better understanding of the nature of the political conﬂicts that have afﬂicted the region for decades. It should be

26

Jeffrey B. Nugent and M. Hashem Pesaran

clear that the resolution of economic problems of the region without a proper understanding of the political, sociological, cultural and religious dimensions of the problems is unlikely to be sustainable. This is largely reﬂected in the multidisciplinary nature of the contributions. It is hoped that this volume encourages further investigation of the problems of the MENA region.

References Acemoglu, D., S. Johnson and J.A. Robinson (2001), ‘‘The colonial origins of comparative development: an empirical investigation’’, American Economic Review, Vol. 91, pp. 1369–1401. Adams Jr., R.H., and J. Page (2003), ‘‘International migration, remittances and poverty in developing countries’’, Washington, DC, World Bank Policy Research Working Paper 3179. Alnasrawi, A. (1994), The Economy of Iraq, Westport: Greenwood. Askari, Nowshirvani and Jaber (1997), ‘‘Economic Development in the GCC: The Blessing and the curse of oil’’, Greenwich, CT: JAI Press. De Soto, H. (2000), The Mystery of Capital: Why Capitalism Triumphs in the West and Fails Everywhere Else, New York: Basic Books. Easterly, W. and R. Levine (2002), ‘‘Tropics, germs, and crops: how endowments inﬂuence economic development’’, NBER Working Papers 9106, National Bureau of Economic Research, Inc. Elbadawi, I.A. (2005), ‘‘Reviving growth in the Arab World’’, Economic Development and Cultural Change, Vol. 53, pp. 293–326. Galal, A. and B. Hoekman (editors) (2003), Arab Economic Integration: Between Hope and Reality, Washington, DC: Brookings. Gallup, J.L. and J.D. Saks (2000), ‘‘Agriculture, Climate, and Technology: Why are the Tropics Falling Behind?’’, American Journal of Agricultural Economics, Vol. 82, pp. 731–777. Henry, C.M. and R. Springborg (2001), Globalization and the Politics of Development in the Middle East, Cambridge: Cambridge University Press. Hoekman, B. and J. Zarrouk (editors) (2000), Catching up with the competition: trade opportunities and challenges for Arab countries, Ann Arbor: University of Michigan Press. Hotelling, H. (1931), ‘‘The economics of exhaustible resources’’, Journal of Political Economy, Vol. 39, pp. 137–175. Issawi, C. (1982), An Economic History of the Middle East and North Africa, London: Methuen. Konan, D.E. (2003), ‘‘Alternative paths to prosperity: economic integration among Arab countries’’, pp. 61–101 in: A. Galal and B. Heokman, editors, Arab Economic Integration, Washington, DC: Brookings. Lofgren, H. (editor) (2003), Food, Agriculture, and Economic Policy in the Middle East and North Africa, Amsterdam: JAI Press, Elsevier Science.

Introduction

27

Maskus, K.E. and D.E. Konan (1997), ‘‘Trade liberalization in Egypt’’, Review of Development Economics, Vol. 1, pp. 275–293. Miniesy, R., J. Nugent and T. Yousef (2004), ‘‘Intra-regional trade integration in the Middle East: past performance and future potential’’, in: H. Hakimian and J. Nugent, editors, Trade Policy and Economic Integration in the Middle East and North Africa: Economic Boundaries in Flux, London: Routledge. Nugent, J.B. and N. Sanchez (1993), ‘‘Tribes, chiefs and transhumance: a comparative institutional analysis’’, Economic Development and Cultural Change, Vol. 42, pp. 87–113. Nugent, J. B. and N. Sanchez, (1998), ‘‘Common property rights as an endogenous response to risk’’, American Journal of Agricultural Economics, Vol. 80, pp. 651–657. (Reprinted in Bruce Larson, editor, Property Rights and Environmental Problems, Aldershot: Ashgate Publishing Ltd., 2003. Nugent, J.B. and N. Sanchez (1999), ‘‘The local variability of rainfall and tribal institutions: the case of Sudan’’, Journal of Economic Behavior and Organization, Vol. 39, pp. 263–291. Owen, R. (1981), The Middle East in the World Economy, 1800– 1914, London: Methuen. Owen, R. and S. Pamuk (1999), A History of the Middle East Economies in the Twentieth Century, Cambridge: Cambridge University Press. Pesaran, M.H. (1982), ‘‘The system of dependent capitalism in pre- and postrevolutionary Iran’’, International Journal of Middle East Studies, Vol. 14, pp. 501–522. Pesaran, M.H. (2006), ‘‘A pair-wise approach to output and growth convergence’’, Journal of Econometrics (forthcoming). World Bank (2005), Unlocking the Employment potential in the Middle East and North Africa: Toward a new Social contract. Washington, D.C.

This page intentionally left blank

PART I: The Main Themes in the MENA Region as a Whole

This page intentionally left blank

CHAPTER 2

Determinants of Growth in the MENA Countries Samir Makdisi, Zeki Fattah and Imed Limam Abstract The overall growth performance of the MENA region over the period 1960–2000 has been both mixed and characterized by a higher degree of volatility compared to the other regions in the world. In comparing the growth pattern of the MENA region within an international perspective, we have found that: capital is less efficient; trade openness less beneficial to growth and the impact of adverse external shocks more pronounced. In addition, Total Factor Productivity Growth (TFPG) in the MENA region, was not an important source of growth in comparison to the other regions owing to among other things, the lower quality of its institutions, modest stock of human capital and its educational system that focused on preparing students for public sector employment. Within the MENA region, non-oil and diversified economies have fared much better than the oil-exporting countries both in terms of output growth and TFPG. Finally, the degree of exposure to internal and external shocks, the extent of economic diversification and international competitiveness were found to be important factors explaining variations in growth performance within the MENA region.

Keywords: economic development, technological change and growth, analysis of growth JEL classifications: O00, O01, R11

Corresponding author. CONTRIBUTIONS TO ECONOMIC ANALYSIS VOLUME 278 ISSN: 0573-8555 DOI:10.1016/S0573-8555(06)78002-6

r 2007 ELSEVIER B.V. ALL RIGHTS RESERVED

32

Samir Makdisi, Zeki Fattah and Imed Limam

1. Introduction Over the last two decades or so, growth performance of the MENA region as a whole has been disappointing and mixed relative to that of the rest of the developing countries.1 The meager growth performance in the region has not kept pace with the high growth rates of its labor force and hence did not address the most important development challenge of the region, namely, its high level of unemployment. In comparison with the other regions in the world, growth rates in the MENA countries have been remarkably volatile and at times lower than that of the poor-performing regions such as Sub-Saharan Africa (SSA). This growth pattern is believed to be inextricably linked to several characteristics of most of the countries in the region; notably, their heavy dependence on oil; weak economic base; high population growth; low rates of returns on investment in physical and human capital; low level of integration in the world economy; underdevelopment of market institutions and, with very few exceptions, the omnipresence of the State. The relatively better growth performance of the MENA countries in the 1960s, 1970s and the ﬁrst half of the 1980s is largely attributed to favorable external environment in the form of high energy prices. This situation has been reversed in the second half of the 1980s and the early 1990s resulting in sharp declines in domestic investment, savings and growth. Oil is justiﬁably perceived by many as the most important source of growth in the MENA region.2 Indeed available data indicate that per capita income in Arab oil-producing countries has consistently been higher than in the non-oil Arab countries by many folds throughout the period under consideration. The margin of difference was even greater when comparing per capita investment. The per capita investment in non-oil producing countries amounted, on average, to only 20 percent of that in the oil-producing countries. However, other factors have also played an important role in shaping the growth pattern of the region. The external environment is not likely to be as favorable as it had been since the early 1970s and until the mid-1980s. Analyzing what have been the main determinants of the recent growth performance is a ﬁrst step toward identifying what needs to be done to make growth more sustainable. This paper will attempt to use the broader framework provided by the recent empirical growth literature to measure the relative contribution of the main

1

Unless otherwise mentioned, the deﬁnition of MENA followed in this paper includes: Algeria, Bahrain, Comoros, Djibouti, Egypt, Iran, Iraq, Jordan, Kuwait, Lebanon, Libya, Mauritania, Morocco, Oman, Qatar, Saudi Arabia, Somalia, Sudan, Syria, Tunisia, Turkey, United Arab Emirates (UAE) and Yemen. Among these the oil exporters are: Algeria, Bahrain, Iran, Iraq, Kuwait, Libya, Oman, Qatar, Saudi Arabia and the UAE. 2 MENA countries currently comprise more than one-third of the proven world oil reserves and account for>30 per cent of the actual world production.

Determinants of Growth in the MENA Countries

33

sources of growth in the MENA region. While subsets of the MENA countries (especially oil versus non-oil countries) are economically heterogeneous, the region’s growth is analyzed within an aggregate framework that puts it into an international perspective in comparison with other regions. Section 2 gives an overview of the growth performance of the countries of the region during the period 1960–2000. Section 3 puts the growth performance of MENA into an international perspective by showing how the region’s performance has been affected by the main determinants of economic growth in comparison to the rest of the world and estimating the relative contribution of the different factors of production to this growth in MENA versus the non-MENA countries. Section 4 analyzes the growth performance of the MENA countries from a regional perspective by explaining variations in the growth rates for the countries of the region included in the sample. Section 5 concludes the chapter. 2. Growth record of the MENA region Over the period 1960–2000, the growth of MENA has followed the growth pattern of the world economy. As shown in Table 1, after a period of relatively high growth rates during the 1960s and 1970s, growth has slowed down during the 1980s and the 1990s. However, the impact of the worldwide recession in the early 1980s has been more pronounced for the MENA countries whose growth performance has not only been below world average, but also weaker than that of the low-performing SSA. Table 2 exhibits the sectoral distribution of production of selected MENA countries according to their average growth performance over the period 1960–1998. The numbers show that countries with better growth performance tend to be more diversiﬁed than the other countries with relatively larger manufacturing and service sectors. In contrast, countries with weaker growth performance have either large agriculture or mining and quarrying sectors. This can explain why the more diversiﬁed countries tend to be more resilient to shocks and hence capable of achieving sustainable growth rates. Another possible explanation of the anemic growth rates recorded by the less-diversiﬁed countries is that most of the growth underachievers are major oil-exporting Table 1.

MENA real GDP per capita growth in international perspective

Group MENA MENA oil-producing countries MENA non-oil producing countries East Asia and Paciﬁc Latin America and Caribbean SSA World

1961–1970

1971–1980

1981–1990

1991–2000

3.9 5.8 2.9 4.3 2.7 1.8 3.1

3.0 0.2 4.2 3.6 2.3 1.6 2.5

1.3 1.6 0.5 2.5 0.5 0.2 1.0

1.1 5.9 1.7 6.4 1.7 -0.4 1.2

Source: World Bank, World Development Indicators (2003).

Samir Makdisi, Zeki Fattah and Imed Limam

34

Table 2. Percentage sectoral distribution of production in selected MENA countries according to growth performance averages for the period 1960–1998 Country

Agriculture

Mining and Quarryinga

Manufacturing

Other Sectors

Countries with better growth performance Egypt 22.9 Jordan 8.6 Morocco 18.2 Oman 17.4 Syria 22.5 Tunisia 15.6 Turkey 25.3 Average 18.6

11.5 12.3 14.1 55 9.9 13.9 9.0 18.0

19.1 12.4 17.2 1.5 13.5 13.0 16.6 13.3

46.5 66.7 50.5 26.1 54.1 57.5 49.1 50.1

Countries with lower growth performance Algeria 11.3 Iran 19.7 Kuwait 0.4 Mauritania 31.4 Saudi Arabia 4.6 Sudan 40.9 UAE 1.4 Average 15.7

34.9 25.2 54.5 18.4 55.0 7.5 59.5 36.4

11.4 10.7 6.9 11.6 7.3 7.2 6.6 8.8

42.4 44.4 38.2 38.6 33.1 44.4 32.5 39.1

Source: World Bank, World Development Indicators (2000). a Includes construction, electricity and gas.

countries with an already high level of GDP, a high proportion of which is essentially resource rent. Under the circumstances it would be difﬁcult for these countries to grow rapidly even when there is a rapid productivity growth in the non-resource sectors.3 The MENA countries have failed to catch up with the developed world during the last two decades. Graph 1 shows that the ratio, RIUS, of the 2002 relative per-capita income, in purchasing power parity (PPP) terms, with respect to the United States to that of 1980 was below unity for almost all the countries in the region pointing to the widening income gap with respect to the United States.4 The only exceptions were Egypt, Oman and Tunisia that have had a relatively better growth performance over the last two decades.

3 4

We owe this point to an anonymous referee. For any country i, this ratio is computed as follows:

RIUSi ¼

GNI02i GNI80i : GNI02US GNI80US

The idea of using such a ratio is borrowed from Nunnenkamp (2004).

Determinants of Growth in the MENA Countries

Graph 1.

35

Evaluation of the relative income (PPP) of MENA region countries with respect to US (1980–2002)

Algeria Bahrain Comoros Egypt Iran Jordan Kuwait Mauritania Morocco Oman SaudiArabia Sudan Syria Tunisia Turkey UAE

0.0

0.65 0.66 0.07 1.12 0.86 0.61 0.49 0.73 0.86 1.15 0.35 0.97 0.76 1.03 0.99 0.33 0.5

1.0

1.5

Another salient feature of the recent growth performance of the MENA countries is its high volatility. Table 3 shows that during the period 1970–2000, the average per capita GDP growth rate of the MENA countries has been characterized by a high variability in comparison to the other regions and with the world average. The average growth variability as measured by the standard deviation of the real per capita GDP growth rate for the MENA region is almost four times that of the world and twice that of the low-performing SSA region. Starting from the second half of the 1980s, although the variability of growth rates has declined somewhat for the MENA region as a whole, it remained higher than the world average. This high variability in the growth performance of the MENA region may be explained by several factors including the lack of economic diversiﬁcation that increases vulnerability to external shocks. This applies especially to the oilexporting countries that are frequently subject to the vagaries of the international oil market as well as to countries that rely heavily on agriculture and whose performance is frequently affected by unfavorable weather conditions. Political instability and the perennial regional conﬂict may further explain the very high variability of the growth performance in a number of countries in the region. Tables 4 shows that the pattern of investment has also followed the growth pattern described above. Investment has declined sharply during the 1980s and the 1990s. Oil countries have borne the brunt of this adjustment with substantial declines in the investment ratios. Non-oil countries have maintained, over the same period, fairly stable investment rates above the level of their low saving ratios.

Samir Makdisi, Zeki Fattah and Imed Limam

36

Table 3.

Average per capita GDP growth rates (1970–2000)

Country Algeria Bahrain Djibouti Egypt, Arab Republic Iran, Islamic Republic Iraq Jordan Kuwait Lebanon Libya Mauritania Morocco Oman Saudi Arabia Sudan Syrian Arab Republic Tunisia Turkey UAE Yemen, Republic MENA East Asia and Paciﬁc Latin America & Caribbean SSA World

Average Growth

Standard Deviation of Growth

1.1 0.3 4.6 3.2 0.3 5.1 2.0 2.6 3.2 5.4 0.2 1.7 2.7 0.4 1.5 2.4 3.1 2.2 3.4 1.6 0.2 5.6 1.5 0.1 1.5

5.7 6.0 3.0 3.0 7.6 21.3 7.7 11.7 18.4 10.0 4.0 4.6 7.3 6.5 6.1 7.9 3.8 3.8 8.7 4.9 4.8 2.3 2.5 2.6 1.3

Source: World Bank, World Development Indicators (2003).

Table 4. Gross domestic investment (per cent of GDP) Group MENA MENA oil-producing countries MENA non-oil producing countries East Asia and Paciﬁc Latin America and Caribbean SSA World

1961–1970

1971–1980

1981–1990

1991–2000

17.4 21.3 14.9 19.1 20.4 16.9 24.3

26.3 29.1 24.1 28.6 23.6 21.1 25.3

24.1 23.6 24.4 31.8 20.3 17.5 23.2

22.5 22.0 22.1 34.4 20.8 17.1 22.6

Source: World Bank, World Development Indicators (2003).

Many studies link the performance of the MENA region to the wide ﬂuctuations in its terms of trade. For instance, comparing terms of trade ﬂuctuations across a number of regions for the period 1980–1995, Gamo et al. (1997, p. 14) have found that the oil-exporting countries in the region have the highest terms of trade variability. In addition, terms of trade variability of the MENA

Determinants of Growth in the MENA Countries

37

non-oil exporting countries were also found to be higher than that of the developing countries as a group. This pattern may be attributed to the excessive reliance of the MENA countries on volatile oil receipts that account for a large part of their total export revenues. Even non-oil MENA countries have been subject to ﬂuctuation in their own terms of trade, being exporters of primary products themselves. Oil price ﬂuctuation has also indirectly affected these countries through its impact on the ﬂows of workers’ remittances, investment and ﬁnancial assistance ﬂows from the oil-exporting countries. All these factors combined show a priori that the MENA region is remarkably vulnerable to external shocks. 3. MENA economic growth in a global context In this section we will put MENA’s growth performance into an international perspective by: ﬁrst, using a cross-country regression framework to analyze the main determinants of growth in the MENA region as compared to the rest of the world and second, using an accounting framework to identify the relative contribution of the different factors of production to the growth of the countries of the region in comparison to the other developing countries. 3.1. Cross-country regression The absence of guidance from the growth theory as to which variables to include, makes the choice among the great number of possible correlates of growth a difﬁcult one. However, our selection will be guided by variables that proved more ‘‘robust’’ than others in recent literature.5 Therefore, we attempted to identify a small set of regressors that would account for most of the variation in cross-country per capita GDP growth. In addition, we focused on variables of relevance to the MENA region’s recent growth performance. With the above strategy in mind, the set of chosen variables included initial conditions, human capital, investment ratio, external shocks, macroeconomic performance, openness, natural resource abundance, and output volatility. We deal with each one of them in detail. Concerning initial conditions, recent empirical growth literature provides ample evidence about the existence of conditional income convergence across countries. Conditional on the rate of investment and the other relevant variables, a poor country would tend to grow relatively faster because the marginal productivity of capital is relatively higher.

5

There is a lot a spuriousness associated with the Barro-type (Barro, 1991) cross-country regression framework. However, it could be used as a suggestive tool to measure the relative contribution of the many sources of growth across countries and regions.

38

Samir Makdisi, Zeki Fattah and Imed Limam

This convergence is evidenced by the negative relationship between the growth rate of per capita GDP and the initial level of GDP per capita after controlling for other relevant variables such as measures of government policies, institutions, politics and variables related to the character of national population. We will consider the 1960 level of real GDP per capita, Y60, as a measure of initial income. Countries with higher initial stock of human capital and knowledge have also been found to be able to forge ahead through higher growth rates. The simple inclusion of a measure of human capital in a Barro-type regression equation, however, overlooks the dual role played by the latter. Human capital not only affects growth as an additional factor of production, together with physical capital and labor, but also the efﬁciency with which these factors are combined. Benhabib and Spiegel (1994) have found evidence that human capital affected Total Factor Productivity Growth (TFPG) through its impact on the capacity of a country to innovate and the capability of using and adapting foreign technology. Arguably, human capital also encourages the accumulation of other factors of production. In our analysis, a measure of human capital will be used to explain both growth as well as TFPG. As a proxy for the level of human capital, we will take the 1960 primary school enrollment ratio, PESENR60. The investment ratio, INVY, is included as a regressor since it has been found by many studies, including that of Levine and Renelt (1992), to have a positive and robust correlation with the growth rates. Since it is generally assumed that positive external shocks are associated with higher growth and vice versa, we have included for each country in the sample the GDP per capita growth of its trading partners weighted by trade shares, GPART. Other conventional proxies for external shocks such as the percentage change in the terms of trade and its variability have not resulted in statistically signiﬁcant impact on growth. Macroeconomic performance plays an important role for growth sustainability. Fischer (1993) has shown that growth is negatively associated with inﬂation, large budget deﬁcits and distorted foreign exchange markets. Among the three measures, we favor inﬂation for several reasons. First, internationally comparable data on budget deﬁcits are scattered and not available for a large number of the MENA countries. In contrast, the widely used blackmarket premium (BMP) rate as a measure of distortion in foreign exchange market is neither a good proxy for the level of distortions in the economy nor an appropriate measure of the adequacy of macroeconomic policies. For instance, the low BMP rates in Egypt or the Arabian Gulf countries reﬂect a greater abundance of foreign exchange more than the absence of distortions in the economy or the presence of stable macroeconomic environment. In addition, the high BMP rates in many countries of the MENA region tend to reﬂect the impact of wars and sociopolitical instability characterizing these countries.

Determinants of Growth in the MENA Countries

39

Openness has been used extensively in the literature as a major determinant of growth performance. Openness affects growth positively to the extent that it magniﬁes the beneﬁts of international knowledge spillover and technological diffusion as well as enforces cost discipline through import competition and the drive to export. The ratio of trade to GDP as a measure of openness may not always be the most appropriate measure to use, especially in the case of the MENA oil-producing countries whose natural resource abundance is generally associated with high trade ratios. In addition, countries could exhibit high trade ratios while simultaneously maintaining highly restrictive trade regimes. Therefore, the trade ratio is not necessarily a good proxy for openness. Following the work of Sachs and Warner (1995), we adopt the deﬁnition of openness, SOPEN, reported in Sachs and Warner (1997a). Openness purports to the fraction of years during the period 1970–1990 in which the country was rated as open according to the following criteria: (a) non-tariff barriers coveringo40 percent or more of traded goods, (b) average tariff rates ofo40 percent, (c) a BMP ofo20 percent, (d) no extreme controls in the form of taxes, quotas or state monopolies on exports and (e) the country is not considered a socialist country. A value of SOPEN ¼ 1 means that the country has remained open to trade during the entire period, while a value SOPEN ¼ 0 means that it has remained completely closed. Among the recently introduced variables into the empirical growth literature is natural resource abundance.6 Sachs and Warner (1997a, 1997b), for instance, have found compelling evidence that countries with high initial ratio of natural resource exports tend to grow slowly in subsequent periods. Earlier ﬁndings of development literature about the disappointing performance of resourceabundant countries have motivated their study on the link between natural resources and economic growth. Natural resource abundance may affect growth performance negatively by causing overvaluation of the national currency thereby impeding the growth of non-oil exports.7 Resource-rich countries are also associated with wasteful consumption and public investment pattern, and provide incentives for rent-seeking and other unproductive activities. In addition, it is widely observed that natural resource availability has tended to forestall reform. Finally, the secular decline of world prices of natural resources and their high volatility translate into high uncertainty which, in turn, impacts growth negatively. The idea of incorporating natural resource abundance as a correlate of economic growth is of great appeal given the high endowment of the MENA countries in natural resources notably, oil. However, despite the fact that they could be a curse for long-term growth, natural resources may also contribute

6 7

Natural resources are deﬁned as primary agriculture, fuels and minerals. Radelet et al. (1997, p. 8.).

40

Samir Makdisi, Zeki Fattah and Imed Limam

positively to growth. Oil export revenues have contributed to the improvement of welfare and helped ﬁnance investment in infrastructure and human capital in most MENA countries. In fact, many of the MENA oil-producing countries such as Kuwait, Bahrain, Qatar and the UAE are ranked by the 1999 Human Development Report, among the top 45 countries in the world in terms of the United Nations (1999)’ Human Development Index. The ultimate impact of natural resources on growth is therefore an empirical question. To measure this impact, we will use the share of exports of primary products in GNP, SXP. This variable is measured for the year 1970 and is borrowed from Sachs and Warner (1997a). A ﬁnal growth agent that is not very frequently used in the recent empirical growth literature is the role of output volatility in explaining output growth. Volatility to the extent that it properly reﬂects uncertainty, should be negatively linked to growth. Uncertainty could affect growth through many channels. For instance, under irreversibility and large sunk costs, uncertainty delays investment decisions and affects output growth. Ramey and Ramey (1995) have found empirical evidence on the negative link between output growth and its volatility. Binder and Pesaran (1999) have provided a theoretical rationale for this link based on the statistical properties of stochastic versions of conventional growth models. In the context of the MENA region, inclusion of a volatility variable is justiﬁed on the ground that volatility of output growth is a salient feature of the region. The factors responsible for this volatility include inter alia ﬂuctuations in world oil prices, weather conditions, ﬂuctuation in workers’ remittances, capital ﬂows, not to mention the prevailing political instability and the involvement of many countries in civil wars and in the regional conﬂict. To measure volatility, we will use the standard deviation of per capita growth rates, STDG, over time for each country in the sample. 3.1.1. Cross-country regression results To give an order of magnitude to the contribution of the above agents of growth, we regress the average growth rate of real per capita GDP on the variables described above. All variables, unless otherwise mentioned, are averaged over the period 1960–1998 or any subperiod within it, in case of unavailable data. Most of the variables used in this paper, unless indicated otherwise, come from the World Bank Global Research Project (GRP) database and World Development Indicators. The sample comprises 86 countries of which 8 are from the MENA region: Algeria, Egypt, Jordan, Morocco, Saudi Arabia, Syria, Tunisia and Turkey. The rest of the MENA countries could not be included in the regression analysis for lack of relevant data. Table 5 gives summary statistics of the variables used in the regression analysis. Table 6 reports the regression results of our model using both ordinary least squares (OLS) and instrumental variables (IV) methods to account for the likely endogeneity of some of the regressors. The ﬁrst column in Table 6 gives the OLS

Determinants of Growth in the MENA Countries

Table 5.

41

Summary statistics of variables used in regressions

Variable

N

Mean

Standard Deviation

Minimum

Maximum

Growth INVY INFL Y60 PESENR60 SXP SOPEN GPART STDG

86 86 86 86 86 86 86 86 86

1.81 19.91 57.35 2547.27 74.42 0.13 0.38 2.43 4.38

1.66 5.59 183.13 2398.73 32.48 0.11 0.45 0.37 1.93

2.70 6.38 3.08 319.00 5.00 0.01 0.00 1.69 1.66

6.43 34.70 1148.22 9895.00 144.00 0.60 1.00 3.69 10.30

Table 6. Cross-country growth regression results dependent variable: real per capita GDP growtha Variable Common Constant INVY INFL Y60 PESENR60 SXP SOPEN GPART STDG MENA specificb INVYMENA SOPENMENA GPARTMENA MENA N Adjusted R2

OLS

Instrumental Variables

OLS with Interaction

1.970 (-2.162) 0.111 (3.854) 0.002 (-2.313) 0.0003 (-4.401) 0.017 (3.446) 2.268 (1.916) 1.100 (2.974) 0.450 (1.313) 0.040 (-0.568)

1.751 (1.710) 0.116 (3.289) 0.002 (2.197) 0.0002 (-1.656) 0.019 (2.824) 3.586 (2.642) 1.016 (1.683) 0.441 (1.066) 0.833 (1.117)

1.805 (2.000) 0.132 (3.913) 0.002 (2.353) 0.0003 (4.749) 0.017 (3.510) 2.779 (2.396) 1.230 (3.458) 0.190 (0.564) 0.009 (0.137)

63 0.59

0.208 0.922 2.294 0.359 86 0.69

86 0.65

(4.369) (2.063) (3.511) (0.190)

a

All Standard errors are heteroskedasticity-consistent. T-ratios are in parentheses. Signiﬁcant at the 5 percent level. b

estimates assuming homogeneity across countries in the response of growth to its main determinants. With respect to the recent empirical literature, our results conﬁrm the relative importance of the investment ratio (INVY), macroeconomic performance (INFL), initial level of income (Y60), human capital (PESENR60), natural resource abundance (SXP) and the degree of openness (SOPEN) in affecting long-term growth performance.

42

Samir Makdisi, Zeki Fattah and Imed Limam

The respective coefﬁcients of these variables were found to have the expected signs and were statistically signiﬁcant. Some of the variables portraying the impact of external shocks (GPART) and volatility (STDG) were found to be insigniﬁcant for the whole sample. We have computed standardized coefﬁcients obtained by regressing standardized values of the dependent variable on the standardized values of the included explanatory variables. The standardized coefﬁcients thus computed depict the effect of one standard deviation change in each of the explanatory variables on growth and allow comparing the relative predictive importance of each one of these variables. Investment and initial level of income turned out to be the most inﬂuential with standardized coefﬁcients of +0.38 and -0.38, respectively. The initial enrollment rate is the next important variable in terms of magnitude with a coefﬁcient of 0.34. The standardized coefﬁcients of the rest of the variables were respectively, 0.30 for openness, 0.20 for inﬂation, 0.15 for natural resource endowment, 0.10 for growth of trade partners and 0.05 for growth variability. To account for the possible endogeneity of some of the regressors that may have affected OLS estimates, notably, the investment ratio; column 2 of Table 6 provides the IV parameter estimates. The IV regression uses trade ratio (X+M)/ GDP, debt ratio, share of Government consumption in GDP and share of manufactures in exports as instruments. The choice of these instruments was guided by two criteria: their lack of correlation with the disturbance term of the regression equation and their high correlation with the explanatory variables. The quality of the instruments was checked by testing for their validity using the procedure described in Gourieroux and Monfort (1996, pp. 616–618). The w2 statistic of the residual-based test shows that the null hypothesis of the validity of instruments could not be rejected at the 10 per cent signiﬁcance level.8 However, given that the use of instruments with incomplete data reduces the sample size from 86 to 63 and given the close similarity of the OLS and IV estimation results, we will only use the OLS method to further extend the analysis of MENA growth performance. It is to be noted that the previous OLS results were reproduced after eliminating observations that do not ﬁt well with the model in terms of unusual response or unusual prediction. These outlier observations were detected using standardized and studentized residuals. An observation is eliminated if it has either a large standardized or studentized residual. Using these criteria, 9 of the 86 observations in the sample could be classiﬁed as unusual.9 Upon the reestimation of the OLS equation without the unusual observations, the parameter estimates remained fairly the same in terms of magnitude, sign and signiﬁcance.

8

The computed w2 value was 0.91 below the tabulated value of 1.64. These include the observations corresponding to China, the Congo Democratic Republic, Jamaica, Jordan, South Korea, Nicaragua, Pakistan, Saudi Arabia and Uganda.

9

Determinants of Growth in the MENA Countries

43

The only exception was inﬂation whose parameter became less signiﬁcant, but remained rightly signed even after the elimination of outliers. This conﬁrms the fact that inﬂation hurts growth especially at unusually high rates. In an attempt to disentangle the regional from the global growth characteristics, we have allowed for differentiated growth impacts of relevant variables depending on whether the country under study is from the MENA region. These variables pertain to some of the factors that are often presented to explain the lower growth performance of the MENA region: the lack of capital efﬁciency, lack of integration in the world economy and vulnerability to external shocks. More speciﬁcally, the MENA region is allowed to have different slope coefﬁcients from any non-MENA region in the sample with respect to the three variables: INVY, SOPEN and GPART. To do this, we have included interaction terms between these three variables and a dummy variable, MENA, taking a value of 1 if the country is from the MENA region and 0 otherwise. The third column of Table 6 gives the previous OLS results with MENA-speciﬁc coefﬁcients. A separate dummy variable for the MENA region was also included in the last regression to see whether the included variables account for a fairly reasonable. Explanation of growth in the region. In contrast with the previous OLS results related to the whole sample, the MENA-speciﬁc interaction terms reported in column 3 of Table 6, seem to suggest that the MENA region differs signiﬁcantly from the rest of the world in the three variables believed to affect their performance. Interestingly, the MENA region has been found to be more vulnerable to uncertainty caused by changes in trade partners’ income. This result notwithstanding, one should bear in mind that the inﬂuence of the above sources of uncertainty and the impact of external shocks may vary from one country to another. To give further evidence on the vulnerability of the MENA region to external shocks, we have tried different measures of terms of trade change. However, all of these measures turned out to be statistically insigniﬁcant. The ambiguous impact of terms of trade on long-term growth has been reported in several instances in the literature.10 There are several plausible explanations for that. The terms of trade variable has been obtained, given the requirements of our cross-country analysis, by computing its average growth over the entire period 1960–1998. Therefore, the impact of terms of trade as an exogenous shock could be diluted owing to the operated smoothing. To take this shortcoming into account, we have used the standard deviation of the terms of trade of each country over the entire period as an alternative. The new variable turned out to be insigniﬁcant as well.

10

For instance, Gamo et al. (1997, p. 27) reported a similar result for the case of several countries in the region.

44

Samir Makdisi, Zeki Fattah and Imed Limam

The impact of terms of trade on economic growth could be proportional to the dependence of any country on a limited number of exported commodities. In other words, a diversiﬁed economy is less likely to be affected by a decline in the terms of trade given that the impact will be limited to a relatively small number of sectors. In contrast, deterioration in the terms of trade will be felt throughout the economy in case of high export concentration or excessive dependence on a limited number of export commodities such as oil. Finally, the ambiguous effect of terms of trade on growth could be due to the asymmetry of this effect. A persistent improvement in terms of trade might lead to an improvement in income and expenditure, while deterioration does not necessarily lead to a proportional reduction in these variables. The 2000 World Bank report on Global Economic Prospects, reports several cases in point from the MENA region.11 For instance, following the 1998 slump in oil prices, Bahrain, Oman and Saudi Arabia used foreign reserves and accumulated foreign assets to alleviate the pressure on ﬁscal deﬁcits and trade balances. In contrast, other oil exporters such as Algeria and Yemen with more binding ﬁnancial constraints had to adjust through expenditure cuts and exchange rate devaluation. This asymmetry in policy reaction might have affected the growth performance in many MENA countries by increasing uncertainty. Our results also show that the investment coefﬁcient for the MENA region is statistically different from and much lower than that of the whole sample. This result is prima facie, at odds with the fact that the MENA countries have persistently maintained high investment ratios that were for the most part above the world average.12 The negative MENA-speciﬁc investment coefﬁcient is largely attributed to the endemic problem of capital inefﬁciency in the region. Many researchers have presented plausible explanations for the low efﬁciency of capital in the MENA region. Page (1998) suggests that this low efﬁciency of capital is due to the dominant role of the public sector and the nature of capital inﬂows in the region destined mainly to ﬁnance public investments and lowproductivity projects in a non-tradable sector such as housing. He also points out that protectionism and lack of integration in world economy precluded these countries from boosting their efﬁciency and competitiveness. Others, such as El Badawi (1999), argue that the low efﬁciency of capital may be attributed to the fact that most of the countries in the region provide an inadequate business environment and institutional support for investment and private sector development. Openness does not seem to play an important role in the MENA region as for the whole sample. The lower impact of openness on growth in the MENA region may be explained by the prolonged application of high tariffs in support

11 12

World Bank (2000, pp. 142–143). Refer to Table 3 in the paper.

Determinants of Growth in the MENA Countries

45

of import-substitution strategies in many MENA countries during the 1960s and the 1970s. Excluding the Arab Gulf States that maintain relatively low tariff rates, the average unweighted tariff rate in the MENA countries is high relative to most regions of the world. Protection remains high even in the most advanced reformers in the region including Egypt, Morocco and Tunisia. Indeed, trade protection in the non-oil MENA countries generally remains higher than in the low-income countries (Table 7). In addition, the rate of reduction in trade barriers has been lower compared to the other regions (Srinivasan, 2002). Finally, lack of statistical signiﬁcance of the MENA dummy shows that the general factors and the MENA-speciﬁc factors mentioned above, provide a fairly good explanation of the region’s growth performance.

Table 7. Trade protection indicators for MENA (most recent year in the late 1990s) Country/Region

NTB Coverage (per cent)

Standard Deviation (per cent)

Weighted Average (per cent)

Simple Average (per cent)

0.0 15.8 32.8 5.5 28.8 13.1 15.6 15.9

15.8 16.7 12.8 31.2 130.6 1.2 2.7 30.1

18.9 17.3 28.8 25.8 13.7 4.7 10.3 17.1

21.6 24.2 29.9 35.7 28.1 4.7 12.0 22.3

5.5

10.9

12.6

15.5

13.4

15.0

12.5

15.3

14.7

12.3

11.6

13.8

15.6

7.0

3.4

4.3

10.9

11.0

6.7

9.8

9.9 17.1 4.5 8.2

16.8 8.5 13.3 11.7

8.7 11.9 14.2 18.8

13.1 13.1 17.7 19.7

Jordan Algeria Tunisia Morocco Egypt Oman Saudi Arabia MENA Comparators By income group Low-income countries Lower middleincome countries Upper middleincome countries High-income countries By region Europe and Central Asia East Asia Latin America SSA South Asia Source: Srinivasan (2002).

46

Samir Makdisi, Zeki Fattah and Imed Limam

3.2. Growth accounting in international perspective To put MENA growth performance in further international perspective, we use the growth accounting framework to see whether factor accumulation or factor productivity has accounted for most of the growth differential between MENA and the other regions in the sample. Growth can be the result of the growth of inputs such as capital and labor or the increase in the efﬁciency of these inputs, namely, their productivity. The debate over the share of inputs as compared with that of their total productivity is still very lively. The empirical evidence is mixed. Mankiw et al. (1992) and many others argue that the share of physical and human capital together with population growth account for as much as 80 percent of international variation in per capita income. Young (1995), in contrast, argues that what is often labeled as the ‘‘Asian miracle,’’ is the outcome of a temporary rapid factor accumulation. Those holding a different view claim that TFP is the key to economic growth and that factor accumulation plays only a less important role.13 The share of capital in total income needed in a typical growth accounting exercise could be imposed a priori, computed from national income accounts or estimated. Each one of these approaches has its own caveats. Imposing a uniform value, typically between 0.3 and 0.4, is not realistic since this value varies a great deal across a wide spectrum of countries. Computation of this value from national accounts, in contrast, is too data demanding especially in light of the availability of internationally comparable data on capital stock and labor that would readily permit the direct estimation of the capital share. In this paper, we have chosen the estimation approach. To carry out the accounting exercise, we have used a two-factor, homogeneous of degree one, Cobb–Douglas production function in per capita form. Capital shares required to measure the relative contribution of factor accumulation and productivity were estimated using the following regression equation: DLogðY it =Lit Þ ¼ li þ ai DLogðK it =Lit Þ þ it The slope coefﬁcient in the above equation represents the capital share in output, Y represents real output, K the capital stock and L the labor. Y is measured by real GDP and is obtained from the World Bank database. The capital stock data are taken from Nehru and Dhareshwar (1994) and L is approximated by total labor force and taken from the World Bank World Development Indicators. Real GDP and capital stock series are based on 1987 constant prices. The sample we used comprises 92 countries and the data cover the period 1960–1997. The list of countries included in our sample was determined on the basis of the availability of capital stock data in Nehru and Dhareshwar (1994)

13

See for instance, Klenow and Clare (1997).

Determinants of Growth in the MENA Countries

Table 8. Region

Whole sample (world) MENA East Asia SSA Latin America

47

Regional average capital shares

Number of Countries

Average Capital Share (Short-term Version)

Average Capital Share (Long-term Version)

92 11 6 21 22

0.67 0.61 0.48 0.59 0.79

0.59 0.54 0.38 0.48 0.78

that used the perpetual inventory method to obtain capital stock series for the period 1950–1990 assuming a ﬁxed annual depreciation rate of 4 percent.14 In this paper, we have completed the capital series from 1991 until 1997 with ﬁxed investment ﬁgures from the World Bank database using the same methodology and assumption. To account for the possible impact of the noise generated by the high variability of yearly data, we have estimated two versions of the above equation. A short-term version using original data for the output per worker and capital per worker and a long-term version were obtained by smoothing the short-term variation of the same variables using three-year moving averages. Table 8 provides the regional averages of capital shares using the two speciﬁcations indicated above. In this table, several remarks are in order. First, our estimates of the world average capital share were found to be above the commonly assumed values of 0.3 or 0.4. This ﬁnding is in line with the estimation results provided, for instance, in Senhadji (1999) and Nehru and Dhareshwar (1994) who have found capital shares above 0.5. Recently, Keller and Nabli (2002) reported an estimated value for the capital share of 0.49 based on a sample of 95 economies. Bisat et al. (1997) have previously found that the average capital share in many countries of the MENA region is well above 0.5 and 0.7 for some oil-producing countries. Second, high-performing East Asian countries hold the lowest capital shares in the group. Third, Latin American and industrialized countries have the highest capital shares. These ﬁndings have an implication for the computation of TFPG. Instead of imposing a uniform value for the share of capital over the entire sample of countries, we have made the less-restrictive hypothesis of assuming region-speciﬁc values based on the estimation results. Table 9 provides estimates of the relative contribution of capital, labor and TFPG to the economic growth of the MENA countries included in the sample in comparison with the other regions.

14

The MENA countries included in the sample are: Algeria, Egypt, Iran, Iraq, Jordan, Kuwait, Libya, Morocco, Sudan, Tunisia and Turkey.

Samir Makdisi, Zeki Fattah and Imed Limam

48

Table 9.

GDP growth rate decomposition (1960–1997)

Country

Growth

Capital

Labor

TFP

Algeria Egypt Iran Iraq Jordan Kuwait Libya Morocco Sudan Tunisia Turkey

0.031 0.057 0.046 0.023 0.052 0.022 0.058 0.049 0.030 0.051 0.050

0.034 0.035 0.049 0.020 0.116 0.015 0.165 0.025 0.028 0.028 0.040

0.006 0.011 0.010 0.020 0.013 0.056 0.011 0.013 0.014 0.012 0.007

0.009 0.011 0.013 0.016 0.051 0.052 0.095 0.011 0.020 0.010 0.003

MENA SSA HASIA LATIN

0.043 0.028 0.066 0.034

0.048 0.020 0.044 0.035

0.011 0.011 0.015 0.005

0.020 0.003 0.008 0.006

World

0.037

0.032

0.007

0.003

Overall, the results show the predominance of capital contribution over that of labor and TFPG in growth performance during the period 1960–1997. The MENA region as a whole, has experienced the lowest contribution of TFP to economic growth in comparison with the rest of the regions. For the 11 MENA countries included in the sample, only the diversiﬁed economies of Egypt, Morocco, Tunisia and Turkey managed to have positive TFPG. Out of the seven remaining MENA countries in the sample that had negative TFPG, ﬁve were oil-exporting countries. Although, the use of oil GDP in the computation of TFPG may be problematic owing to its large and persistent ﬂuctuation (Bisat et al., 1997, p. 17), lack of ample available data on employment and capital stock in the oil sector did not enable us to compute TFPG after excluding oil. To assess the relative contribution of the variables accounting for interregional TFPG performance, we have regressed TFPG on relevant variables based on recently established results in the literature.15 We conducted these regressions using values for capital shares ranging from 0.3 to 0.7. Each hypothetical value for capital share was applied uniformly over the different countries in the sample. This was done to see whether different values of the capital share affect the impact of the regressors on TFPG.

15

Only six MENA countries were included in the analysis given the lack of data. They were: Algeria, Egypt, Jordan, Morocco, Tunisia and Turkey.

Determinants of Growth in the MENA Countries

Table 10. Variable ICRG INFL Y60 PRIM60 SXP EASIA SSA LATIN Adjusted R2 Number of observations

49

Determinants of TFPG at different values of capital share pendent variable: TFPG

a,b

de-

a ¼ 0.3

a ¼ 0.4

a ¼ 0.5

a ¼ 0.6

a ¼ 0.7

0.24 (3.00) 0.001 (2.08) 0.0004 (6.62) 0.02 (2.75) 1.71 (2.01) 0.19 (0.51) 0.89 (2.60) 0.67 (2.27) 0.58 75

0.18 (2.31) 0.001 (1.75) 0.0003 (5.37) 0.02 (2.59) 1.12 (1.31) 0.49 (1.42) 0.68 (1.87) 0.61 (1.99) 0.45 75

0.13 (1.56) 0.001 (1.44) 0.0002 (4.05) 0.02 (2.41) 0.55 (0.61) 0.79 (2.38) 0.47 (1.20) 0.56 (1.71) 0.29 75

0.07 (0.83) 0.0005 (1.17) 0.0002 (2.8) 0.02 (2.21) 0.03 (0.04) 1.09 (3.30) 0.26 (0.61) 0.50 (1.43) 0.16 75

0.02 (0.20) 0.0004 (0.94) 0.0001 (1.70) 0.01 (2.01) 0.61 (0.62) 1.39 (4.07) 0.05 (0.11) 0.45 (1.19) 0.09 75

a

Constant terms not included and t-ratios in parentheses. Estimation based on White heteroskedasticity-consistent standard errors. Signiﬁcant at the 5 percent level, Signiﬁcant at the10 percent level. b

The included regressors were the quality of institutions, International Country Risk Guide, ICRG;16 inﬂation rate, INFL; the initial income, Y60; the initial enrollment rate in primary school, PESENR60 and the adopted measure of natural resource abundance, SXP. Other conventional variables such as openness, growth in terms of trade and political stability have been tried, but were dropped for lack of statistical signiﬁcance. Regional dummies were also included to account for interregional differences. Table 10 reveals that for lower values of the capital share, the parameter estimates tend to be signiﬁcant and of the expected signs. Institutions and the stock of human capital, as approximated by the initial enrollment rate, positively affect the TFPG. The negative sign attached to the initial income points to the existence of a catching-up effect at the TFPG level. Inﬂation was also found to affect the TFPG negatively. Finally, the natural resource curse was found to apply at the productivity level, too. In other words, natural resource abundance affects TFPG negatively. At higher values of the capital share, the explanatory power of the model drops. This is due to the fact that at higher values of the capital shares, capital accounts for a higher portion of the overall economic growth as well as the TFPG. Hence the other variables become less relevant. However, it should be

16

Our measure of institutions, ICRG, is a composite of four indicators: (a) government repudiation of contracts, (b) risk of expropriation, (c) rule of law and (d) bureaucratic quality. This measure is computed for the decade of the 1980s and is borrowed from Easterly and Levine (1997). Knack and Keefer (1995) were the ﬁrst to introduce these institutional variables into growth empirics.

50

Table 11. Variable ICRG INFL Y60 PESENR60 SXP a

Samir Makdisi, Zeki Fattah and Imed Limam

Relative contribution of relevant variables to TFP growth differential a MENA–World

MENA–SSA

MENA–Hasia

MENA–Latin

0.18 0.03 0.39 0.31 0.01

0.04 0.04 0.07 0.19 0.06

0.48 0.00 0.05 0.54 0.09

0.49 0.05 0.01 0.21 0.13

Assumptions: MENA a ¼ 0.5; SSA a ¼ 0.5; Hasia a ¼ 0.4; Latin a ¼ 0.7; world a ¼ 0.5.

mentioned that the only two variables that remained signiﬁcant for different values of the capital share were the initial income and human capital. This result is widely in line with the recent empirical ﬁndings on the TFPG.17 While the initial level of income affects the potential of catching up notably through higher productivity, human capital affects the TFP by determining the capacity of a country to innovate and the speed of technological diffusion. To put the TFPG performance of the MENA countries into global perspective, we have computed the contribution of the relevant variables to TFPG for different values of the capital share. Table 11 reports the results after applying different regional values for the capital share. Based on our estimation, we have applied the value of 0.5 to the MENA, Sub-Saharan regions as well as to the whole sample. The values applied for the other regions were 0.4 for East Asia and 0.7 for Latin America. Overall, the results exhibited in Table 11, point to the overriding importance of the quality of institutions and the stock of human capital in explaining the lower productivity performance of the MENA countries in comparison with the high-performing East Asian countries and with the rest of the world in general. More speciﬁcally, Table 11 shows that human capital, as approximated by initial enrollment ratios, accounts for the MENA region’s TFPG underperformance with respect to East Asia and Latin America. Despite the net improvement in many educational institutions in the MENA countries, illiteracy ratios remain very high and the educational attainment of the labor force very low in comparison with the other regions in the world. For the year 2000, UNESCO predicted that the illiteracy ratio in the MENA region would be around 31 percent for adults above 15 years compared to 26 percent in the group of developing countries and 13 percent in the East Asia and Paciﬁc region. These statistics are even more alarming given the high gender gap in terms of literacy and in comparison with the other regions in the world. The average illiteracy rate among females in the MENA region for the year 2000 was expected to be around 40 percent; almost double the average

17

See for instance the ﬁndings of Senhadji (1999) and Benhabib and Spiegel (1994).

Determinants of Growth in the MENA Countries

51

illiteracy rate for males.18 Illiteracy among women is linked to poor health and low education attainment among children, and hence the low quality of human capital. The recent statistics published by Barro and Lee (2000) provide further evidence of the relative weaker educational attainment of the labor force in the MENA region in comparison with the other regions in the world. For instance, they estimated that the average years of schooling for the population aged 15 and over, for the year 2000, for the high-performing MENA countries such as Egypt, Jordan, Syria, Tunisia and Turkey will be: 5.51, 6.91, 5.77, 5.02 and 5.29, respectively. These rates do not compare favorably with 8.83 for Argentina, 7.55 for Chile, 10.84 in Korea, 7.05 for Singapore and 8.76 in Taiwan.19 Using panel data for six countries in the MENA region, El-Erian et al. (1998, p. 11) have found that the rapid expansion in education did not result in higher productivity or more rapid economic growth.20 They argue that the weak link that they found between education and growth may be due to the low quality of education that focuses more on repetition of deﬁnitions, knowledge of facts and concepts and less on developing critical thinking and problem-solving capacity. In addition, the educational system had been focused on preparing students for public sector employment. This outcome is largely attributed to distorted educational choices in these countries caused by the higher wages prevailing in the dominant public sector that are set without consideration for alternative employment opportunities in other sectors. Economic Research Forum (ERF) for the Arab countries, Iran and Turkey (2000, p. 6) reports that for the early 1990s, the average share of civilian government employment in the MENA region was about 17.5 percent compared with o9 percent for the developing countries as a group. In addition, the share of central government wages in GDP was at 10 percent; almost double that of the world average. As rightly argued by the report, this situation has resulted in the prevalence of acute job redundancies and increasing unemployment rates in the region owing to the glaring mismatch between the outcomes of the educational system and the requirements of the rapidly changing labor market. To explain the lower quality of education, Ridha (1998) argues that in most countries of the region education systems are overpoliticized to the extent that they deviate from the objectives they are supposed to achieve. The educational tools and practices such as the curriculum, textbooks and teaching methods are often manipulated to reach political ends.

18

These statistics are derived from the UNESCO Statistical Yearbook (1999). These updated statistics are available on the following web site: http://www2.cid.harvard.edu/ ciddata/. The data set could also be accessed through the web page of Jong-Wha Lee: web. korea.ac.kr/jwlee. 20 The countries included in their studies are: Algeria, Egypt, Jordan, Kuwait, Syria and Tunisia. 19

52

Samir Makdisi, Zeki Fattah and Imed Limam

Another argument advanced by Pritchett (2001) can possibly explain the weak link between education and productivity growth in the MENA countries. He argues that in a perverse institutional environment such as the one prevailing in many MENA countries, education and accumulated capital could be used in wasteful and counterproductive activities. In addition, the fact that most of these countries are natural resource-abundant, provides further incentives for the proliferation of rent-seeking activities. The second important factor explaining the underperformance of the MENA region in terms of the TFPG is the quality of its institutions. Institutions can be deﬁned as the regular and patterned forms of social behavior and interaction among human beings established by formal and informal rules.21 Institutions matter for growth and productivity because they affect incentives of actors. For they affect the behavior of people in a society and very often lock their behavior within a regularized pattern, institutions may produce path dependence that could explain prolonged periods of poor economic performance and hence the inability of the poor countries to catch up. Many published indicators for the quality of institutions show that the MENA region lags behind most of the regions in the world. The average value of the variable used in our paper to depict the overall quality of institutions, ICRG, for MENA countries was 4.31. This score is below that of the world average, 5.59, and the high-performing East Asian countries, 6.37. A recent report by the World Bank (2002) using a composite index of governance that includes several relevant indicators of the quality of institutions such as the quality of administration and accountability shows that the MENA region ranks at the bottom when compared with countries with the same level of income. A similar conclusion can be reached based on the results of a study by Kaufman et al. (2002) to estimate six dimensions of governance that represent reasonable indicators for the quality of institutions. In comparing the score of the different regions in the world for the year 2000, we have found that MENA lags behind most of the regions in terms of voice and accountability, that is, the ability of citizens to participate in decision making and hold accountable public ofﬁcials and regulatory quality that has to do with regulations governing the business sector, government intervention, quality of legislation, investment environment and tax effectiveness. Table 12 shows the average scores for ﬁve institutional indicators published by the Political Risk Services: government stability, investment proﬁle, corruption in government, rule of law and quality of bureaucracy. Despite the fact that the MENA countries have better scores than the world averages for three out of the ﬁve reported indicators, the indicators of corruption and bureaucracy

21

North (1990).

Determinants of Growth in the MENA Countries

Table 12. Group

Government stability Investment proﬁle Corruption in government Rule of law Quality of bureaucracy

53

Institutional indicators of the MENA regiona

MENA (1995)

East Asiab (1995)

World (1995)

MENA (2000)

East Asia (2000)

World (2000)

6.9

7.1

6.2

7.5

10.3

7.4

5.7

6.5

5.5

10.2

8.0

9.9

3.1

3.9

3.5

2.5

2.8

3.0

4.4 2.0

5.1 2.8

4.3 2.3

4.1 1.7

4.1 3.0

4.0 2.2

Source: PRS Group (2002), International Country Risk Guide. a Higher scores mean better institutions and vice versa. b Includes only the eight best-performing East Asian countries.

remain, however, below world averages and much lower than those of the highperforming East Asian countries. Empirical work has shown that governance and institutions affect the quality of policy which has a direct bearing on investment and business climate (Knack and Keefer, 1995, 1997). As regards the MENA countries, reducing bureaucratic ineffectiveness, red tape, corruption, excessive government intrusion and improving the deplorable state of government services remain, in this regard, major challenging tasks for the MENA region. 4. Explaining intraregional MENA growth performance In the previous section, the growth performance of MENA countries was compared with the other reference regions. In this section, we will dwell on the relative performance of individual MENA countries with respect to the region’s average performance. As shown previously in Table 3, there is a considerable variation in the growth performance of MENA countries during 1970–2000. It was also shown that growth in the oil-exporting countries was subject to a higher variability than that of the non-oil exporting countries. In addition, the average growth performance of the oil-exporting countries has been below that of the non-oil exporting countries. The MENA countries that were able to achieve an average real per capita GDP growth rate over 2 percent a year during the period 1960–1998 were the non-oil exporting countries, except, Oman. The best growth performers were Egypt, Jordan, Morocco, Oman, Syria, Tunisia and Turkey. Oman’s average growth rate was the highest followed by that of Egypt and Tunisia. Interestingly enough, only countries with relatively high growth rates have managed to achieve positive TFPG over the last three decades or so. Oil-exporting countries have, in general, had negative TFPG. These facts are

54

Samir Makdisi, Zeki Fattah and Imed Limam

corroborated by our own estimates and those of the few studies such as, Keller and Nabli (2002), Bisat et al. (1997) and Nehru and Dhareshwar (1994) that have provided estimates for TFP growth rates for countries in the MENA region. In addition, MENA growth overachievers had in general the lowest growth variability among the sample group except, Oman. The higher variability of the latter can be attributed to the fact that it is an oil-exporter and therefore subject to the effect of oil price ﬂuctuation. Available data and the scope of this study do not permit the establishment of systematic links between growth performance and the country-speciﬁc characteristics (structural, policy, initial conditions, institutional, political and other internal and external factors). Preliminary analysis indicates, however, that the better achievers tend to have above-average indicators for integration in the world economy; be it through the crude measure of openness, share of FDI in GDP or share of manufactured exports in total exported commodity. They also tend to be more diversiﬁed and have initiated economic reform earlier than the other countries. To account for the relative growth performance within the MENA region, we have used a cross-country regression framework applied to a panel of data comprising 13 MENA countries and spanning the period 1970–1998. For the sake of increasing sample variation, we have used for each country six 5-year period averages for all variables and for the periods 1970–1974, 1975–1979, 1980–1984, 1985–1989, 1990–1994 and 1995–1998. The last period is one year shorter than the previous periods for lack of relevant data. We have adopted the same cross-country regression framework used previously. However, some of the variables were discarded for lack of statistical signiﬁcance. Other variables were appended to reﬂect either speciﬁc characteristics of the region or cross-country differences in terms of factors affecting growth. For instance, we have included government consumption as a ratio to GDP, GCY, to reﬂect the predominant role of the state in the region and the impact of government-induced distortions. We have also included the share of manufactures in total merchandise exports, MANUF, to reﬂect the impact of economic diversiﬁcation (or the lack of it) on growth as well as the extent of competitiveness in international markets. The debt-service ratio, DEBTS, was included to reﬂect the impact of debt overhang on the growth of many countries in the region. The other variables we have used are: the share of investment in GDP, INVY; the GDP per capita growth of each country’s trading partners weighted by trade shares, GPART and the level of income per capita in the beginning of each period, YINI. The model was estimated using the random-effect panel regression method.22 Although the estimation results of this dynamic panel may have been plagued by

22

Hausman’s speciﬁcation statistic for the test of ﬁxed versus random-effect model was (1.23 at 6 degrees of freedom) in favor of the latter with a p- value of 0.98.

Determinants of Growth in the MENA Countries

Table 13.

55

Panel data estimation of growth determinants in the MENA region dependent variable: per capita GDP growth rate

Variable

Coefﬁcient

T-Ratio

INVY DEBTS GPART MANUF YINI GCY

0.15 0.10 0.91 0.03 0.0004 0.16

2.46 2.91 2.48 1.72 4.48 2.33

Number of countries ¼ 15; number of observations used ¼ 61; adjusted R2 ¼ 0.51; country intercepts not reported. Signiﬁcant at the 5 percent level. Signiﬁcant at the 10 percent level.

Table 14. Country Algeria Bahrain Egypt Iran Jordan Morocco Oman Saudi Arabia Sudan Syria Tunisia Turkey UAE

Relative contribution of relevant variables to growth differential in selected MENA countries Predicted Growth INVY DEBTS GPART MANUF YINI GCY Individual Effect 0.36 1.21 2.93 0.03 2.19 1.66 3.15 0.02 0.45 1.67 3.09 2.33 3.09

4.88 3.94 3.27 3.72 4.52 3.56 4.16 3.18 1.96 3.47 4.11 2.95 4.23

4.38 0.00 1.76 0.27 1.56 3.02 0.35 0.00 1.75 1.36 1.91 2.58 0.00

1.65 0.36 1.90 1.50 1.52 1.45 2.64 2.37 1.35 1.52 1.55 1.39 2.28

0.07 0.27 0.91 0.08 1.40 1.34 0.55 1.15 0.04 0.67 1.85 1.58 0.99

0.98 3.74 0.60 1.69 1.10 0.71 2.46 3.37 0.30 1.58 0.98 1.13 9.25

2.44 3.41 2.60 3.13 4.23 2.72 4.23 4.09 1.93 2.60 2.45 1.65 2.55

1.54 1.36 1.81 0.23 1.64 1.76 2.85 1.77 1.09 1.54 0.92 1.78 1.22

the inconsistency of the estimates owing to short sample size, we use these results to convey some of the views presented above and to single out country-speciﬁc factors that have shaped growth in certain MENA countries.23 The estimation results reported in Table 13 show that the parameters of all the variables considered were statistically signiﬁcant and of the expected sign. Based on the data available for 15 countries, Table 14 shows that high investment ratios have contributed signiﬁcantly to the relatively better growth performance of countries such as Oman and Tunisia. In contrast, export

23

We owe this argument to an anonymous referee.

56

Samir Makdisi, Zeki Fattah and Imed Limam

diversiﬁcation and international competitiveness explain the relatively better performance of countries such as Tunisia, Turkey, Jordan and Morocco. Debt overhang has negatively impacted growth in Algeria, Morocco and Turkey. Large government size has had a detrimental effect on growth in Jordan, Oman and Saudi Arabia. Variation in trading partner growth has had a notable impact on the growth of oil-exporting countries such as Oman, Saudi Arabia and the UAE. Finally, the high variation in individual country’s intercepts in Table 14, points to the fact that the country-speciﬁc growth determinants or the unexplained growth, remain relatively high. These individual factors should be tackled at a more disaggregated country-speciﬁc level.

5. Concluding remarks The overall growth performance of the MENA region has been mixed as well as characterized by a higher degree of volatility in comparison with the other regions of the world. Several sources of uncertainty in the region can explain this volatility. These include, among others, ﬂuctuations in world oil prices, weather conditions, workers’ remittances, capital ﬂows, not to mention factors contributing to sociopolitical instability in the region such as civil wars and regional conﬂicts. In analyzing the growth pattern of the MENA region within an international perspective, we have found that: capital is less efﬁcient; trade openness less beneﬁcial to growth and the impact of adverse external shocks more pronounced. In addition and in comparison with the other regions, TFPG was the least important source of growth in MENA. Among the MENA countries included in the sample, only Egypt, Morocco, Tunisia and Turkey had positive TFPG. The quality of institutions and human capital accounted for the lower performance of the MENA countries in terms of TFPG and in comparison with the other regions of the world. MENA still lags behind the rest of the world in terms of the quality of its bureaucracy and a business environment that provides weak incentives for investment and depresses its return. Human capital was also found to contribute to the relative underperformance of the MENA region. Educational attainment of the labor force remains relatively low in comparison with the other regions and illiteracy rates, especially among women, continue to be relatively high. The dominance of the public sector in the economy together with guaranteed employment of university graduates in most of the MENA countries may have had a negative impact on productivity. The degree of exposure to internal and external shocks, the extent of economic diversiﬁcation and international competitiveness were also found to be relevant factors in explaining variations in growth performance within the MENA region.

Determinants of Growth in the MENA Countries

57

In view of the aggregate nature of our study and in light of the above ﬁndings, several relevant issues deserve further analysis within a country-speciﬁc context. Future research on the region needs to shed important additional lights on the determinants of growth in each of the countries concerned. Among the areas that deserve further investigation are the within-regional variation in the role of human and physical capital, the inﬂuence of the State, institutions and the relative impact of external and internal shocks as they relate to economic growth. This analysis would contribute to our understanding of the growth process in the MENA countries. It is equally useful to show, for instance, how economic policy, institutions, politics and other country characteristics affect the way factors of productions are used and combined. It would also be desirable to analyze the impact of the sectoral decomposition of growth in the MENA countries. Such a decomposition is useful for identifying the sectors that have been successful in achieving better growth performance, expanding investment and employment and raising productivity and earnings. This decomposition should also help explain why certain sectors have been more successful than the others. Last but not the least is the urgent need for policy makers and researchers alike to ﬁnd ways to deal with the major challenges in the region, namely, improving resilience to external and internal shocks, diversifying economic activity and reducing excessive reliance on natural resources, improving the business environment and the quality of institutions for higher and more productive investment in physical and human capital and devising an adequate incentive structure for a more job-creating growth. Acknowledgements The authors would like to thank Bill Easterly, Dani Rodrik, Jong-Wha Lee, Seppo Honkapohja, two anonymous referees and others for their comments on the earlier drafts of the paper including those present during the World Banksponsored workshop on the Global Development Network held in Prague during the period June 9–11, 2000. References Barro, R. (1991), ‘‘Economic growth in a cross-section of countries’’, Quarterly Journal of Economics, Vol. 104, pp. 407–433. Barro, R. and J.W. Lee (2000), ‘‘International data on educational attainment: updates and implications’’, Working Paper No. 42, Center for International Development at Harvard University, April. Benhabib, J. and M. Spiegel (1994), ‘‘The role of human capital in economic development: evidence from aggregate cross-country data’’, Journal of Monetary Economics, Vol. 34, pp. 143–173.

58

Samir Makdisi, Zeki Fattah and Imed Limam

Binder, M. and M.H. Pesaran (1999), ‘‘Stochastic growth models and their econometric implications’’, Journal of Economic Growth, Vol. 4, pp. 139–183. Bisat, A., M.A. El-Erian and T. Helbling (1997), ‘‘Growth investment, and saving in the Arab economies’’, IMF Working Paper WP/97/85. Easterly, W. and R. Levine (1997), ‘‘Africa’s growth tragedy: policies and ethnic divisions’’, Quarterly Journal of Economics, Vol. 112(4), pp. 1203–1250. Economic Research Forum (ERF) for the Arab Countries, Iran and Turkey (2000), Economic Trends in the MENA Region, Cairo. El Badawi, I. (1999), ‘‘Can reforming countries perform an Asian miracle?: role of institutions and governance on private investment’’, in: I. Limam, editor, Institutional Reform and Development in the MENA Region, Arab Planning Institute and Economic Research Forum for the Arab Countries, Iran and Turkey. El-Erian, M., T. Helbling and J. Page (1998), ‘‘Education, human capital development and growth in the Arab economies’’, Paper presented in the Joint Arab Monetary Fund, Arab Fund for Economic and Social Development Seminar on ‘‘Human Resource Development and Economic Growth’’, May 17–18, 1998, Abu Dhabi, UAE. Fischer, S. (1993), ‘‘The role of macroeconomic factors in growth’’, Journal of Monetary Economics, Vol. 32(3), pp. 485–512. Gamo, P.A., A. Fedelino and S.P. Horvitz (1997), ‘‘Globalization and growth prospects in Arab countries’’, IMF Working Paper WP/79/127. Gourieroux, C. and A. Monfort (1996), Statistique et Mode`les Econome´triques, Vol. 2, Economica, 2d edition, Paris. Kaufman, D., A. Kraay and P. Zoido-Lobaton (2002), ‘‘Governance matters II: updated Indicators for 2000/2001’’, available through internet at: http:// www.worldbank.org. Keller, J. and M.K. Nabli (2002), ‘‘The macroeconomics of labor markets outcomes in MENA over the 1990s: how growth has failed to keep pace with a burgeoning labor market’’, World Bank, available through internet at http:// www.worldbank.org/mdf/mdf4/papers/keller-nabli.pdf. Klenow, P.J. and A.R. Clare (1997), ‘‘The neoclassical revival in growth economics: has it gone too far?’’ in: NBER, Macroeconomics Annual 1997, Cambridge, MA: MIT Press. Knack, S. and P. Keefer (1995), ‘‘Institutions and economic performance: crosscountry test using alternative institutional measures’’, Economies and Politics, Vol. 7(3), pp. 207–227. Knack, S. and P. Keefer (1997), ‘‘Does social capital have an economic payoff: a cross country investigation’’, Quarterly Journal of Economics, Vol. 112(4), pp. 1251–1288. Levine, R. and D. Renelt (1992), ‘‘A sensitivity analysis of cross-country growth regressions’’, American Economic Review, Vol. 82(4), pp. 942–963.

Determinants of Growth in the MENA Countries

59

Mankiw, N.G., D. Romer and D.N. Weil (1992), ‘‘A contribution to the empirics of economic growth’’, Quarterly Journal of Economics, Vol. 107(2), pp. 407–437. Nehru, V. and Dhareshwar A. (1994), ‘‘New estimates of total factor productivity growth for developing and industrial countries’’, Policy Research Working Paper # 1313, World Bank. North, D.C. (1990), Institutions, Institutional Change and Economic Performance, Cambridge: Cambridge University Press, Cambridge. Nunnenkamp, P. (2004), ‘‘Why economic growth has been weak in Arab countries: the role of exogenous shocks, economic policy failure and institutional deﬁciencies’’, Journal of Development and Economic Policy, Vol. 7(2), pp. 1–18. Page, J. (1998), ‘‘From Boom to Bust- and Back? The Crisis of Growth in the Middle East and North Africa’’, in: N. Shaﬁk, editor, Prospects for Middle Eastern and North African Economies: From Boom to Bust and Back?, Economic Research Forum for Arab Countries, Iran and Turkey, Cairo. Political Risk Services (PRS) Group (2002), International Country Risk Guide, New York. Pritchett, L. (2001), ‘‘Where has all the education gone?’’, World Bank Economic Review, Vol. 15(3), pp. 367–391. Radelet, S., J. Sachs and J.W. Lee (1997), Economic Growth in Asia, Boston: Harvard Institute for International Development, Boston. Ramey, G. and V. Ramey (1995), ‘‘Cross-country evidence on the link between volatility and growth’’, American Economic Review, Vol. 85, pp. 1138–1151. Ridha, M.J. (1998), ‘‘Charting the future education and change in the Arab countries: a platform for the 21st century’’, Paper presented in the Joint Arab Monetary Fund, Arab Fund for Economic and Social Development Seminar on ‘‘Human Resource Development and Economic Growth,’’ May 17–18, 1998, Abu Dhabi, UAE. Sachs, J.D. and A.M. Warner (1995), ‘‘Economic reform and the process of global integration’’, Brookings Papers on Economic Activity, Vol. 10(1), pp. 118, Washington DC. Sachs, J.D. and A.M. Warner (1997a), ‘‘Natural resource abundance and economic growth’’, Center for International Development and Harvard Institute for International Development, Boston. Sachs, J.D. and A.M. Warner (1997b), ‘‘Sources of slow growth in african economies’’, Journal of African Economies, Vol. 6(3), pp. 335–376, Oxford. Senhadji, A. (1999), ‘‘Sources of economic growth: an extensive growth accounting exercise’’, IMF Working Paper WP/99/77. Srinivasan, T.G. (2002), ‘‘Globalization in MENA: a long term perspective’’, Paper presented at the fourth Mediterranean Development Forum, October 6–9, 2002, Amman, Jordan. UNESCO (1999), Statistical Yearbook, several issues, Paris. United Nations (1999), Human Development Report 1999, New York.

60

Samir Makdisi, Zeki Fattah and Imed Limam

World Bank (2000), Global Economic Prospects and the Developing Countries 2000, : World Bank Publication. World Bank (2002), ‘‘Better governance for development in the Middle East and North Africa,’’ available through internet at http://www.worldbank.org. World Bank, World Development Indicators CD-ROM, several issues Washington DC. World Bank, Global Research Project (GRP) Database Washington DC. Young, A. (1995), ‘‘The tyranny of numbers: confronting the statistical realities of the East Asian growth experience’’, Quarterly Journal of Economics, Vol. 110(3), pp. 641–680.

CHAPTER 3

A Reexamination of the Political Economy of Growth in the MENA Countries Hadi Salehi Esfahani Abstract Since the 1980s, economic growth in the Middle East and North African (MENA) region has been low and volatile. The problem has been attributed to bad economic policies such as excessive government interventions, large public sectors, restrictive trade policies, inefficient mass subsidies, and badly managed fiscal policies. This paper examines the political economy factors that have shaped those policies and their outcomes. It offers a framework based on the recent political economy literature to explore the deeper forces behind the puzzles of growth variations in the region, going beyond the claims that powerful interest groups have been blocking reform. It focuses on the reasons why interest groups and politicians may not have been able to adopt better policies and generate larger surpluses for them to reap. The usefulness of the framework is illustrated by applications to key political economy issues in MENA. The paper examines the mechanisms through which resource rents have affected government policies and economic growth in the region and argues that the impact has varied depending on other factors. The analysis includes the roles of authoritarianism, social heterogeneity, geopolitical rents and threats, human capital, and administrative and judicial institutions. It shows how resource rents and other factors have shaped government interventions and economic growth prior to the 1980s and how they made adjustment and growth costly later on to varying degrees. Keywords: political economy, growth, Middle East, North Africa JEL classifications: O1, O4, O5, O53, P16

Corresponding author. CONTRIBUTIONS TO ECONOMIC ANALYSIS VOLUME 278 ISSN: 0573-8555 DOI:10.1016/S0573-8555(06)78003-8

r 2007 ELSEVIER B.V. ALL RIGHTS RESERVED

Hadi Salehi Esfahani

62

1. Introduction Over the past two decades, economic growth in the Middle East and North African (MENA) countries has been generally low and relatively volatile (see Figure 1 and Table 1). This has been the case compared to the rest of the developing world as well as MENA’s own prior experience. A widely held view among the economists studying the region links the problem to bad economic policies adopted by the MENA governments. The view suggests that those policies can typically be characterized by extensive government intervention, large public sectors, restrictive trade policies, inefﬁcient mass subsidies (especially for energy and food), and precarious ﬁscal conditions (Henry and Springborg, 2001; Abed and Davoodi, 2003; Sala-i-Martı´ n and Artadi, 2003; Yousef, 2004). However, if such policies can indeed be held responsible for low growth, one has to wonder why many governments have opted for them. Moreover, to the extent that actual policies have been different from this description, there is still a question why those policies were selected and why they did not work in many cases. Figure 1.

Weighted average PPP GDP per worker in MENA and non-MENA developing countries

11

Log of PPP GDP per Worker in Constant 2000 US dollars

MENA Major Oil Exporters* Excluding Bahrain, UAE, and Yemen 10 MENA Major Oil Exporters* MENA Non-Oil Exporters** 9

8 Non-MENA Major Oil-Exporting Developing Countries†

Non-MENA Non-Oil Exporting Developing Countries 7 1965

1970

1975

1980

1985 Year

1990

1995

2000

2005

Note: The weights for averaging are countries’ PPP GDP. * Algeria, Bahrain, Iran, Kuwait, Oman, Saudi Arabia, the United Arab Emirates (UAE), and Yemen. ** Egypt, Jordan, Morocco, Sudan, Syria, Tunisia, and Turkey. Other countries are excluded owing to lack of sufﬁcient data. There is data for Lebanon after 1988, but adding it to the chart does not change the picture much for those years. y Angola, the Republic of Congo, Gabon, Nigeria, Trinidad and Tobago, and Venezuela. Source: Calculated based on World Bank’s World Development Indicators (2006).

A Reexamination of the Political Economy of Growth in the MENA Countries

Table 1.

Per worker GDP growth and volatility in MENA countries

a

Country

Average Per Worker GDP Growth Rate Period

Oil exporting Algeria Bahrain Iran Kuwait Oman Saudi Arabia UAE Yemen Non-oil exporting Egypt Jordan Lebanon Morocco Sudan Syria Tunisia Turkey Non-MENA oil exportersb Non-MENA nonoil exportersb

63

Standard Deviation for the Years Available

1955–1964

1965–1974

1975–1984

1985–1994

1995–2004

1.52 n.a. 6.95 n.a. 4.30 n.a. n.a. n.a.

4.71 n.a. 9.87 4.17 16.83 13.52 6.64 n.a.

2.20 6.30 1.66 8.81 4.48 4.37 4.66 n.a.

3.16 1.01 0.26 1.12 1.72 0.32 4.58 1.90

0.00 1.36 0.89 2.95 0.44 0.41 0.39 1.80

8.51 6.67 7.50 10.57 16.27 8.69 8.60 4.46

3.42 6.86 n.a. 0.67 1.28 6.78 6.05 4.69

1.35 0.62 n.a. 2.58 0.11 4.44 4.07 3.74

6.49 8.24 n.a. 1.95 1.28 3.01 1.94 2.17

0.78 3.13 8.08 1.60 1.80 1.35 1.14 1.53

1.42 0.72 0.59 0.36 2.90 0.26 1.87 1.86

3.56 8.99 18.55 4.43 5.71 8.01 3.52 4.74

5.32

1.94

2.84

1.08

0.08

6.79

4.43

2.98

2.14

2.89

3.18

4.61

n.a., not available. Source: Calculated based on World Bank’s World Development Indicators (2006), Augmented for years before 1965 by growth rate data from Penn World Tables, http://pwt.econ.upenn.edu/. a Djibouti, Iraq, Libya, Oman, Qatar, and West Bank and Gaza are not included owing to lack of sufﬁcient data. b Weighted averages, with each country’s PPP GDP serving as its weight. Oil exporters consist of Angola, the Republic of Congo, Gabon, Nigeria, Trinidad and Tobago, and Venezuela.

A number of political economy studies of the MENA countries have sought to answer the above question by searching for politically organized groups that ﬁnd growth-enhancing policies inimical to their interests and, thus, try to block the adoption of such policies (see, e.g., Richards, 1991; Sadowski, 1991; Waterbury, 1993; Henry and Springborg, 2001). However, the answers proposed in these studies are often incomplete because they do not explain why in the slowgrowing countries the interest groups and politicians cannot ﬁnd ways to adopt better policies that generate larger surpluses for them to reap. Another popular explanation of bad policies and growth failures is based on the so-called ‘‘natural resource curse’’ hypothesis. The argument is that large

64

Hadi Salehi Esfahani

oil rents can induce a voracious rent seeking that erodes institutional development through effects such as massive corruption, patronage, etc. (Leite and Weidmann, 1999). This effect is believed to undermine growth because it hampers the government’s ability to implement efﬁcient policies. The problem is further compounded by the oil-rich countries’ need to manage the revenue volatility and the real exchange appreciation that oil rents entail (Ersel and Kandil, 2007; Makdisi, et al., 2007). While these effects have often been detected empirically in the data, their theoretical foundations have not been clear. In particular, when one observes countries such as Norway manage their oil rents well without institutional deterioration, it becomes clear that there is a need to explore the conditions under which the explanation applies (Sala-i-Martin and Subramanian, 2003). A third interpretation of MENA’s growth pattern combines the interest group and resource curse explanations into a more structured perspective. This approach argues that the rising natural resource rents prior to 1980 allowed high investment and high growth while strengthening redistributive policies in the form of implicit ‘‘social contracts’’ between the rulers, the public, and various interest groups in MENA societies. Later, those social contracts came to haunt MENA economies when resource rents declined. The reason was that the region’s politicians were still expected to deliver on their distributive promises in exchange for public acquiescence to their rule. This may have severely constrained the room for the reorientation of policies toward higher efﬁciency and growth (Yousef, 2004). However, this view still leaves the details regarding the nature of social contracts and the sources of their variation across countries and over time unspeciﬁed. This is important because the view does not show why MENA polities did not continue to demand growth in the past quarter century and often settled with inefﬁcient redistributive deals. Even if the initial contracts had a redistributive character, one wonders why their bases did not shift toward growth, especially when resource rents diminished. This is particularly puzzling because social contracts were also common among East Asian countries, which succeeded with little natural resources by emphasizing shared growth rather than pure redistribution (Campos and Root, 1996). Another factor often mentioned as a cause of slow economic growth in MENA is the large amounts of labor and public funds devoted to military (Lebovic and Ishaq, 1987; United Nations Development Programme (UNDP), 2002; Abu-Bader and Abu-Qarn, 2003). Indeed, as Figure 2 shows, the MENA countries stand out among developing countries in terms of the military’s share in employment and GDP. This is often attributed to the presence of domestic and international conﬂicts, especially the Arab–Israeli confrontations. Such conﬂicts and their associated military expenditures are believed to have contributed to economic risks and shortage of funds in the region, thus reducing the incentives and the ability to invest and enhance productivity. They may also have hindered the institutional developments needed for improved policymaking and economic performance. However, some have argued that given the presence

A Reexamination of the Political Economy of Growth in the MENA Countries

65

Military Expenditure as Percent of GDP

Figure 2. Military expenditure, military personnel, and GDP growth in developing countries (1985–2004 averages) KWT

20

15

OMN SAU

10

ISR

JOR

YEM

SYR ARE

5

IRN

DZA

BHR MARTUR EGY

LBN

TU N

0

Military Personnel as Percent of Labor Force

-4

-2 0 2 4 6 1985-2004 Average Growth Rate of GDP per Worker

JOR

10

8

10

8

10

SYR

OMN

5

BHR

ARE

LBN SAU T KWDZA

IRN EGYTUR MAR YEM TUN

0 -4

-2 0 2 4 6 1985-2004 Average Growth Rate of GDP per Worker MENA Oil-Exporting Non-MENA Non-Oil-Exporting

MENA Non-Oil-Exporting Non-MENA Oil-Exporting

Source: World Bank, World Development Indicators (2006).

of conﬂicts, military expenditure may have played a positive role by bringing about stability and order (see especially the recent panel data analysis by Yildirim, et al., 2005). Again, it is clear that there is a need to further explore the political economy factors that give rise to conﬂicts in the MENA region and shape their impacts on economic growth.

66

Hadi Salehi Esfahani

This paper proposes a framework based on the recent political economy literature that offers ways of ﬁnding deeper answers for the puzzles of growth variation and slowdown in the MENA region. The approach presented here is part of a growing number of perspectives on the political economy of growth that seek to explain differences in growth-related policies and institutions across countries. Recent examples of such work that also provide surveys of the literature are Castanheira and Esfahani (2003), Rodrik (2003, 2005) and Acemoglu et al. (2005a). Of course, the political economy literature is vast and the references given here are meant to be representative rather than exhaustive. The present paper reﬁnes the existing studies by integrating their ﬁndings around a key theme – namely, incomplete contracting – and by incorporating additional effects implied by that theme into the framework. The value of the framework is highlighted by applying it to some of the major growth issues of the MENA countries. The purpose of this application is not to ﬁnd a deﬁnitive explanation for MENA’s growth puzzle. Rather, the aim is to suggest that the proposed political economy framework is a promising approach for further case study and cross-country research on the subject. Some results of such case studies are synthesized in Esfahani (2006). Section 2 discusses the analytical framework. Section 3 examines a number of policies and institutions in the MENA countries to show how the framework can shed light on the forces that may cause inefﬁciency in the interactions among the politicians, interest groups, and the public. Section 4 concludes.

2. A framework for analyzing the political economy of growth The political economy literature studies the role of collective action processes (i.e., interest group activity, policymaking institutions, and the like) in resource allocation and rent distribution. One part of that literature focuses on economic growth and examines the impact of the collective action processes on the incentives of economic agents to invest and to improve productivity. That is, the key question in the political economy of growth is how much of an economy’s potential surplus (i.e., rents and quasi-rents or the value produced in excess of the recurrent costs of production) is realized and guided toward growth-enhancing activities. While political economy in general is concerned with growth as well as distribution of the society’s surplus, from a growth standpoint, distribution matters only to the extent that it affects the surplus being channeled toward productivity-enhancing investments. To keep things simple, I identify three main categories of agents in the political economy systems: the public (or the population at large who act as voters when there are elections or other opportunities to show public approval or disapproval of policies or policymakers), interest groups (organized groups that inﬂuence policy decisions on a systematic basis but do not control it directly), and political elites (or, for short, politicians: elected or self-proclaimed

A Reexamination of the Political Economy of Growth in the MENA Countries

67

policymakers, administrators, and political parties).1 Politicians are interested in expanding their control over the government apparatus, which requires public and/or interest group support. Each member of the public and each interest group wants to maximize its net beneﬁts from the economy and can offer its support to politicians in exchange for receiving the beneﬁts induced by the policy. This implies that in choosing economic policies, politicians are largely motivated by the relative ability of interest groups and various segments of voters to support them. Struggles to capture a larger share of the economic surplus can arise both within and across interest groups and voter segments, and the eventual allocation is determined by the rules of the political game. The main premise of the approach developed here is that contracting problems are the key impediments to economic growth. The argument is that it is the incompleteness of contracts, rather than the presence of social contracts, that has hindered MENA’s growth. Contracting problems include asymmetric information and moral hazard as well as the difﬁculties and costs of writing complete contracts owing to complexities in foreseeing and specifying all possible contingencies in dynamic human transactions. As is well known from the Coase Theorem, in the absence of contracting problems, members of a society should be able to motivate each other and remove all inefﬁciencies in investment and production. This should render the largest social surplus that is technologically possible, and especially provide opportunities for rapid growth and catch-up when the country is not already at the production possibility frontier. Thus, when growth remains low in a developing country, there must be contracting problems that prevent the realization or proper allocation of the maximum surplus. The claim is that the more serious the contracting problems are, the lower the growth rate is likely to be. The extent of contracting problems in a country depends on the characteristics of its resources, politics, and institutions. Here I deﬁne these concepts and discuss their relevant characteristics. 2.1. Resources The available resources – including natural and human resources, physical capital, and technology – matter because they shape the possibilities for production and the nature of contracting and incentives required to reach efﬁciency. In

1

One can further subdivide these categories to account for the relative powers or the ranks of the ofﬁcials (e.g., ministers vs. civil servants, the dominant party vs. the others), or the real power of different interest groups (e.g., unions vs. the army or other groups present in the country). These divisions will be discussed below where needed. The main categories highlighted here capture the functional differences in their access to power: voters have power at election times, and much less power when elections are away. Instead, interest groups can maintain their inﬂuence (sometimes by monetary means, sometimes by demonstrations and strikes) on a more regular basis. Political elites, on their side, directly control the policy levers, favoring one or another group.

68

Hadi Salehi Esfahani

political economy studies, the nature of available resources normally receives attention only when natural resources play a major role in the economy. This aspect, of course, is particularly important for the MENA countries that are endowed with oil as well as the others that are indirectly affected through their exchanges with oil-exporting countries (e.g., via foreign aid, capital ﬂows, and labor migration). However, this often overshadows the fact that human resources are also important because they help bring together other resources and make them productive by applying technology. Moreover, human capital can facilitate the identiﬁcation of growth problems and the design and implementation of effective solutions for them. Most MENA countries face shortages of human capital in a variety of areas. Although there have been major efforts to expand education in the region, the achievements have been uneven partly because the policy circles themselves are often short of the expertise necessary to design and implement productive policies. The key resource characteristics that need to be recognized and examined for understanding growth are those that determine how much rent and quasi-rent can be generated by the available resources, how elastic their supplies are, and how difﬁcult it is to contract over the process by which they become available for production. For example, natural resources tend to have inelastic supplies and often generate large surpluses relative to costs of production, but the degree of contracting difﬁculty varies across different types of resources. For example, forests can be quickly cut, but oil requires long-term investment in exploration and extraction (Deacon and Bohn, 2000). The supply of human capital, in contrast, may be relatively more elastic and contracting over its delivery often faces major imperfections, especially in activities that require creativity and innovation (Hart, Shleifer, & Vishny, 1997). For physical capital, the supply is often elastic and contracting problems exist ex ante, but those elements diminish ex post once capital is in place. 2.2. Politics Politics shapes the nature of group conﬂicts that need to be resolved in the collective action process. The diversity of preferences and fragmentation of the society (e.g., differences in language, religion, ethnicity, resource endowments, and others that lead to cleavages among groups) raise the likelihood of difﬁculties in making deals both at the micro and macro levels. A more contentious set of interest groups may exert greater effort to keep each other at bay and, as a result, may raise bigger obstacles for the adoption of collectively beneﬁcial policies by the government. In particular, such demands can obstruct appropriate regulations (or give rise to excessive and inappropriate regulations) in the economy and, thus, result in signiﬁcant inefﬁciency and slow growth (Alesina and Drazen, 1991; Alesina, et al., 2003). In contrast, when there is harmony of interests or when the dominant groups have sufﬁciently encompassing interests, selected policies are more likely to be efﬁcient (McGuire and Oslon, 1996).

A Reexamination of the Political Economy of Growth in the MENA Countries

69

It should be noted that although the boundaries of interest groups and the extent of their heterogeneity may be given in the short and medium runs, they do not necessarily remain ﬁxed factors over decades, even when groups are formed based on language, religion, or ethnicity. In fact, many interest groups form around organizations, where the members develop common interests and speciﬁc capital associated with their organizations, as in industry-speciﬁc lobbies, the government bureaucracies, or the military. The dynamics of such groups and their interactions can be quite rich and complex, particularly because they strive to inﬂuence government policies and expenditures that may in turn affect their own position and strength (see Acemoglu and Robinson, 2001, and references therein). Analyzing the role of heterogeneity in political and economic performance raises a number of key issues: which dimensions of diversity matter, in which ways, and how those dimensions must be measured. Until recently, the common practice was to use Herﬁndahl indices of fractionalization for ethnic, linguistic, and religious aspects of heterogeneity, which are more observable and stable over time (see Alesina et al., 2003). This index is often found to be negatively and signiﬁcantly related to GDP growth. However, it does not appear to be well correlated with most aspects of social and political conﬂict that are often claimed to be the links between fractionalization and growth. For this reason, to shed light on the underlying connections between heterogeneity and political economy outcomes, a number of studies have offered new measures that focus on other aspects of heterogeneity. In particular, the polarization indices developed by Esteban and Ray (1994) and Montalvo and Reynal-Querol (2005) try to capture the fact that when fractionalization is very high, in a way the society may look relatively homogeneous since the differences among groups may become smaller or their conﬂicts may be more localized and less consequential for the system as a whole. (It should be pointed out that these polarization indices focus on the probability of intergroup alienation and within-group identity, not on the actual intensity of differences and loyalties.) Montalvo and Reynal-Querol (2005) show that their index explains within-society conﬂict much better than fractionalization, though it is not well correlated with cross-country growth. The upper panel of Figure 3 compares two indices for fractionalization and polarization for 134 countries and highlights the non-linear relationship between the two. Heterogeneity factors are not conﬁned to ethno-linguistic and religious polarization. Other notable sources of diversity are ideological and cultural orientation (e.g., modernist vs. traditional), asset ownership, location (region, urban vs. rural), industry, and organizational afﬁliation (e.g., the military or political parties and cliques). Each of these factors, which are at least partially endogenous, may have rich interactions with other elements of heterogeneity. For example, political parties may add fuel to the ﬁre of intergroup rivalry or may help enhance national identity and, thereby, reduce ethno-linguistic and religious alienation, class conﬂict, etc. Naturally, the manner in which the economy’s resources are applied and the institutional mechanisms needed to achieve good performance are different in those varying political conditions.

Hadi Salehi Esfahani

70

Figure 3. Fractionalization, polarization, and GDP growth in developing countries fractional index is the average of ethnic, linguistic, and religious Herfindahl indices for the three dimensions of heterogeneity; polarization index is the average of polarity of ethnic and religious distributions according to the measure proposed by Montalvo and Reynal-Querol, 2005; averaging is done as a rough aggregation method to save space; similar relations emerge when individual components are used. 1 KWT

Polarization Index

0.8

IRQ BHR ARE JOR

0.6 MAR

0.4

EGY

IRN SYR OMN

DZA

0.2

TUR SAU

TUN

0 0

0.2

0.4

0.6

0.8

Fractionalization Index

1985-2004 Average Growth Rate of GDP per Worker

10 8 6 4 2

TUN

0

TUR

EGY OM N SYR

YEM MAR IRN

BHR

SAU DZA

-2

JOR

ARE

KWT

-4 0

0.2

0.4 0.6 Polarization Index

0.8

MENA Oil-Exporting Non-MENA Non-Oil-Exporting

MENA Non-Oil-Exporting Non-MENA Oil-Exporting

Regression Line for Non-MENA

Regression Line for MENA

1

Source: Alesina et al. (2003), Montalvo and Reynal-Querol (2005), and World Development Indicators (2006).

A Reexamination of the Political Economy of Growth in the MENA Countries

71

In the MENA region, the extent of preference diversity and potential conﬂicts vary greatly. However, by the fractionalization measure of ethno-linguistic and religious diversity, the MENA countries generally do not rank very high. So, to the extent that this factor matters for growth, they should fare well. But, by the polarization index, parts of the MENA region lie in the upper half of the sample, and even that may understate the actual intergroup tensions in some countries where societies are divided into contending groups with distinct fault lines, as in Algeria, Iraq, and Lebanon. Concentration of land and capital ownership has also served as a source of conﬂict, resulting in many conﬁscation and redistribution episodes by national leaders who sought to broaden their public support or dispossess their political rivals, or both. The cases in point are the land reforms and nationalizations in Egypt during 1952–1963, in Iraq, Syria, and Tunisia in the late 1950s, in Algeria after independence in 1962 and again in the early 1970s, and in Iran in 1963 and again after the 1979 Revolution. Of course, the intensity and role of group interactions depend on prevailing ideologies at the group and national levels. In particular, as North (2005) points out, nationalist ideologies can play important roles in mitigating potential conﬂicts arising from polarization. In MENA, at many points in time, political leaders have managed to use nationalist ideologies to overcome or suppress major internal conﬂicts and, thereby, facilitate mobilization of economic resources. These efforts seem to have been most successful in situations when new regimes have come to power following long wars or crises, where the need for establishing order has provided potent arguments and drives for mufﬂing internal conﬂicts, at least for a while. Although order in these cases has typically come about through dictatorship, sometimes they have been able to strengthen national identities and to dilute religious or ethnic conﬂicts on a long-term basis (e.g., Turkey under Ataturk and Iran under Reza Shah). In the case of Arab countries, similar ideological tools have also been used for forming cross-country movements. However, these movements have rarely gained strength (with the exceptions of Baath and Nasserist movements) and even when they have gained some force they have intensiﬁed conﬂict at other fault lines and have ultimately disintegrated. 2.3. Institutions Institutions are the rules that assign roles to individuals in the society and structure their interactions. These rules range from informal cultural norms to formal legal and political procedures governing the country. The allocation of decisionmaking powers and the constraints that institutions place on individual behavior can help or hinder the potential for resolving conﬂicts and reaching efﬁcient and enforceable deals. Institutions have many aspects and dimensions. To make the analysis manageable, I focus on governance institutions because they play fundamental roles in alleviating or exacerbating deﬁciencies in other rules. Furthermore, I concentrate on institutional aspects that affect representation,

72

Hadi Salehi Esfahani

commitment, and coordination in the interactions of various groups through the government. These three dimensions of governance institutions are crucial for the extent to which the government can overcome contracting problems and help various groups strike efﬁcient deals that take the economy closer to its production possibility frontier. The following discussion expounds this point. 2.3.1. Representation For efﬁcient collective decision making in a heterogeneous society, various segments of the population must be somehow represented in the government; otherwise important information about the preferences of some groups may not be incorporated in government decisions. This can lead to distortions in the production incentives, both among those groups as well as others that interact with them. For example, when the government ignores the educational needs of one part of the population, they may be discouraged from engaging in productive activities and may become foot soldiers in rivalries among more powerful groups. Representation can be viewed as the set of political or social institutions that enable various groups to conclude effective contracts with the politicians, offering support or acquiescence to political power in exchange for policies that address those groups’ concerns. Effectiveness of such institutions depends on the capability of various parts of the population to ensure that their demands are given due consideration and weight by the policymakers. In the economic growth literature, representation is often identiﬁed with electoral democracy, which does not seem to have a direct relationship with growth (Barro, 1996; Acemoglu et al., 2005b). Although democracy certainly enhances representation, it also entails many other institutional aspects with additional multifaceted effects, as we will see below. Besides, having a chance to vote for politicians is only one, often imperfect, mechanism for the population to inﬂuence policymakers. In fact, in addition to the opportunity to react to policy outcomes, thorough representation requires channels that connect the public to the process of policymaking through political, professional, and social organizations. These latter channels – whether trade unions, business associations, religious organizations, or tribal structures – may also exist and provide some representation even in the absence of elections and other formal democratic institutions. In the MENA countries, representation mechanisms are often believed to be weak. This is particularly an issue in the region’s larger, more diversiﬁed countries that lack democratic institutions. There also seem to be differences in representation across monarchies and republics in the region. In monarchies, the rulers depend more on traditional and family relations to connect with broader segments of the population. In contrast, in countries where republics have been established through military coups, revolutions, or national liberation movements, preexisting power relations have crumbled and the political elite have had to rely more on new forms of political and administrative organization to manage their relations with the population at large (Owen, 2004). However, since these organizations

A Reexamination of the Political Economy of Growth in the MENA Countries

73

have been initiated from above, they have typically served more as means of political control by the elite rather than representation (Ayubi, 1995). 2.3.2. Commitment Commitment consists of any cost that policymakers must bear if they decide to reverse an adopted policy in ways that take away the returns to investments made in response to that policy (Levy and Spiller, 1996; Esfahani, 2000). The presence of such costs is crucial for economic growth because otherwise, important investments might be deterred. The reason is that most investments become largely sunk once they are in place, so their quasi-rents can be easily taxed without much economic consequence in the short run. If the politicians can change policies with no cost (or some net beneﬁt) to themselves, they may ﬁnd it convenient to ignore the losses of the producers. However, this possibility discourages entrepreneurs from investment in the ﬁrst place. To ensure commitment, a country needs systematic checks and balances and separation of powers so that the decision makers can turn their policies into enforceable contracts (North and Weingast, 1989). However, a dictator with a secure, long-term position may also be able to offer some commitment to efﬁcient policies based on his incentive to maintain a reputation for future dealings. In recent years, most of the research on the role of institutions in economic growth has focused on the commitment issue – often under the rubrics of property rights’ security and constraints on governments – as the key determinant of long-run income levels. (See, in particular, the large volume of work spun by Acemoglu and his associates; e.g., Acemoglu et al., 2001, 2005a, and b and Acemoglu and Robinson, 2005. See also Rodrik, et al., 2002, among others.) As in the case of representation, some authors associate commitment capability with democratic rule. However, the more recent contributions to this literature show that in underdeveloped economies, limitations on democracy may enable the elite to protect their investments against the taxation power of the poor majority, thus ensuring the incentives needed to initiate investment and growth (Acemoglu, 2005). Of course, once the economy develops to some extent, further growth is likely to require transition to democracy. Commitment through non-democratic mechanisms (reputation of the autocrat and rent sharing with the elite) has been prevalent among the MENA countries, though it has not always worked well, especially in the cases of detailed policy matters. Examples of such cases will be discussed in Section 3. 2.3.3. Coordination Coordination is another crucial institutional feature needed for minimizing resource waste because distributive struggles typically create a ‘‘tragedy of the commons’’ problem where the parties involved tend to overexploit the rent source over which they compete or overuse their own resource to secure a larger share

74

Hadi Salehi Esfahani

(Persson and Svensson, 1989; Alesina and Tabellini, 1990). For example, when government budgets are made in parliaments where the lawmakers can independently and freely propose amendments, total expenditure and the deﬁcit tend to be inefﬁciently large (von Hagen, 1992; Alesina and Perotti, 1995; Velasco, 2000). The reason is that in such a situation, the cost of an additional dollar spent on the favorite project of each policymaker is shared by the population at large and, as a result, is not fully internalized. Similar problems exist in trade policy formation and many other public policies. Coordination across policy areas (e.g., coordinating trade, labor, and ﬁscal policies) tends to pose even harder problems and could be the source of major inefﬁciency in the economy if the country lacks appropriate institutional mechanisms. Of course, coordination would not arise as a problem to the extent that decision making on public policy could be decentralized without external effects – e.g., in a market-like fashion. However, this is not possible in many areas of public policy; hence the need for other institutional mechanisms to bring about coordination. One mechanism is ex ante constraints (e.g., balanced-budget laws) that force everyone to act within limits, though such constraints are typically too inﬂexible and inefﬁcient because specifying ex ante rules that anticipate all possible future situations is very rarely practical. A more ﬂexible mechanism is ‘‘delegation,’’ where one or a few individuals make the aggregate decisions that bring everyone into line. This obviously implies some centralization and, to be effective, requires proper incentives for those in charge of the aggregate decisions. An intermediate mechanism is ex ante limits that are renegotiated on a regular basis and only delegate the task of enforcing the agreed limits. Of course, the effectiveness of all these arrangements varies depending on the nature of other political institutions such as political parties. For example, coordination is much easier when a few parties are in a position to aggregate the population’s interests in a decentralized way and negotiate them with each other. In the MENA countries, high degrees of power concentration have often facilitated coordination over major policies. But, those institutional arrangements have not been suitable for more detailed actions of government agencies because processing the details of minor policy matters quickly becomes overwhelming for the center. More on this will follow. In making governance work effectively, capabilities in all three institutional dimensions discussed above – representation, coordination, and commitment – are needed. Indeed, there are complementarities among the three dimensions. In particular, higher commitment capability tends to be complementary with strength in both representation and coordination because it allows the government and interest groups to coordinate their actions over time as well as across policy areas. In addition, it enhances the opportunities for the politicians to assure their constituencies that their interests will continue to be represented in the policymaking process. Conversely, better representation and coordination help ensure that the adopted policies enjoy broader support and greater efﬁciency and, therefore, are less likely to be reversed in arbitrary manner and without due process. Broader representation of the population in the

A Reexamination of the Political Economy of Growth in the MENA Countries

75

policymaking process can also help coordination to be achieved with more information and over larger sets of policy issues. However, there are some tradeoffs among the three institutional dimensions as well, especially between representation and coordination. Better representation can make coordination more difﬁcult because it requires the decision-making process to take account of more political demands. Conversely, coordination mechanisms may constrain the possibilities for broad representation. Of course, such tradeoffs and complementarities vary across countries depending on their economic and institutional conditions. For example, in a country facing chaos, establishing some mechanism for coordination is likely to be crucial and sacriﬁcing it to ensure better representation may come at a huge cost in terms of economic growth. In contrast, when the basic coordination problems have been solved, weak representation can become a key source of instability and inefﬁciency. Another example is the increased value of commitment capability when the government does not have access to cheap resource rents and cannot create employment and promote growth through subsidized private or public investment (Esfahani and Toossi-Ardakani, 2002). Some countries (e.g., those with less political polarization and better rule-of-law tradition) are able to resolve the tension among the institutional functions relatively well and reach high levels of capability in all dimensions. In other countries, long-term growth performance is constrained by the fact that some institutional capabilities are missing and trying to build them comes at a high cost in terms of other capabilities (e.g., forcing coordination at the cost of representation or vice versa). 2.4. Dynamics in the political economy of growth The above discussion offers a framework for understanding what the growth potentials of a country are at a point in time given its resources, politics, and institutions. However, the latter factors are by no means static or exogenous. Changes in demographics and resource characteristics, partly as a result of the growth process itself, inﬂuence the relative powers and interests of various groups, which lead to changes in politics. Those forces in turn may make the existing institutions obsolete and prompt interest groups to seek changes in rules. In analyzing this process, it is important to note that the existing power structure at each time has important implications for the set of likely institutions that emerge subsequently. As Acemoglu et al. (2005a) have emphasized, the desire of dominant groups to maintain their powers plays an important role in enhancing or hindering institutional change. Of course, this effect should be weighed against the shifts in potential balance of power brought about by developments in resources. Moreover, there are important institutional and political changes that come about not through the desire of politically active groups, but rather as byproducts of unintended socioeconomic developments. For example, the turn to religion as a political tool in recent decades could not have happened without deep socioeconomic developments that made such a

76

Hadi Salehi Esfahani

trend feasible, in some instances in clear opposition to the desire of the ruling elite. Another important issue in the analysis of growth dynamics is that institutions are by deﬁnition costly to change. Indeed, rules that are easy to change cannot become institutions that constrain the behavior of actors in an economy. In other words, the institutions that matter must be rules that are difﬁcult to change either because they are self-enforcing or because they are supported by other institutions that require substantial effort and coordination to change. A key implication of this observation is that institutional change comes about only when shifts in resources and politics have raised the cost–beneﬁt ratio of the status quo for a sufﬁciently large group of actors. A further consideration in growth dynamics is the fact that the onset of economic growth does not require deep and extensive institutional reform, as Rodrik (2005) and Glaeser (2004) have noted. In many instances, small and limited policy changes have been observed to initiate remarkable growth episodes. One crucial factor that may have been at work in such cases is innovation in business– government and intragovernment relations in ways that made contracting among them more complete via performance indicators that were more easily observable to both businesses and policymakers. For example, a key reason for the success of export promotion policies of Korea and Taiwan in the 1960s and 1970s may have been the use of net foreign exchange earnings as an indicator of performance. These policies offered rewards to domestic entrepreneurs based on a criterion that was far more observable than the deals given under import substitution policies (which at least in theory were based on their contribution to production capacity and foreign exchange savings). Although both types of policy offered rents to domestic producers and could help mitigate their investment constraints (especially ﬁnance and foreign exchange problems) relative to their foreign competitors, export promotion policies proved more successful because they offered more complete ‘‘policy contracts.’’ The latter policies were also more effective in enabling the entrepreneurs search for their sources of comparative advantage and break into international markets. Of course, this was not possible in all industries because not all industries could be expected to become exporters and, besides, government subsidy funds were limited. As a result, policymakers maintained import restrictions for quite some time on a wide range of products to protect many industries and collect public funds for subsidizing the rest. Interestingly, the innovation in trade policies seems to have helped address some governance problems as well. By observing export performance and collecting feedback from exporters on their experiences with public services, policymakers could assess the performance of various bureaucracies better and motivate the administrators more effectively (Amsden, 1989; Wade, 1990). Similar effects may help explain the success of Turkey and Tunisia after their moves toward export promotion in the 1980s. Another mechanism through which a move toward export promotion may induce more extensive change is the incentives it creates for producers to develop export interests and reduce their stakes in import substitution. This can shift the

A Reexamination of the Political Economy of Growth in the MENA Countries

77

political calculus of trade policy in favor of more liberal trade, especially when the rising export revenues make it easier to import and starving the domestic demand for imports becomes increasingly unattractive (Esfahani and Squire, forthcoming). The import liberalization trends in Turkey and Tunisia during the 1990s following their shift toward export promotion seem to ﬁt this pattern well. A more general version of the above process is the so-called ‘‘dual-track’’ reform strategy, whereby some marginal areas in a controlled economy are ﬁrst liberalized and allowed to expand without generating much risk in the rest of the economy. The success of the liberalized part then generates new resources and stronger incentives for reform in the remaining parts, eventually improving the entire economy in a politically palatable and low-risk manner. China has come to represent a highly successful case of this strategy (Lau et al., 2000), though the MENA countries were indeed pioneers in adopting this approach, starting with Tunisia in 1969 followed by Egypt and Syria in 1974, and Algeria in 1978–1979 (Owen, 2004, pp. 115–119). However, unlike China, lack of emphasis on export promotion in MENA limited the beneﬁts of the approach and turned into a very gradual process of change. Tunisia later shifted to export promotion later on in the mid-1980s and reaped more beneﬁts, probably because of its lack of resource and geopolitical rents.2 While policy innovations can initiate strong growth episodes, sustaining high growth in the face of changes in circumstances ultimately requires good institutions. Such institutions may, of course, be built as part of a burgeoning growth process (Rodrik, 2005). But, this point raises another question that has not received much attention in the literature. The circumstances under which small policy changes trigger a virtuous circle of growth and institution building seem to depend largely on informal institutions (e.g., social and political culture). Indeed, it is the informal rules and norms imbedded in culture, religion, etc., that ultimately help formal institutions to function and evolve. In particular, without informal institutions that keep the fabric of society together, it is unlikely that reforms in formal institutions can come about. Some research along these lines has started under the rubric of social capital, though further insights from that ﬁeld await theoretical development and more detailed characterization and quantiﬁcation of informal institutions. However, it is important to remain aware of the critical roles that such institutions play in the growth process. Finally, it should be noted that the multifaceted interactions of resources, politics, and institutions imply that the key factors constraining economic growth may vary greatly according to place and time. To identify the cause of stagnation and appropriate solutions to a country’s growth problems at a particular time, one needs to focus on speciﬁc circumstances, although it is crucial

2

The small size of its domestic market may have also rendered inward-looking policy less practical, though this effect does not seem to have been important for Syria. Also, China’s large domestic market does not seem to have deterred its commitment to export promotion.

78

Hadi Salehi Esfahani

to work within a theoretically consistent framework that can inform the choice of relevant factors and suggest hypotheses to be examined. Such studies can help improve the growth process by providing better information to the political and economic actors in each country and, thereby, reducing their contracting problems. This observation also suggests a solution to the dilemma noted by Pritchett (2003) regarding the existing political economy models. He points out that those models are incapable of generating useful policy implications because they assume that everyone is already optimizing subject to the constraints of the system. In other words, any proposed policy improvement must be ignoring or violating some constraint. However, in a world with a myriad of dimensions, even optimizing agents may leave many alternatives unexplored, especially when it comes to the options that require collective action. Just the perception that some forms of coordination are costly may give the players sufﬁcient reason to ignore a host of possibilities, some of which may be potentially productive. The players in such complex games may not know exactly what it is that they must strive to know. Systematic research and lessons from experience can lower the cost of considering such alternatives and render them feasible; hence a role for political economy research to contribute to economic and institutional reform. The approach outlined above suggests that to understand the determinants of growth, one has to identify the political, institutional, and economic characteristics that are most relevant for growth-related contracting in a country. Then, the task is to analyze such characteristics and to show how their evolution has given rise to the country’s pattern of economic growth. Below, I provide an overview for the application of this approach to the MENA countries. 3. MENA’s growth experience: the roles of resources, politics, and institutions This section applies the above political economy framework to examine the growth consequences of a number of economic, political, and institutional characteristics that are of broad concern across the MENA countries. I start with a discussion of the role of natural resource rents and then examine the geopolitical factors, domestic political divisions, and governance institutions and human resources. The stylized facts used here regarding the nature of the economic situations and policy responses are largely drawn from the other chapters in this volume. In particular, the empirical connections between those conditions and economic growth are based on Ersel and Kandil (2007); Makdisi et al. (2007); and Salehi-Isfahani (2007). 3.1. Natural resource rents One of the most notable features common to many MENA countries since the mid-20th century is the richness of its natural resource endowments, specially oil and natural gas. Even countries that lack such resources beneﬁt from the region’s endowments indirectly through aid from richer countries or remittances from their migrant workers. While there are interesting issues concerning how

A Reexamination of the Political Economy of Growth in the MENA Countries

79

oil resources are developed and exploited in the MENA countries, they are far outweighed by another question: Why have MENA countries not always used their natural resource rents to help their economies grow faster systematically compared to their less-endowed counterparts in the developing world? Here, I explore the possible answers to this question as an illustration of how the incomplete contracting framework may be shedding light on the subject. To understand why faster growth is not a systematic consequence of natural resource endowments in MENA, we need to ﬁrst understand what the MENA countries do with the rents. The ﬁrst notable pattern in this regard is the relatively low rates of taxation of non-oil income in oil-rich countries. Since natural resources in these countries are generally under public ownership, abundance of the related rents has reduced the need for taxation. Although statutory tax rates are in some cases rather high, especially for the high-income brackets, the typical direct and indirect taxes actually paid remain generally low and ﬁnance a relatively small share of the government budget. Interestingly, the low tax rates have not translated into encouragement for investment and growth in many MENA countries. This seems to be partly because of uncertainty in the net returns to production, which could be related to ﬂuctuations in oil revenues. We will explore the reasons for this below. Another important part of oil rents is spent on large government bureaucracies as well as ﬁnancing infrastructure and production projects. The latter part could, in principle, be a growth stimulant. However, that role is often undermined owing to the inefﬁciencies in investment and operation of public enterprises (PEs) and infrastructure. Finally, the bulk of the remaining revenues are used in the form of transfers and mass subsidies. As Esfahani (2003) shows, in oil-rich countries of MENA the subsidy on gasoline consumption alone ranges between 1 and 5 percent of the GDP. Subsidies are also offered for a variety of other items, especially other forms of energy, staple foods, and credit. This use of resource rents has had limited potential for generating economic growth. So, why do MENA countries often use their resource rents in these ways rather than channeling them more effectively toward growth? From a contracting perspective, inefﬁciency arises when the politicians who need to elicit support from the public or politically inﬂuential interest groups cannot make them the residual beneﬁciaries of efﬁcient uses of public funds. For example, the funds subsidizing redundant workers in PEs could be used for creating more productive jobs, but the workers may not consent to such reallocation if they believe that they may not be the beneﬁciaries of those jobs, or if they suspect that the funds may simply be diverted to other inefﬁcient uses. Similarly, the public may understand that mass subsidies on fuel are inefﬁcient, but if it cannot trust that the politicians will use the saved subsidies to produce a better outcome, broad support for the subsidies may remain strong (Esfahani, 2003). This may not be a concern for democratic regimes that are under close public scrutiny. It may also not matter for authoritarian regimes that can rule with narrow support, at least

80

Hadi Salehi Esfahani

for a while, as in the case of the Shah’s government in the 1960s, which raised the price of fuel and generated substantial funds for its projects. However, most MENA governments rely on some form of public support or acquiescence, without being fully trusted. As a result, when they have access to rents, they feel the need to offer subsidies because the public does not have high hopes of seeing those funds applied toward more efﬁcient alternatives of broad interest. Even if the government intends to make good use of the funds, specifying the details of what must be done and assessing the outcome are often extremely difﬁcult because the factors involved and the possible states of the world are rarely practical. As long as there are opportunities for misallocation or embezzlement of rents, the public is likely to prefer upfront transfers through inefﬁcient subsidies or expenditures, rather than allowing the funds to remain in the hands of the politicians and waiting for possible efﬁcient uses. Upfront subsidies offered in the form of low food and energy prices also act as effective coordination mechanisms for the public to react collectively to policy changes. This may explain why the removal of subsidies has often triggered widespread riots, forcing governments to rescind price hikes at least partially in many cases, as in Iran in 1969, Egypt in 1977, Morocco and Tunisia in 1984, Jordan in 1989 and 1996, and Yemen in 1998 and 2005. It is also notable that the expanded role elections in Iran after the 1979 Revolution has been associated with increased mass subsidies. The above discussion suggests an important reason why resource rents may not have been channeled toward growth. It can also shed light on the slowdown of growth in MENA economies in the 1980s and the 1990s and their vulnerability to the decline in oil prices in those years. Note that the optimal policy for spending oil revenues would have entailed saving part of the funds during resource booms and to create a stabilization fund to be spent during bust years (or paying back during booms the loans accumulated during troughs). However, this type of policy would have faced the same difﬁculty as the attempt to put them into more productive uses: The public and most interest groups would have been skeptical about ever beneﬁting from the saved funds and would not have easily consented to moving them out of current consumption. Indeed, few oil-exporting developing countries other than Kuwait have reached sufﬁcient coordination in their ﬁscal policies to restrain their expenditures in major ways during boom times. Many MENA countries have done the exact opposite, borrowing during booms when their creditworthiness has risen and paying back with difﬁculty during busts when foreign lending has dried up. In some cases, this pattern has further forced the government to inﬂationary ﬁnancing, as in the cases of Algeria, Egypt, Iran, and Syria during many years in the 1980s and 1990s (see Table 2).3 For some countries without oil revenues the inﬂation

3

Note that the overall record of inﬂation in MENA is much better than that of most of the developing world, as Table 2 shows. The point here concerns the rise and decline of inﬂation in some countries during some periods.

Table 2. Country

Price volatility in the MENA countries

a

Average CPI Inﬂation Rate Period 1965–1970

1970–1974

1975–1979

1980–1984

1985–1989

1990–1994

1995–1999

2000–2004

Oil exporting Algeria Bahrain Iran Kuwait Libya Oman Qatar Saudi Arabia Yemen

n.a. 2.36 1.53 0.00 8.23 n.a. n.a. 1.82 n.a.

4.75 10.25 7.27 10.63 1.36 n.a. n.a. 9.39 n.a.

11.70 14.88 14.73 1.78 8.40 n.a. n.a. 15.41 n.a.

8.96 5.48 19.16 5.60 11.39 n.a. 4.97 1.32 n.a.

9.10 0.98 20.48 1.58 4.88 n.a. 2.82 1.17 n.a.

24.75 0.98 20.64 4.25 9.18 1.42 2.19 1.70 37.64

12.36 0.61 26.78 2.01 4.23 0.49 3.58 0.89 20.53

2.43 0.25 14.26 1.40 7.17 0.58 2.48 0.21 9.89

8.38 6.99 10.31 5.79 7.30 1.56 2.31 8.32 17.82

Non-oil exporting Egypt Jordan Morocco Syria Tunisia Turkey

5.26 n.a. 1.02 1.45 n.a. 5.06

4.83 9.79 6.17 9.65 n.a. 13.12

10.74 11.84 9.41 8.84 n.a. 33.52

15.82 7.02 10.22 13.47 8.90 51.47

18.92 7.02 4.93 31.75 7.32 51.07

14.08 7.04 6.19 13.59 5.85 73.74

6.90 3.12 2.72 2.72 3.89 80.74

4.69 1.85 1.62 0.04 2.78 37.63

6.58 5.89 4.24 12.23 2.21 30.01

2.41

6.64

13.78

13.77

27.56

39.32

38.78

23.72

19.63

20.71

16.69

33.66

43.24

130.55

68.83

12.50

8.23

121.62

Non-MENA oil exportersb Non-MENA non-oil exportersb

81

n.a., not available. Source: Calculated based on World Bank’s World Development Indicators (2006). a Djibouti, Iraq, Lebanon, UAE, and West Bank and Gaza are not included owing to lack of sufﬁcient data. b Weighted averages, with each country’s PPP GDP serving as its weight. Oil exporters consist of Angola, the Republic of Congo, Gabon, Nigeria, Trinidad and Tobago, and Venezuela.

A Reexamination of the Political Economy of Growth in the MENA Countries

Standard deviation of CPI for the Years Available

82

Hadi Salehi Esfahani

problem has been more chronic, as in Turkey. In either case, the consequence in most countries has been economic slowdown and macroeconomic instability. The above discussion points to some speciﬁc ways in which politics and institutions have interacted with resource rents, shaping the pattern of economic performance in the region. Further causes and consequences of contracting problems, especially the coordination failures that have contributed to the outcome, will be discussed below. In the absence of well-developed credit and insurance markets (Ersel and Kandil, 2007), the considerable terms of trade ﬂuctuations, the consequent instability, and the availability of public resources during booms must have jointly contributed to the rise and maintenance of extensive PE sectors and large general public employment as social insurance substitute in the MENA countries. (For the available public employment data, see Table 3. A similar conclusion can be drawn, particularly in the case of MENA’s oil exporting countries, from the subjective indicators of government intervention and ownership shown in the ﬁrst column of Table 4 based on the rankings carried out by the Heritage Foundation).4 In addition, many MENA governments, especially non-oil exporters, have tried to address the demands for the economic security among private sector lowincome workers by restricting layoffs. This can be seen in the last column of Table 4, which shows the employment rigidity rankings developed by the World Bank. Of course, such policies provide job security only for the employed workers at the cost of employers and potential entrants into the job market. The need to manage economic risk through the next best governmentcontrolled mechanisms may have also strengthened protectionist incentives during the 1990s (Esfahani and Squire, forthcoming). However, the recovery of oil prices in recent years seems to have changed this trend and has enabled MENA’s major oil exporters to ease trade restrictions (see the middle columns of Table 4). For the rest, except for a few countries, liberalization has not proceeded as fast after the 1990s. Interestingly, the overall rates of protection in the MENA countries before the 1990s used to be below those in most other developing regions (Esfahani and Squire, forthcoming). Indeed, the presence of PEs and other redistribution mechanisms helped MENA governments reduce the impact of openness on parts of population that could potentially lose as a result of foreign competition. That situation, however, changed when per capita oil revenues declined in the 1990s and the MENA countries began to look more inward while most of the developing world raced toward greater economic integration. Besides their impacts on policy outcomes, natural resource rents have inﬂuenced the evolution of political institutions in the MENA countries. Because the

4

Objective and accurate data on public enterprises is limited, especially for the MENA countries. Despite those imperfections, the picture constructed by Tables 3 and 4 is suggestive about the extensive nature of public employment and ownership in MENA, especially in the case of oilexporting countries.

A Reexamination of the Political Economy of Growth in the MENA Countries

Table 3.

83

Public employment in MENA countries compared to the weighted average of non-MENA developing countriesa

Country

Total Public Employment as Percent of Labor Force

Public Enterprise Employment as Percent of Labor Force

1991–1995

1996–2000

1991–1995

1996–2000

Oil exporting Algeria Bahrain Iran Kuwait Libya Oman Qatar UAE Yemen

n.a. 26.9 26.3 n.a. 44.9 21.5 n.a. n.a. n.a.

19.5 n.a. 21.9 40.0 n.a. 16.4 29.4 n.a. 14.5

7.0 n.a. n.a. n.a. 1.41 n.a. n.a. 1.98 n.a.

3.5 n.a. 4.2 3.7 n.a. 7.9 0.8 n.a. 1.66

Non-oil exporting Egypt Jordan Lebanon Morocco Syria Tunisia Turkey

23.8 17.1 8.3 n.a. n.a. n.a. 10.2

25.8 16.9 8.1 7.2 n.a. n.a. 9.6

13.3 n.a. 0.6 n.a. n.a. 5.1 3.0

5.0 1.4 0.3 0.7 10.0 n.a. 2.3

13.8

11.4

4.0

2.7

Weighted average of developing non-MENA countriesa

n.a., not available. Sources: ILO, Public Sector Employment Database, the website for the International Labor Statistics; World Bank, Public Sector Employment Database, 2002, and World Bank, World Development Indicators 2000. a The developing countries included in the average are those with data for both subperiods. For China, public employment data excludes those in town and village enterprises. Labor force is used as weights across countries.

ruling elites did not have to rely much on taxation, they have found it easier to continue their rule without much representation (Mahdavi, 1970; Elbadawi and Makdisi, forthcoming). The rents have allowed the ruling politicians to buy support from interest groups, rather than being responsible to the tax-paying public. In this way, abundance of natural resources may have contributed to the survival of authoritarian regimes in the region. More on this is discussed in Section 3.4. 3.2. Heterogeneity and polarization As we have seen in Section 2, MENA societies are not particularly fractionalized, but several of them are quite polarized (Figure 3). The latter pattern has

Hadi Salehi Esfahani

84

Table 4. Government intervention, trade restrictions, and employment rigidity in the MENA countries compared to the weighted average of non-MENA developing countriesa Country

Government Intervention and Public Ownership

Trade Restrictions

World Bank’s Index of Labor Market Rigidity

1995–2000

2001–2005

1995–2000

2001–2005

2004

Oil exporting Algeria Bahrain Iran Iraq Kuwait Libya Oman Qatar Saudi Arabia UAE Yemen

4.3 3.7 4.2 4.4 5.0 4.7 5.0 4.4 3.5 4.4 4.0 4.0

4.6 4.1 4.3 4.9 5.0 4.5 4.6 4.5 4.6 4.7 4.0 4.1

4.7 5.0 2.0 5.0 5.0 2.0 5.0 2.6 2.5 4.2 2.0 4.8

3.8 4.6 3.0 3.4 5.0 2.2 5.0 3.0 2.8 4.0 2.0 3.4

38.7 55.0 n.a. 40.0 n.a. 20.0 n.a. 35.0 n.a. 13.0 33.0 37.0

Non-oil exporting (excluding Turkey) Egypt Jordan Lebanon Morocco Syria Tunisia Turkey

3.1

3.3

4.7

4.5

54.2

3.0 3.7 2.5 3.1 3.8 3.5 2.0

3.4 3.5 3.0 2.7 4.3 3.1 2.5

5.0 4.0 3.8 4.5 4.4 5.0 1.8

4.4 4.2 3.8 4.8 4.6 5.0 2.6

53.0 34.0 28.0 70.0 37.0 54.0 55.0

3.1

4.0

4.8

4.7

47.5

3.3

3.2

4.2

4.1

41.4

Non-MENA oil exportersa Non-MENA nonoil exportersa

n.a., not available. Sources: The Heritage Foundation, Index of Economic Freedom, and the World Bank, World Development Indicators (2006). a World Bank’s rigidity of employment index ranges from 0 ( ¼ least rigid) to 100 ( ¼ most rigid). Other indices range from 1–5, with higher numbers representing less market-friendly policies. For variable deﬁnitions, see Appendix Table A.1. The developing countries included in the average are those with available data for both subperiods. Labor force is used as weights across countries. Non-MENA oil exporters are Angola, the Republic of Congo, Gabon, Nigeria, Trinidad and Tobago, and Venezuela.

been a major source of internal conﬂicts in the region. In particular, Kurdish minorities and other ethnic groups in Iran, Iraq, and Turkey have at times engaged in political and even armed conﬂict with their respective governments. In addition, Lebanon, Yemen, and Oman have been the scenes of destructive

A Reexamination of the Political Economy of Growth in the MENA Countries

85

civil wars. Other examples are the war in Western Sahara and the (often less intense) conﬂicts between Berbers and Arabs in North Africa. Along with international conﬂicts (especially, the Arab–Israeli conﬂict), these tensions have strengthened the role of military establishments in several MENA countries and have provided justiﬁcation for repressive government policies. This is reﬂected in Figure 4, which maps military expenditure as share of GDP against the polarization index of Montalvo and Reynal-Querol (2005). It is notable that the two indices are not correlated outside the MENA region, but are positively related within the region in a way that affects the pattern for the developing world as a whole. Moreover, as the lower panel of Figure 3 shows, polarization seems to be negatively related to growth within the MENA region, even though the relationship of the two is very weak in the rest of the developing world. While there is some evidence at the macro level indicating that ethnic conﬂicts adversely affect policy outcomes, the details of the connections are not yet very well understood. As Bates (1999) argues, ethnic divisions need not lead to conﬂict and may in fact positively affect economic performance under some conditions. Certainly, ethnic divisions tend to increase internal cohesion within each ethnic group, which may help solve a great deal of free-rider problems that arise in more amorphous societies. These issues are worth pursuing in future political economy studies to develop a better understanding of the positive and Figure 4. Polarization and military expenditure in developing countries (polarization index is the average of polarity of ethnic and religious distributions according to the measure proposed by Montalvo and Reynal-Querol, 2005)

Military Expenditure as Percent of GDP

IRQ JOR

SYR

10

OMN

BHR ARE

5 SAU

IRN TUR

DZA

EGY

YEM MAR

KWT

TUN

0 0

0.2

0.4 0.6 Polarization Index

MENA Oil-Exporting Non-MENA Non-Oil-Exporting

0.8

MENA Non-Oil-Exporting Non-MENA Oil-Exporting

Source: Montalvo and Reynal-Querol (2005) and World Development Indicators (2006).

1

86

Hadi Salehi Esfahani

negative effects of ethnicity on economic policy and to search for ways in which the negative effects can be mitigated. Some considerations along these lines will be discussed in the next section on geopolitical issues and the role of military. Polarization may have more far-reaching effects on economic policy and growth that have not yet been explored. For example, increased polarization may raise the incentives for intense government intervention in social and political organizations as well as in markets, as a means of controlling potential conﬂict. At least, governments seem to extend their direct controls over economic activity during periods of heightened political tension. More long-term consequences of such incentives as well as other possible effects of polarization merit further study. 3.3. Geopolitical rents and threats and the role of military As we have seen, military expenditure as a percentage of GDP in MENA region is by far the highest in the world. Commensurately, the region has had its share of wars. Besides the internal factors noted above, the demand for military expenditure in the region originates in the international conﬂicts. The Arab–Israeli conﬂict remains the most signiﬁcant and stiﬂing one, as reﬂected in the large resource allocation to armed forces in Israel and its neighbors. Disputes between Iran and its Arab neighbors have also had similar, though less pronounced, effects in those countries. However, militarization has also been supported by superpower strategies and rivalries, which at times have subsidized those expenditures and offered rents to strategic MENA countries. Prior to the fall of the Soviet Union, the list included Iran, Iraq, and Syria, but currently Egypt, Israel, and Turkey remain the most prominent beneﬁciaries of such rents. (Iraq may rejoin the group once it recovers from war.) The role of geopolitical rents goes beyond military subsidies and includes support for the ruling politicians in the MENA countries. Along with the natural resource rents, this is another factor that may have helped the longevity of dictatorships in the region (Elbadawi and Makdisi, forthcoming). A number of scholars have further suggested that external rents may have weakened the urge to implement market-oriented reforms in the region (Richards, 1991; Lo¨fgren, 1993). This issue, which has been more broadly debated in the context of the role of foreign aid, remains to be fully analyzed. The issue is that countries with easier access to foreign aid can be in a better position to pursue objectives such as rent redistribution and stability with less distortion in domestic production or foreign trade. The reason why they maintain distortions may lie in institutional weaknesses, as in the case of resource rent use. Domestic and international conﬂicts have provided the military establishments in MENA with reasons to grow and to gain control over resources not only by way of weapons and personnel, but also in the form of economic activities. In many MENA countries, the military owns and runs business conglomerates that produce military as well as civilian products (Richards and Waterbury, 1990, pp. 360–367). Because of their superior political position and subsidized inputs

A Reexamination of the Political Economy of Growth in the MENA Countries

87

(including draftees who come at low cost), these conglomerates compete well in domestic markets and government contract auctions. As a result, they have turned into economic forces that have inﬂuenced the path of trade and industrial policies. As pointed out earlier, greater control of resources by the military does not seem to be negatively correlated with economic growth in MENA (Yildirim et al., 2005). This may be due to the role of military in providing basic internal and external security and in forcing solutions to persistent internal conﬂicts. The latter issue is particularly important in the context of polarized societies where there are severe contracting problems among contending groups and the military ends up providing a context for renegotiation and conﬂict resolution. For example, the political turmoil in Turkey in the 1970s ultimately came to an end through a military coup in 1980. Solutions of this type imposed by the armed forces may not be fair, but they can be more productive than the continuation of turmoil. 3.4. Authoritarianism Political institutions in most MENA countries have been relatively stable but autocratic compared to those in comparable countries. (See the ﬁrst two columns of Table 5. The measures presented in the table are derived from a combination of subjective and qualitative indices. Such institutional measures are, of course, not precise and should be treated only as indicative.) These characteristics must be an important part of any explanation of policy and performance in the region over the past few decades. Most conspicuously, regime longevity and centralization of decision making seem to have afforded reasonable coordination at the macro level, mitigating the possible negative effects of polarization. This is particularly evident in the relatively low inﬂation rates in many MENA countries compared to the rest of the developing world. Many Arab governments do have high levels of debt (see Figure 5), but as Esfahani (2005) shows, this fact actually indicates creditworthiness, which can be attributed to reasonable levels of coordination and institutional capability, though not sufﬁcient discipline to limit borrowing to the amounts needed strictly for tax smoothing. Interestingly, in Iran and Turkey where discipline has been weaker, government debt has been constrained at much lower levels (Figure 5).5 Similar coordination effects may also have helped many MENA countries score better than their counterparts elsewhere in terms of qualitative indices of government effectiveness, rule of law, and control of corruption (Table 5). The ability of many MENA countries to solve some macro contracting problems through authoritarian rule has come partly at the cost of misinformed

5

Note that the more democratic countries such as Turkey have also experienced higher inﬂation rates. Inﬂation in Iran also accelerated after the 1979 Revolution when its political system became less autocratic. These correlations do not mean that democracy is always a cause of price instability. Rather, the point is that democratization with weak policymaking institutions is likely to entail coordination failures in many policy areas.

88

Table 5. Governance indicators in MENA countries, 1996–2004 averagesa Political Stability

Government Effectiveness

Regulatory Quality

Rule of Law

Control of Corruption

Oil exporting Algeria Bahrain Iran Iraq Kuwait Libya Oman Qatar Saudi Arabia UAE Yemen

1.2 1.2 1.0 1.0 1.9 0.3 1.6 0.7 0.7 1.4 0.7 0.8

0.9 2.0 0.0 0.5 2.4 0.4 0.7 0.9 1.1 0.1 1.0 1.2

0.5 0.7 0.6 0.4 1.6 0.2 1.1 0.9 0.8 0.1 0.7 0.7

1.1 0.8 0.9 1.4 2.7 0.1 2.0 0.6 0.2 0.1 0.8 0.6

0.6 0.7 0.8 0.6 1.7 0.9 0.9 1.1 1.0 0.6 1.1 1.0

0.5 0.6 0.5 0.6 1.4 0.9 0.9 0.7 0.6 0.2 0.8 0.6

Non-oil exporting (excluding Turkey) Egypt Jordan Lebanon Morocco Syria Tunisia Turkey

0.8

0.2

0.1

0.2

0.1

0.1

0.9 0.3 0.5 0.5 1.6 0.8 0.5

0.4 0.0 0.5 0.1 0.4 0.4 0.9

0.1 0.4 0.2 0.1 0.8 0.8 0.1

0.2 0.3 0.0 0.1 1.0 0.1 0.3

0.1 0.4 0.2 0.2 0.4 0.3 0.1

0.2 0.1 0.4 0.1 0.6 0.3 0.2

Non-MENA oil exportersb Non-MENA Non-oil exportersb

0.9 0.1

1.3 0.5

1.1 0.2

0.8 0.2

1.2 0.3

1.1 0.4

Country

Source: D. Kaufmann, A. Kraay, and M. Mastruzzi (2005). ‘‘Governance Matters IV: Governance Indicators for 1996–2004,’’ World Bank Website, http:// www.worldbank.org/wbi/governance/govdata/. a Units of all indices are expressed in terms of standard deviations of rankings around zero. b The developing countries included in the average are those with available data for both sub-periods. Labor force is used as weights across countries. NonMENA Oil exporters are Angola, The Republic of Congo, Gabon, Nigeria, Trinidad, Tobago, and Venezuela.

Hadi Salehi Esfahani

Voice and Accountability

A Reexamination of the Political Economy of Growth in the MENA Countries

Figure 5.

89

Central government debt as percentage of GDP across regions

80 70

Percent of GDP

60 50 40 30 20 10 0 European Arab Sub-Saharan Eastern South Asia Monetary Countries** Africa Europe & Union Central Asia

Iran and East Asia & Latin Turkey Pacific*** America & Caribbean

Regional Weighted Averages over 1997–1999 (averaged using GDP weights over countries for which data are available) Sources: World Bank, World Development Indicators (2004), and Joint BIS-IMF-OECD-WB ** Algeria, Bahrain, Jordan, Lebanon, Morocco, Oman, and Tunisia. *** Excluding China. Central government of China has a low debt and, if included, dominates the region, lowering the average public debt down to 16.7 percent of GDP. Statistics on External Debt, http://www.oecd.org/site/0,2865,en_21571361_31596493_1_1_1_1_1,00.html.

and poor-quality policymaking at the micro level. Figure 6 depicts a way by which this tradeoff manifests itself. It shows that across countries, accountability is inversely related to the difﬁculty of doing business (measured by rank in terms of a survey-based index developed by the World Bank). Speciﬁcally, MENA countries mostly fall on the upper left part of the chart, reﬂecting a situation where businesses that are independent of the government and ruling politicians have little say in policy and face difﬁculties starting up, growing, and adjusting to economic conditions. Similar problems can be observed in terms of poor quality of services with public good components such as education and healthcare (World Bank, 2003). Such effects mute the dynamism of the economy at the micro level and counteract with the gains from coordination at the macro level. Another way of viewing the cost of doing business under centralized MENA regimes is the weaknesses of policy commitment at the micro level. Indeed, changing policies without due process has been a major complaint of domestic and foreign businesses in the MENA countries. Esfahani (1997) offers a host of examples. For instance, in Egypt, following the rise in foreign investments in 1981, the Minister of Energy decided to raise the price of energy and petroleum

Hadi Salehi Esfahani

90

Figure 6.

Democratic accountability and ease of doing business

150 Difficulty of Doing Business**

EG Y DZ AG WB

SYR IRQ

IR N

100 YEM

MAR LBN TUR

ARE TUN OMN

50

JOR

KWT

SAU

0 -2

-1

0 Democratic Accountability*

MENA Oil-Exporting Non-MENA Non-Oil-Exporting

1

2

MENA Non-Oil-Exporting Non-MENA Oil-Exporting

Sources: D. Kaufmann, A. Kraay, and M. Mastruzzi, 2005. ‘‘Governance Matters IV: Governance Indicators for 1996-2004’’, World Bank Website, http://www.worldbank.org/wbi/governance/govdata/; World Bank, ‘‘Doing business: economy rankings’’, http://www.doingbusiness.org/EconomyRankings/. * Combined index of voice and accountability calculated by Kaufmann et al. Units are standard deviations of rankings around zero. ** Based on rank calculated by the World Bank in terms of ease of doing business. Higher numbers indicate greater difﬁculty of doing business.

products for FDI (foreign direct investment) projects by a factor of 3, bringing it to the world market level. He argued that the substantially subsidized domestic price was for users who did not enjoy the privileges of the foreign investment code (Middle East News Agency, Economic Weekly, 1989, May 19, p. 13). Given that the privileges were supposed to be incentives for foreign investors and those who had invested in Egypt had assumed a uniform price of energy, the policy change placed FDI projects at a disadvantage vis-a`-vis their Egyptian competitors. Some investors were so disappointed that they ceased their operations (Sadowski, 1991, p. 115).6 There are also many other examples where

6

Another relevant example from Egypt is the cancellation of a US ﬁrm’s contract for the management of duty-free shops at the Cairo airport. The contract was subsequently awarded to Egypt’s major public sector airline, EgyptAir, which has been beneﬁting from the support of airline industry regulators in many other respects, e.g., the limitation of charter ﬂights to Cairo (Business Middle East, The Economist Intelligence Unit, 1995, November 16–30, p. 10).

A Reexamination of the Political Economy of Growth in the MENA Countries

91

governments simply cancel contracts granted to foreign investors. Just recently, the government of Iran helped a local mobile phone company, Irancell, to cancel its $3 billion contract with a Turkish company, Turkcell (Turkish Industrialists’ and Businessmen’s Association (TUSIAD), Turkey News, September 6–18, 2005). Another contract with the Turkish operator of Khomeini International Airport was repealed last year owing to pressure from Iran’s Revolutionary Guards, claiming security concerns (TUSIAD, Turkey News, January 25– February 1, 2005). The impact of authoritarianism on economic activity has not been uniform across the MENA countries. In particular, the extent and form of government involvement in the economy seems to have systematically varied between monarchies and republics. In monarchies, the survival of the traditional relations of power and the established elite has ensured support for private property, whereas in republics, state ownership has often been used as a means of dispossessing old propertied classes and establishing new orders by subordinating production to the political imperatives of new regimes (Owen, 2004, pp. 45–53). Indeed, the extension of government controls over the economy has been quite visible after the formation of republics in Algeria, Egypt, Iran, Syria, and Tunisia. There are, of course, other forces at work that explain the public sector growth in monarchies or in the case of less authoritarian republics (Turkey in the 1960s) as well – e.g., partial solutions to market failures and the connection between resource rents and public ownership discussed earlier. Many details of such effects and their interactions (e.g., the interaction of authoritarianism and polarization) in shaping economic policy await careful scrutiny. Although the survival of authoritarianism in MENA has been facilitated by natural resource rents and internal and external conﬂicts, the institution has deeper social and cultural roots as well. These roots have long been debated in the political economy literature, going at least as far back as Marx’s idea of ‘‘Asiatic Mode of Production.’’ Whether the root causes will continue to exert their inﬂuence in the coming decades or not is an important question. However, it appears that the changing composition of endowments in the region is counteracting with the old causes and is bringing about institutional change. In recent years, there have been persistent efforts to bring about democracy in a number of countries, as in the case of Iran. Many observers have attributed these efforts to the gradual rise of human capital relative to the natural resource rents as a source of income. The more educated labor forces of the MENA countries demand greater participation, organize and articulate their positions better, and when governments ignore them, their increased mobility imposes larger costs on local economies and ultimately on governments. While democratization is desirable for a variety of reasons, from an economic policy perspective it also poses major challenges. The improved representation under democratic conditions makes coordination more difﬁcult. For example, there is a great deal of historical and cross-country evidence that the expansion

92

Hadi Salehi Esfahani

of franchise increases demands on public expenditures (Shelton, 2005). Also, compared to stable autocracies, democracy tends to increase the turnover in the leadership. This shortens the horizons of policymakers in ofﬁce and can weaken their incentives to coordinate or even represent the polity’s interests. In MENA countries, where tax administration is underdeveloped and resource rents are declining, these demands may lead to unsustainable budget deﬁcits, a problem that Turkey has experienced for a few decades as it has moved toward a more established democracy. Countering these effects requires a great deal of institution building to enable competing interest groups coordinate their actions in the macroeconomic policy area. Similarly, without enhancing the coordination mechanisms, democratization may increase the demand for protection and undermine the prospects of trade liberalization. Another implication of democratization in a world of imperfect contracts is the particular difﬁculties that countries with large resource rents tend to face. As discussed in Section 3.1, when the public knows that the government has access to large rents that may be diverted to particular interests, the voters become keen to capture their shares through the imperfect means that a nascent democracy provides. As a result, mass subsidies and various forms of protectionism may expand in the early stages of democratization process, giving rise to greater distortions in the economy (Esfahani, 2003). This implies that the democratic transition may be particularly difﬁcult times for MENA economies with wellendowed resources. The impact of democratization on policy commitment is also complicated and depends on the details of the institutions that structure the government’s decision-making process. On the one hand, the democratic process is going to make policy change more difﬁcult because a larger number of interest groups become involved and their conﬂicting preferences may preserve existing policies. On the other hand, the shorter time horizon of policymakers may induce them to choose lower quality policies and refuse to honor promises made by their predecessors. Moreover, unstable coalitions may emerge and undermine the credibility of long-term policies. Again, there are complementary institutions that can mitigate this problem (e.g., an independent and competent judiciary, a strong party system, a capable and meritocratic bureaucracy, etc.). But, MENA nations are not particularly well endowed with such institutions and need to build them as democratization proceeds. Along with democratization, a process of decentralization is likely to start in some MENA countries. This is a natural development and may contribute to the quality of public services at the local level. However, like democratization, its success is contingent on a set of complementary institutions that ensure coordination at the macro level. The unfortunate experiences of Brazil and Argentina with decentralization in the 1980s and their contrast with the operation of decentralized systems in China and India, among many others, contain important lessons for arrangements that must accompany decentralization to ensure success. (For important lessons from the case of China, see Qian, 2003).

A Reexamination of the Political Economy of Growth in the MENA Countries

Table 6.

93

Average years of schooling in MENA countries, 1970–2000a

Country

1970

1980

1990

2000

Oil exporting Algeria Bahrain Iran Iraq Kuwait Libya

0.97 0.82 1.82 1.01 0.76 2.88 1.38

1.91 1.56 3.12 1.93 1.83 4.29 2.22

3.34 3.01 4.87 3.36 3.20 5.99 3.37

4.69 4.72 6.09 4.66 4.34 7.05 4.67

Non-oil exporting Egypt Jordan Syria Tunisia Turkey

1.53 1.25 2.29 1.67 0.91 2.16

2.24 2.21 2.93 2.86 1.92 2.80

3.40 3.57 5.39 4.35 3.02 3.95

4.51 5.05 7.37 5.74 4.20 4.80

Non-MENA oil exportersb Non-MENA non-oil exportersb

1.20 2.94

1.91 3.77

2.54 4.80

3.52 5.37

Sources: Barro-Lee education data available from the NBER website, http://www.nber.org. Labor force data extracted from the World Bank, World Development Indicators website. a MENA countries shown in the table are only those for which data are available. b The developing countries included in the average are those with available data for both subperiods. Labor force is used as weights across countries. Non-MENA oil exporters are Angola, the Republic of Congo, Gabon, Nigeria, Trinidad and Tobago, and Venezuela.

3.5. Human capital Until about 1980, educational attainments in MENA were quite low compared with the other developing countries (Table 6).7 This was largely the legacy of the region’s traditional and tribal social structures, though in some areas it has been reinforced by colonial policies (Ayubi, 1995). However, over the past few decades, most MENA countries have been quickly catching up with their peers in terms of schooling quantity, though the achievement is less solid on the quality dimension (Salehi-Isfahani, 2007). The attention given by MENA governments to education in recent decades started with the post-WWII social contracts promising improvement in the living conditions for broad segments of the population. Expansion of education also served as part of the state’s effort to extend its control and shape the development process. Like other dimensions of government controls, the extension of education came easier in quantitative terms and was to some extent complementary to those dimensions (e.g., educating the workforce that was to ﬁll the newly created government jobs). However, extensive government controls

7

This is based on years of schooling. Data on literacy rates reveal similar patterns.

94

Hadi Salehi Esfahani

of overeconomic activity seem to have generated negative interactions with education quality. Government jobs focused the goals of the educational system on minimum skills and degrees, rather than entrepreneurship and creativity (SalehiIsfahani, 2007). The widespread interventions also diverted the government’s attention and resources away from more fundamental improvements in educational quality.8 Weakness of administrative institutions discussed below has further contributed to the problem, and may at the same time have been partly caused by educational deﬁciencies. In any event, a manifestation of the phenomenon seems to be the apparent puzzle of low economic growth in the face of rapidly expanding education in MENA (Pritchett, 1999). If this conjecture is correct, we should before long observe much higher social returns to education in countries that manage to expand the role of private markets. 3.6. Administrative and judicial institutions Administrative and judicial institutions are not particularly effective in the MENA region. (Figure 7 shows two indicators in these respects, though the numbers are subjective and subject to the usual caveats.) Typically, court cases take long time to settle and there is considerable concern among the public that judgment may not be competent and impartial. Moreover, in most MENA countries, bureaucrats are viewed as lacking sufﬁcient competence and impartiality (World Bank, 2003). Effectiveness in judicial and administrative services is highly desirable because these are valuable and crucial public goods particularly for improving contracting conditions in both private and public arenas. However, building a strong bureaucracy and judiciary has posed a number of dilemmas for developing countries. One fundamental problem is that reforming these institutions requires some capability to design and to carry out policies (or deals and contracts) that in turn depend on having capable administrators and state organizations in the ﬁrst place. Although there are strategies of starting from some parts of these institutions and gradually rebuilding the system, implementing these strategies often faces a number of problems. In authoritarian countries, it is usually difﬁcult for the rulers to grant sufﬁcient autonomy and power to bureaucrats and judges to enable them to move forward with the necessary reforms. This is particularly the case because some partiality in those organizations is needed for maintaining the structure of power. In new democracies, in contrast, the difﬁcult task is to coordinate a multitude of players and convince each that all the others will be cooperating to ensure a desirable long-term outcome for the reforms. Currently, there is limited knowledge concerning the requirements of reform and the details of judicial and administrative institutions that match each

8

Financial resources, of course, were less of a constraint for oil-exporting MENA countries, which have managed to achieve substantial improvement in schooling since the 1970s (see Table 6).

A Reexamination of the Political Economy of Growth in the MENA Countries

Figure 7.

95

Court independence and bureaucratic quality across developing countries

1984-2004 Average of ICRG Bureacratic Quality**

4

3 TUR JOR MAR EGY

TUN

2

SAU OMN KWT DZA

IRN SYR YEM

1

IRQ

0 1

2

3

4

Independence of Courts* MENA Oil-Exporting Non-MENA Non-Oil-Exporting

MENA Non-Oil-Exporting Non-MENA Oil-Exporting

Sources: Charles Humana, World Human Rights Guide (1992) and Political Risk Services, Inc. Knack and Keefer (1995). * Court independence is a subjective ranking of autonomy of courts for the early 1990s. The range is 0–4 and higher scores indicate greater independence. ** Bureaucratic quality is a survey-based subjective measure of autonomy from political pressure and strength and expertise to govern as well as the existence of an established mechanism for recruiting and training. The range is 0–4 and higher scores indicate higher quality.

country’s conditions. Despite voluminous literature on these subjects, careful analytical work on them remains inadequate. Recently, some systematic research on these subjects has started (see, for example, La Porta et al., 2004). More work along these lines and applications to the MENA countries should be an important part of further research on the political economy of growth in the region. 4. Conclusion The factors shaping economic growth are truly numerous and vary greatly across countries and over time. Moreover, these factors interact with each other in multifaceted ways, making every country’s situation at each point in time by and large speciﬁc. This fact has very important implications for the ways in which research on growth needs to be conducted and for the policy implications for each case. Most importantly, it implies that the causes of stagnation and growth are ultimately situation-speciﬁc and must be understood locally. Furthermore, the actions needed to enhance growth must also be assessed and applied at the local level by policymakers who are intimately familiar with the situation.

96

Hadi Salehi Esfahani

The situation-speciﬁcity discussed above does not mean that no generalizations can be made and there is no role for general frameworks for understanding growth. Indeed, the framework proposed in this paper and the illustration of its use for analyzing major issues of growth in MENA suggest that there are ﬂexible perspectives that can help organize thinking about economic development at general levels. Moreover, such approaches can shed light on why conditions become speciﬁc and which lessons from other cases can be transferred to others and which ones are not applicable elsewhere. The proposed framework and its emphasis on incomplete contracting owing to complexity also highlight the importance of carrying out in-depth case studies for understanding growth in speciﬁc context while keeping in mind the lessons of the underlying economic theory. Another important general implication is that solutions to growth problems need to be eventually formulated and implemented at the local level by people who are intimately familiar with the situation at hand and, at the same time, have a good grasp of the available theoretical frameworks and empirical results. While the incomplete contracting framework proposed here is fairly general, it has basic differences from the earlier approaches to the political economy of growth. Perhaps the most important feature of the framework is that it goes beyond the simple claims that powerful interest groups have been blocking reform. Furthermore, it offers ideas about why interest groups and politicians may not ﬁnd ways to adopt more efﬁcient policies that yield larger surpluses for themselves. In particular, in the case of MENA, the framework shows that the interactions among the politicians, different interest groups, and the public are far more intricate than the image of authoritarian governments forcing selfserving bad policies on the society. Rather, it may be the need for maintaining public support in the absence of a complete social contract that leads politicians to opt for inefﬁcient policies. This further implies that although some of the institutions or policies one ﬁnds in the region may not be desirable from a longrun perspective, under the circumstances of the short and medium run they may be better than feasible alternatives. However, change will be necessary as the underlying problems are better understood and addressed. In fact, there are already some institutions and policies that have better substitutes if the existing political economy knowledge is better disseminated among the politicians and the public. Future research on the political economy of growth must proceed in a number of directions. To begin with, there is a need for more detailed case studies to explore the roots of failures to improve policies in speciﬁc circumstances. Second, among political actors there are many details of interactions that need to be explored theoretically and empirically, especially in the context of non-democratic political systems. This should particularly include studies of how the political structure shapes economic policies and how in turn economic outcomes affect the contracting capability of the political economy actors. Finally, there is a great need for documenting and quantifying institutional details

A Reexamination of the Political Economy of Growth in the MENA Countries

97

and their dynamics in MENA, especially in ways that would be comparable across countries. This would provide better opportunities for testing hypotheses and digging deeper into the causes of growth and stagnation.

Acknowledgments I would like to thank Maria Augusztinovics, Micael Castanheira, Shanta Devarajan, Heba Handoussa, Gary McMahon, Jeff Nugent, Lyn Squire, and two anonymous referees for their helpful comments on the earlier drafts of this paper. I am also indebted to the participants in the GDN conferences in Prague and Cairo and the economics workshops at the American University in Washington, American University in Beirut, and the University of Maryland for helpful discussions.

References Abed, G.T. and H.R. Davoodi (2003), Challenges of Growth and Globalization in the Middle East and North Africa, Washington, DC: International Monetary Fund. Abu-Bader, S. and A.S. Abu-Qarn (2003), ‘‘Government expenditures, military spending and economic growth: causality evidence from Egypt, Israel and Syria’’, Journal of Policy Modelling, Vol. 25, pp. 567–583. Acemoglu, D. (2005), ‘‘The form of property rights: oligarchic vs. democratic societies’’, Manuscript, MIT. Acemoglu, D., S. Johnson and J. Robinson (2001), ‘‘The colonial origins of comparative development: an empirical investigation’’, American Economic Review, Vol. 91(5), pp. 1369–1401. Acemoglu, D., S. Johnson, and J. Robinson (2005a), ‘‘Institutions as the fundamental cause of long-run growth’’, in: P. Aghion and S. Durlauf, editors, Handbook of Economic Growth, Amsterdam: Elsevier. Acemoglu, D., S. Johnson, J.A. Robinson, and P. Yared (2005b), ‘‘Income and democracy’’, Manuscript, MIT. Acemoglu, D. and J.A. Robinson (2001), ‘‘Inefﬁcient redistribution’’, American Political Science Review, Vol. 95, pp. 649–661. Acemoglu, D. and J. Robinson (2005), Economic Origins of Dictatorship and Democracy, Cambridge, UK: Cambridge University Press. Alesina, A., A. Devleeschauwer, W. Easterly, S. Kurlat, and R. Wacziarg (2003), ‘‘Fractionalization, Journal of Economic Growth’’, Vol. 8, pp. 155-194. Alesina, A. and A. Drazen (1991), ‘‘Why are stabilizations delayed?’’, American Economic Review, Vol. 81, pp. 1170–1188. Alesina, A. and R. Perotti (1995), ‘‘The political economy of budget deﬁcits’’, International Monetary Fund Staff Papers, Vol. 42(1), pp. 1–31.

98

Hadi Salehi Esfahani

Alesina, A. and G. Tabellini (1990), ‘‘Voting on the budget deﬁcit’’, American Economic Review, Vol. 80(1), pp. 37–49. Amsden, A. (1989), Asia’s Next Giant: South Korea and Late Industrialization, New York: Oxford University Press. Ayubi, N.N. (1995), Over-stating the Arab State: Politics and Society in the Middle East, London: I.B. Tauris Publishers. Barro, R.J. (1996), ‘‘Democracy and growth’’, Journal of Economic Growth, Vol. 1(1), pp. 1–27. Bates, R.H. (1999). ‘‘Ethnicity, capital formation, and conﬂict’’, CID Working Paper No. 27, Harvard University. Campos, J.E. and H.L. Root (1996), The Key to the Asian Miracle: Making Shared Growth Credible, Washington, DC: Brookings Institution Press. Castanheira, M. and H.S. Esfahani (2003), ‘‘Political economy of growth: lessons learned and challenges ahead’’, pp. 159–212 in: G. McMahon and L. Squire, editors, Explaining Growth: A Global Research Project, New York, NY: Palgrave Macmillan. Deacon, R.T. and H. Bohn (2000), ‘‘Ownership risk, investment, and the use of natural resources’’, American Economic Review, Vol. 90(3), pp. 526–550. Elbadawi, I.A. and S. Makdisi (2006), ‘‘Democracy and development in the Arab world’’, Quarterly Review of Economics and Finance, Vol. 46(5), . Ersel, H. and M. Kandil (2007), ‘‘Financial development and economic growth in the MENA countries’’, in: J. Nugent and H. Pesaran, editors, Explaining Growth in the Middle East, Contributions to Economic Analysis, Amsterdam: Elsevier. Esfahani, H.S. (2000), ‘‘Institutions and government controls’’, Journal of Development Economics, University of Illinois at Urbana-Champaign, Vol. 63, pp. 197–229. Esfahani, H.S. and A. Toossi-Ardakani (2002), ‘‘What determines the extent of public ownership?’’, Manuscript, University of Illinois. Esfahani, H.S. (2003), ‘‘A political economy model of resource pricing with evidence from the fuel market’’, Manuscript. Esfahani, H.S. (2005), ‘‘Institutions and ﬁscal discipline in MENA region’’, Manuscript, University of Illinois at Urbana-Champaign. Esfahani, H.S. (2006), ‘‘Understanding common trends and variations in the growth experience of MENA countries’’, Manuscript, University of Illinois at Urbana-Champaign. Esfahani, H.S. and L. Squire (2006), ‘‘explaining trade policy in the Middle East and North Africa’’, Quarterly Review of Economics and Finance, Vol. 46(5). Esteban, J. and D. Ray (1994), ‘‘On the measurement of polarization’’, Econometrica, Vol. 62(4), pp. 819–851. Glaeser, E.L., R. La Porta, F. Lopez-de-Silanes and A. Shleifer (2004), ‘‘Do institutions cause growth?’’, Journal of Economic Growth, Vol. 9, pp. 271–303.

A Reexamination of the Political Economy of Growth in the MENA Countries

99

Hart, O., A. Shleifer and R.W. Vishny (1997), ‘‘The proper scope of government: theory and an application to prisons’’, Quarterly Journal of Economics, Vol. 112, pp. 1127–1161. Henry, C.M. and R. Springborg (2001), Globalization and the Politics of Development in the Middle East, Cambridge, UK: Cambridge University Press. Knack, S. and P. Keefer (1995), ‘‘Institutions and economic performance: crosscountry tests using alternative institutional measures’’, Economics and Politics, Vol. 7(3), pp. 207–227. La Porta, R., F. Lo´pez-de-Silanes, C. Pop-Eleches and A. Shleifer (2004), ‘‘Judicial checks and balances’’, Journal of Political Economy, Vol. 112(2), pp. 445–470. Lau, L., Y. Qian and G. Roland (2000), ‘‘Reform without losers: an interpretation of China’s dual-track approach to transition’’, Journal of Political Economy, Vol. 108(1), pp. 120–143. Lebovic, J.H. and A. Ishaq (1987), ‘‘Military burden security needs and economic growth in the Middle East’’, Journal of Conflict Resolution, Vol. 31(1), pp. 106–138. Leite, C. and M. Weidmann (1999). ‘‘Does mother nature corrupt? Natural resources, corruption and economic growth’’, IMF Working Paper WP/99/ 85, Washington, DC. Levy, B. and P. Spiller (editors) (1996), Regulations, Institutions and Commitment: The Case of Telecommunications, Cambridge, UK: Cambridge University Press. Lo¨fgren, H. (1993), ‘‘Economic policy in Egypt: a breakdown in reform resistance?’’, International Journal of Middle Eastern Studies, Vol. 25(3), pp. 407–421. Mahdavi, H. (1970), ‘‘The patterns and problems of economic development in rentier states: the case of Iran’’, in: M.A. Cook, editor, Studies in the Economic History of the Middle East, London, UK: Oxford University Press. Makdisi, S., Z. Fattah and I. Limam (2007), ‘‘Determinants of growth in the MENA countries’’, in: J. Nugent and H. Pesaran, editors, Explaining Growth in the Middle East, Contributions to Economic Analysis, Amsterdam: Elsevier. Montalvo, J.G. and M. Reynal-Querol (2005), ‘‘Ethnic polarization, potential conﬂict, and civil wars’’, American Economic Review, Vol. 95(3), pp. 796–816. McGuire, M.C. and J.M. Olson (1996), ‘‘The economics of autocracy and majority rule: the invisible hand and the use of force’’, Journal of Economic Literature, Vol. 34(1), pp. 72–96. North, D.C. (2005), Understanding the process of economic change, Princeton, NJ: Princenton University Press. North, D.C. and B. Weingast (1989), ‘‘Constitutions and commitment: the evolution of institutions governing public choice in seventeenth-century England’’, The Journal of Economic History, Vol. 49(4), pp. 803–831. Owen, E.R. (2004), State, Power and Politics in the Making of the Modern Middle East, London: Routledge.

100

Hadi Salehi Esfahani

Persson, T. and L.E.O. Svensson (1989), ‘‘Why a stubborn conservative would run a deﬁcit: policy with time-inconsistent preferences’’, The Quarterly Journal of Economics, Vol. 104(2), pp. 325–345. Political Risk Services, Inc. (1995), International Country Risk Guide Data, 1982–95, Computer ﬁle, Syracuse, NY: Compiled by IRIS Center, University of Maryland, College Park, MD. Pritchett, L. (1999), ‘‘Has education had a growth payoff in the MENA region?’’, MENA Working Paper Series No. 18, Washington, DC: The World Bank. Pritchett, L. (2003), ‘‘A conclusion to cross national growth research: a foreword ‘to the countries themselves’’, in: G. McMahon and L. Squire, editors, Explaining Growth: A Global Research Project, New York, NY: Palgrave Macmillan. Qian, Y. (2003), ‘‘How reform worked in China?’’, pp. 297–333 in: D. Rodrik, editor, In Search of Prosperity: Analytic Narratives on Economic Growth, Princeton: Princeton University Press. Richards, A. (1991), ‘‘The political economy of dilatory reform: Egypt in the 1980’s’’, World Development, Vol. 19(12), pp. 1721–1730. Richards, A. and J. Waterbury (1990), A Political Economy of the Middle East, Boulder, CO: Westview Press. Rodrik, D. (editor) (2003), In Search of Prosperity: Analytic Narratives on Economic Growth, Princeton: University Press Princeton. Rodrik, D. (2005), ‘‘Growth strategies’’, in: P. Aghion and S. Durlauf, Handbook of Economic Growth. Rodrik, D., A. Subramanian and F. Trebbi (2002), ‘‘Institutions rule: the primacy of institutions over geography and integration in economic development’’, Manuscript, Harvard University. Sadowski, Y. (1991), Political Vegetables? Businessman & Bureaucrat in the Development of Egyptian Agriculture, Washington, DC: Brookings Institution Press. Sala-i-Martı´ n, X., and E.V. Artadi (2003), ‘‘Economic growth and investment in the Arab world’’, Economics Working Paper 683, Department of Economics and Business, Universitat Pompeu Fabra. Sala-i-Martin, X. and A. Subramanian (2003), Addressing the natural resource curse: an illustration from Nigeria’’, NBER Working Paper No. 9804. Salehi-Isfahani, D. (2007), ‘‘Microeconomics of Growth in MENA: the Role of households’’, in: J. Nugent and H. Pesaran, editors, Explaining Growth in the Middle East, Contributions to Economic Analysis, Amsterdam: Elsevier. Shelton, C. (2005), ‘‘The Size and Composition of Government Expenditure’’, Manuscript, Stanford University. United Nations Development Programme (UNDP) (2002), Arab Human Development Report 2002, New York: United Nations. Velasco, A. (2000), ‘‘Debts and Deﬁcits with Fragmented Fiscal Policymaking’’, Journal of Public Economics, Vol. 76(1), pp. 105–125.

A Reexamination of the Political Economy of Growth in the MENA Countries

101

von Hagen, J. (1992). ‘‘Budgeting procedures and ﬁscal performance in the European communities’’, Economic Papers No. 96, Brussels: Commission of the European Communities Directorate General for Economic and Financial Affairs. Wade, R. (1990), Governing the Market: Economic Theory and the Role of Government in East Asian Industrialization, Princeton: Princeton University Press. Waterbury, J. (1993), Exposed to Innumerable Delusions: Public Enterprise and State Power in Egypt, India, Mexico, and Turkey, Cambridge, UK: Cambridge University Press. World Bank (2003), Better Governance for Development in the Middle East and North Africa, Washington, DC: The World Bank. Yildirim, J., S. Sezgin and N. O¨cal (2005), ‘‘Military expenditure and economic growth in Middle Eastern countries: a dynamic panel data analysis’’, Defence and Peace Economics, Vol. 16(4), pp. 283–295. Yousef, T.M. (2004), ‘‘Development, growth and policy reform in the Middle East and North Africa since 1950’’, Journal of Economic Perspectives, Vol. 18(3), pp. 91–116. Appendix: The variables are deﬁned and measured according to the following criteria (See Table A.1). Table A.1. Variable definitions for Table 4 Score

1

2

3

Government Intervention and Public Ownership

Trade Policy

Level of Government Intervention in the Economy

Average Tariff Rate (percent)

Non-Tariff Barriers

Very low:o10 percent of GDP; virtually no government-owned enterprises Low: 11 – 25 percent of GDP; a few government-owned enterprises, like the postal service; aggressive privatization program in place Moderate: 26 – 35 percent of GDP; several government-owned enterprises like telecommunications, some banks, and energy production; stalled or limited privatization program

o4

Very low

5–9

Low

10–14

Moderate

102

Hadi Salehi Esfahani

Table A.1. Score

4

5

Continued

Government Intervention and Public Ownership

Trade Policy

Level of Government Intervention in the Economy

Average Tariff Rate (percent)

Non-Tariff Barriers

High: 36 – 45 percent of GDP; many government-owned enterprises like transportation, goods distributors, and manufacturing companies Very high: 46 percent or more of GDP; mostly government-owned industries; few private companies

15–19

High

>20

Very high

Source: The Heritage Foundation website, Index of Economic Freedom (2006).

CHAPTER 4

Financial Development and Economic Growth in the MENA Countries Hasan Ersel and Magda Kandil Abstract Financial development contributes to economic growth through two channels. The direct channel is through mobilizing and allocating financial resources. The indirect channel is through creating the appropriate environment for monetary policy to be effective. The paper analyzes the performance of the MENA countries in mobilizing and allocating financial resources using aggregative indicators. The findings indicate that the MENA countries were quite successful in mobilizing financial resources, but relatively less efficient in allocating these resources. However, on average, the direct contribution of financial development to economic growth in the MENA countries is not significantly different from the other developing economies. Further, the paper analyzes the effect of financial development on the effectiveness of monetary policy in a subset of MENA countries. The findings indicate that monetary policy in these countries is capable of inducing long-lasting non-neutral effects that differentiate trend output growth across countries over time. Monetary policy appears, therefore, to be an important determinant of economic growth in the MENA countries. The paper concludes by pointing out directions to improve financial development in the MENA countries by promoting competition in the financial sector, strengthening supervision, and increasing the independence of central banks. On the other hand, taking further steps to create an appropriate legal environment (e.g., improve the quality of legislation and the judiciary system,

Corresponding author. The views in the paper are those of the authors and do not necessarily represent those of the IMF

or IMF policy. CONTRIBUTIONS TO ECONOMIC ANALYSIS VOLUME 278 ISSN: 0573-8555 DOI:10.1016/S0573-8555(06)78004-X

r 2007 ELSEVIER B.V. ALL RIGHTS RESERVED

104

Hasan Ersel and Magda Kandil

etc.) may further help the MENA countries to reap the utmost benefits by maximizing the potential of a healthy financial system.

Keywords: monetary policy, ﬁnancial intermediation, economic growth, price inﬂation, MENA countries, resource allocation, regulations JEL Classifications: G20, E50, 011 1. Introduction A ﬁnancial system provides means of payment and enables the economic decision makers to establish links between current and future economic activities. In that sense, the necessity of ﬁnancial development for economic growth is selfevident. However, the existence of an association between these two variables is not sufﬁcient to draw conclusions concerning the role of ﬁnance in the growth process. A well-performing ﬁnancial system is expected to promote growth by inﬂuencing positively the rate of accumulation and efﬁciency of capital by enabling decision makers to take advantage of existing (or potential) ﬁnancing opportunities. In an imperfect world, a ﬁnancial system is expected to play an important role in making the decision makers aware of such opportunities and the risks associated with them. Toward achieving such a goal, a ﬁnancial system is expected to facilitate the mobilization of capital as well as its efﬁcient allocation among different users. To serve this end, a ﬁnancial system is expected to fulﬁll the following objectives: (1) Increase the amount of ﬁnancial savings (resource mobilization function). (2) Allocate ﬁnancial savings to those projects with the highest return and, therefore, help economic agents to realize the value of their decisions in the best possible way. (resource allocation function). A ﬁnancial system, in contrast, has its own operating costs. Therefore, a desirable property of the ﬁnancial system should be to achieve the aforementioned objectives in a cost-efﬁcient manner. In other words, a well-designed ﬁnancial system should (3) Enable ﬁnancial intermediaries to operate at the lowest cost (operational efﬁciency). In such a setting, a ﬁnancial system can be considered as a resource allocation mechanism designed to achieve the two objectives within a given environment (i.e., institutions and information transmission facilities). The success of the ﬁnancial system in achieving these objectives provides its direct contribution to economic growth. While the role of ﬁnancial development is crucial to mobilize and allocate private resources, ﬁnancial deepening is equally important to enhance the

Financial Development and Economic Growth in the MENA Countries

105

efﬁcacy of the monetary policy. In a market economy, where economic decisions are taken in a decentralized way by numerous and heterogeneous decisionmaking units, monetary policy is conducted by sending signals to affect their behavior. Therefore, the success of the monetary policy depends not only on the appropriateness of the decisions taken by the monetary authority, but also on the ﬁnancial system’s capability of transmitting policy signals to economic decision-making units as well as on the nature of their responses. Conceptually, one can distinguish two phases in the implementation of a monetary policy decision. In the ﬁrst stage, the central bank (the authority responsible for conducting monetary policy) sends its signals through (open market operations, the discount rate, and the required reserve ratio) to ﬁnancial institutions. In the second phase, ﬁnancial institutions, primarily banks, serve as ﬁnancial intermediaries in transmitting central banks’ signals to non-ﬁnancial decision makers. More speciﬁcally, banks translate policy signals into actions that eventually affect the volume and allocation of funds that are being transferred. The ﬁnancial system, therefore, by enhancing the efﬁciency of monetary policy, can indirectly contribute to economic growth. The purpose of this paper is to examine the relation between ﬁnancial development and economic growth in a sample of Middle East and North Africa (MENA) countries for which data are available.1 Section 2 analyzes the direct contribution of ﬁnancial development to economic growth. Section 3 is devoted to monetary policy to understand the role of the ﬁnancial system in the transmission mechanism of policy signals. Section 4 provides the conclusions and a summary of policy issues that are relevant to the MENA countries. 2. Financial development and economic growth: direct contribution Financial development refers to a rather complex and multidimensional process that involves deepening and broadening of the ﬁnancial markets via changes in the institutions and the regulatory environment as well as the introduction of new ﬁnancial instruments. Therefore, one can approach the problem of describing ﬁnancial development by listing all the relevant variables and introducing proxy measures to represent them. That was the approach followed in Creane et al. (2003), namely collecting and analyzing information for 100 quantitative and qualitative variables under six themes to describe ﬁnancial development in the MENA countries. An alternative approach is to reduce the informational requirements of describing the ﬁnancial development by treating the process in an aggregative framework. In this paper, we follow the aggregative approach to analyze the relation between ﬁnancial development and growth. Namely, we focus on the two main functions of a ﬁnancial system in the growth

1

Data problems prevented coverage of all MENA countries.

106

Hasan Ersel and Magda Kandil

process (i.e., resource mobilization and allocation functions) and evaluate the efﬁciency of the ﬁnancial system in achieving these functions. The framework incorporates available data for the MENA countries to evaluate the performance of the ﬁnancial system in achieving resource mobilization, resource allocation, and operational efﬁciency. 2.1. Resource mobilization function A ﬁnancial system intermediates between savings-surplus and savings-deﬁcit agents in channeling funds provided by the former group to the latter. Financial systems offer different types of instruments to make such transfers suitable to the needs and tastes of the parties involved. One major distinction is between ‘‘direct’’ and ‘‘indirect’’ ﬁnancing. In direct ﬁnancing, the securities markets play a fundamental role. In indirect ﬁnancing, intermediaries, i.e., banks, assume the risk of making transformations to satisfy the interests of agents on the two sides of the fund transfer. In the MENA countries, indirect ﬁnancing is the dominant form of ﬁnance. Therefore, in the rest of the paper, the focus will be on banking. Since markets for ﬁnancial securities are also developing in these countries, we expect that the role of banks in resource mobilization will gradually decrease over time.2 One can follow two different approaches to measure the resource mobilization performance of a ﬁnancial system. The ﬁrst approach is to use the volume of funds controlled by the ﬁnancial system, proxied by Broad Money (currency

2

To quantify the importance of the securities markets in allocating ﬁnancial resources, two measures are frequently used. These are market capitalization and turnover ratios. In the former ratio, market capitalization is measured by share prices times the number of shares outstanding. The second ratio is deﬁned as total value of shares traded during the period divided by the average market capitalization for the period. The following table indicates that the markets for securities, even under the most generous assumptions, are evident to play a marginal role in allocating ﬁnancial resources in developing countries.

YEAR 2001

MENA countriesa 11 Middle-income countriesa EMU countries

Market Capitalization/ GDP (%)

Turnover Ratio (%)

34.6 48.5 89.9

6.2 7.7 90.6

Source: The World Bank: World Development Indicators (WDI), Washington, DC, The World Bank, 2002. a Refers to the corresponding sets of countries that are listed in Table 1.

Financial Development and Economic Growth in the MENA Countries

107

plus demand and time deposits), as a performance indicator for the domestic ﬁnancial system. The second approach is to evaluate the performance of a ﬁnancial system by looking at the difference between its ‘‘intermediation potential’’ and its realization. In Table 1 Broad Money/GDP for nine MENA countries and a set of emerging market economies are given. As can be seen from the ﬁgures in Table 1, the ratio of Broad Money to GDP varies considerably among countries. This may be due to other factors that are independent from the quality of their ﬁnancial systems (such as accustomed transaction practices, the level of inﬂation, excess liquidity, migrants’ ﬂows, the interest rate, the exchange rate, and supply-side constraints). However, the ﬁgures in Table 1 do not indicate a signiﬁcant difference between the resource mobilization capabilities of the ﬁnancial systems of the MENA countries and emerging countries in other parts of the world. It is clear that the domestic ﬁnancial systems of countries cannot be considered as the only channels that transfer ﬁnancial resources. One can think of two competing forms of ﬁnance that compete with the domestic ﬁnancial system. Table 1. Broad Money/GDP Ratio (%) (1997–2001 Averages) MENA Countries Algeria Egypt Iran Jordan Kuwait Lebanon Morocco Saudi Arabia Tunisia Average Middle-income Countries Argentina Brazil Chile India Korea Malaysia Mexico Phillipines Thailand Turkey Venezuela Average Source: International Institute of Finance, Data Base.

% 44.34 75.87 38.85 116.35 85.76 170.62 78.34 47.54 51.71 78.82 29.11 29.76 41.81 59.37 173.43 139.72 24.89 60.71 100.70 46.77 18.40 69.56

108

Hasan Ersel and Magda Kandil

The ﬁrst is the traditional ﬁnance. It may take various forms ranging from usury to the trade ﬁnance among corporations.3 The effectiveness of the traditional ﬁnance depends on many diverse factors, ranging from the development level of the ﬁnancial system to the size of the underground economy. Although, there are insurmountable difﬁculties in collecting data for the volume of traditional ﬁnance, it is reasonable to assume that its importance declines as the ﬁnancial system develops. The second major competitor is the international ﬁnancial system, whose importance signiﬁcantly increased for the developing countries during the era of ﬁnancial globalization. One way of demonstrating its importance for developing countries may be as follows: Consider the case of a frictionless global ﬁnancial system where ﬁnancial institutions are able to offer their services to any customer, irrespective of its location. Then each economic decision-making unit will be in a position to choose to operate with the ﬁnancial institution that serves its needs best. In today’s imperfectly competitive environment, however, national ﬁnancial systems are protected (both de jure and de facto) against foreign competition in mobilizing residents’ ﬁnancial resources. But, however perfect (and/or protected) it may be, no country’s ﬁnancial system can be the only intermediary for its residents in a global ﬁnancial environment. Therefore, major differences between the potential of a ﬁnancial system and its actual performance in mobilizing ﬁnancial resources can be considered as an indicator of its effectiveness.4 The Institute of International Finance releases data on deposits held abroad by residents5 of countries in commercial banks that report to the Bank of International Settlements (BIS). Since these deposits belong to those residents that reveal their identity it is highly unlikely that their existence is the result of some illegal operations. Therefore, they can be considered leakage of resources that could have been mobilized by the domestic ﬁnancial system instead. In Table 2, the amount of deposits held abroad is measured as ratios to Broad Money and GDP for seven MENA countries and eleven middle-income countries from other regions of the world. From the ﬁgures given in the table it is clear that neither these MENA countries nor those from other regions, are homogeneous with respect to their ﬁnancial system’s ability in mobilizing ﬁnancial resources. The data also reveal that in terms of absolute levels, even after the exceptionally high ﬁgure for Jordan is excluded, the leakage is quite

3

In many countries, trade credit among non-ﬁnancial companies in various forms is used rather widely. In fact, it may even be the major external source for working capital needs of companies. Since standard ﬁnancial statistics do not cover such trade credits, for the countries where such practice is widely used, the credit to GNP ratio stays at a rather low level. 4 It should also be pointed out that the reasons behind these divergences may range from mistrust of the economic policies followed in that country, or, in an open-economy context, to the effectiveness of external competition. 5 These ﬁgures exclude assets held abroad by resident deposit money banks, which are given separately.

Financial Development and Economic Growth in the MENA Countries

Table 2.

109

Deposits held abroad by residents in commericial banks that report to the BIS (1997–2001 Averages)

Country

As % of Broad money

As % of GDP

MENA countries Algeria Egypt Iran Jordan Kuwait Lebanon Morocco Saudi Arabiaa Tunisiaa

27.26 32.20 21.67 94.01 64.25 57.66 23.59 82.83 28.77

11.69 24.93 9.33 109.40 54.91 98.32 18.21 39.21 13.82

Average (except Saudi Arabia and Tunisia)

45.80

46.68

42.16 28.29 50.48 9.11 5.37 11.74 41.58 29.06 10.79 28.41 145.15

12.30 8.25 20.25 5.23 7.94 15.49 10.09 16.38 10.32 10.72 25.61

36.55

12.96

Middle-income countries Argentina Brazil Chile India Korea Malaysia Mexico Phillipines Thailand Turkey Venezuela Average

Source: International Institute of Finance, Data Base. For Saudi Arabia and Tunisia: Comparative Statistics for Emerging Market Economies, International Institute of Finance, April 2000, pp. 132. a Time series data are not available for these countries. The ﬁgures are the averages of 1997 and 1998.

high for the MENA countries. This indicates that the ﬁnancial systems in these countries have scope to enhance their performance in mobilizing ﬁnancial resources. Nonetheless, a thorough evaluation of the difference would require analysis of factors (such as the legal environment, uncertainty, agents’ conﬁdence, economic policies, etc.) that affect residents’ decisions in evaluating the quality of domestic and foreign ﬁnancial institutions. 2.2. Resource allocation function A ﬁnancial system is said to fulﬁll its resource allocation function if it allocates the resources it mobilized to the most proﬁtable and productive users. A ﬁnancial system should be in a position of delineating the most proﬁtable users of the funds that it transfers and it can, therefore, be considered relatively more

110

Hasan Ersel and Magda Kandil

efficient if it achieves a better use of ﬁnancial resources. A ﬁnancial system may allocate resources based on market information, where competitive forces play more role, or on non-market information, where the weight of other factors (social, political, administrative, etc.) is more prevalent. In repressed/protected economies, a ﬁnancial system allocates ﬁnancial resources, but these decisions are based on non-market criteria. In contrast, in many of the developing countries, even after implementing liberalization programs, non-market allocation schemes remain in activities such as ﬁnancing of state-owned companies and agricultural support programs. One way of approaching the problem of allocation efﬁciency is to assume that a ﬁnancial system has a higher chance in identifying better customers, if its decisions are market based.6 To approximate the market-based portion of the funds allocated to the domestic economy, the share of claims on the private sector7 to total domestic credits is used. This choice is based on the assumption that the ﬁnancial system allocates resources to the private sector according to market criteria. Evidently, such an assumption may not correspond to reality for various reasons. For emerging countries, the most important violation is the ‘‘related party lending’’ practice of private banks. In many countries private banks are afﬁliated with industrial groups and may frequently extend loans to the group companies based on non-market criteria. In contrast, state-supported programs to the private sector, although included in credits to the private sector, can hardly fulﬁll the proﬁtability criteria in the usual sense. In Table 3, claims on the private sector as a percentage of domestic credits is given for 9 countries from the MENA region and for 12 major emerging countries from other regions of the world. The ﬁgures in the table indicate that the MENA countries are far from being homogeneous with respect to the role of their ﬁnancial systems.8 The share of private claims in total credits also exhibits a wide variation, ranging between 28.87 (Syria) and 92.49% (Tunisia). However, as the ﬁgures for other emerging economies indicate, heterogeneity is not peculiar to the MENA countries. Emerging countries from the other regions of the world also exhibit considerable differences. The ﬁgures also reveal that the ratio of the claims on the private

6

Basing decisions on non-market criteria does not necessarily imply an irrational behavior. Many developing countries, which followed planned development strategies to promote industrialization through import substitution policy, created protective environments to allocate their resources. In such environments, the priority was given to ﬁnancing activities or projects in industries or in regions, which were considered socially important in development plans. Therefore, ﬁnancial proﬁtability was left out partly (or completely) as a criterion for allocation in such decisions and was replaced by the social beneﬁt. Such a system inevitably generated the concept of ‘‘duty loss’’ even for private companies, including private banks. The implied incentive structure was based on the principle of ‘‘profits are private and losses are social,’’ which created the well-known moral hazard problems. 7 Line 32d in the International Financial Statistics (IFS) data. 8 See Creane et al. (2003) for a supporting argument.

Financial Development and Economic Growth in the MENA Countries

Table 3.

111

Claims on privatge sector as % of domestic credits (1997–2001 average)

Country

%

MENA Countries Algeria Egypt Iran Jordan Kuwait Lebanon Morocco Saudi Arabia Syria Tunisia

14.62 55.53 49.79 83.07 55.91 52.26 61.70 69.38 28.87 92.49

Average

55.77

Middle-income Countries Argentina Brazil Chile India Korea Malaysiaa Mexico Phillipines Thailanda Turkey Venezuela Average

67.86 60.23 90.62 56.62 89.71 103.42 69.62 73.03 100.06 51.28 79.69 76.56

Source: International Institute of Finance, Data Base. a A larger than 100% ratio in the second column for Malaysia and Thailand indicate, that the public sector is supplying credit in these countries.

sector to total domestic credits is signiﬁcantly low for the MENA countries compared to other emerging countries. In fact only the ratios for Jordan and Tunisia, exceeds the average of the other emerging countries in Table 3. 2.3. Operational efficiency Operational efﬁciency of a ﬁnancial system refers to its ability to provide ﬁnance at the lowest possible cost. The cost of providing ﬁnance consists of two components. First, is the interest paid to the ultimate lender and second is the intermediation cost. The operational efﬁciency of the ﬁnancial system refers to the latter, since the former is determined by factors that are beyond the control of ﬁnancial institutions (such as macroeconomic conditions, the depth of the ﬁnancial system, etc.).

Hasan Ersel and Magda Kandil

112

Table 4. Country

MENA countries Algeria Egypt Jordan Kuwait Lebanon Morocco Syria Middle-income countries Argentina Brazil Chile Korea Mexico Phillipines Thailand Venezuela

Indicators of operational efficiency

Real Interest Rate on Deposits (%) (1997–2001 Average)

Cost of Intermediation Ratio (Lending Rate/ Deposit Rate) (1997–2001 Average)

Lending Rate/TBill Rate

2.98 5.50 5.06 4.04 n.a. 4.47 3.05

1.35 1.40 1.59 1.53 1.26 1.84 2.25

1.26 1.55 n.a. n.a. 1.56 n.a. n.a.

9.8 17.1 5.8 5.3 3.99 2.98 2.91 7.1

1.43 3.15 1.53 1.17 2.46 1.43 1.96 1.49

n.a. 3.04 n.a. n.a. 1.2 1.8 n.a. n.a.

Source: IFS Data Base; n.a., not available

Unfortunately, data concerning operational efﬁciency indicators are either not available or not reliable for many countries. Therefore, the rather incomplete information given in Table 4 should be treated cautiously. In the ﬁrst column of Table 4, real interest rates on deposits for the MENA countries are given as averages for the 1997–2001 period. The average value of this ratio for the MENA countries is 4.37%. The London Interbank Offered Rate (LIBOR) rate for US dollar deposits, after adjusting for US inﬂation, is 3.14% for the same period. As one expects, the cost of resource mobilization is higher for the MENA countries, but not excessively higher than the LIBOR rate.9 In the second column of Table 4, the average ratio of lending rate/deposit rate (cost of intermediation ratio) is given for the 1997–2001 period. This ratio, under quite strict assumptions, can be used as a proxy for the intermediation cost.10 The cost of intermediation ratio is expected to be lower for more efﬁcient

9

Countries from Latin America in the sample have either excessively high real deposit rates (Argentina and Brazil) or signiﬁcantly negative real rates (Mexico and Venezuela). 10 This ratio is affected by the currency composition of the assets and liabilities of banks when valuation gains (or losses) owing to exchange rate developments are not taken into account in calculating the interest rate. Valuation problems are more distorting for the high-inﬂation economies.

Financial Development and Economic Growth in the MENA Countries

113

ﬁnancial systems. However, the ﬁndings in this table are puzzling. The cost of intermediation ratios for emerging economies in the table, except for Brazil, Mexico, and Syria, are lower than that of the Euro Area average, which is 2.1.11 In the third column of Table 4 the lending rate/T-Bill rate is reported. Suppose that banks treat government securities as a risk-free asset and loans are extended primarily to private agents. Then, this ratio will reﬂect the risk premium in lending to private agents. A high ratio, therefore, may indicate a weakness in the ﬁnancial system’s ability/willingness to ﬁnance private agents. For the Euro Area, the average value of this ratio is 1.3 for the 1997–2001 period. Unfortunately, data are not available to calculate this ratio for most of the emerging countries in the sample. Data are available for three MENA countries and three other emerging economies. The ratios for Egypt and Mexico are below the Euro Zone average. The only exceptionally high ratio is for Brazil, which indicates that during the period in question the risk premium on private lending was extremely high.

2.4. Economic growth and financial development indicators The next logical step is to see whether these three dimensions have some effect on economic growth in the MENA countries. In this exercise the MENA countries, irrespective of their per capita income, are treated within the world of developing countries and the following two questions are asked: (1) Do these three dimensions contribute to the growth performance of developing economies? (2) Is there a difference between the MENA countries and other developing countries with respect to the contribution of these direct channels on the their growth performance? To answer these questions a set of indicators is deﬁned to represent each dimension. The relation between these variables and the growth performance of the countries in the sample is tested by using cross-section data for developing countries. The choice of the indicators as well as the sample was restricted by data availability. The reason behind the choice of indicators is as follows:

11

In the IFS data base, the ‘‘prime rate’’ data collected from banks is given as a proxy for the ‘‘lending rate.’’ In practice such a rate may not be correlated at all with the actual rates that banks charge in their lending activities. First of all, the share of best customers may be too small to represent the actual rates. Secondly, banks may only be using the ‘‘prime rate’’ for ofﬁcial reporting and not in determining their actual lending rates. Finally, extension of a loan, in most instances, is a component of a basket of services that the bank offers to its client, where both parties are more concerned with the pricing of the basket and not its individual components.

Hasan Ersel and Magda Kandil

114

2.4.1. Resource mobilization function To represent this function the following variable is used: RM ¼

Broad Money ðMoney þ Quasi MoneyÞ GDP

The choice of such measure is based on the following assumption: the savings rate in an economy is determined by a set of economic, sociological, and political factors. The ﬁnancial system development has little, if any effect, on the savings rate. What the ﬁnancial system does is to transform ﬁnancial savings and make them available to be used by economic decision makers. The data for Broad Money is obtained from IFS, whereas the GDP ﬁgures are from WDI. 2.4.2. Resource allocation function The following indicator is used to represent the resource allocation function: RA ¼

Credit to the Private Sector Credit to the Private Sector þ Claims on Government and Other Public Entities

This indicator is based on an asymmetric view of treating private and public sectors. The private sector, in this treatment, is assumed to attain the status of being credible through the quality of their operations. The public sector, in contrast, derives its credibility from being part of the government. Therefore, as the ﬁnancial system extends more credit to the private sector, it can be considered as contributing more to the growth of the economy.12 The data are obtained from WDI. 2.4.3. Operational efficiency The following, widely used, interest margin variable is used as a proxy for operational efﬁciency: OE ¼

12

Lending Rate Deposit Rate

In fact, this assumption can be challenged on various grounds. First of all, the banking system may play a purely passive role in channeling preferential credits to the private sector. In other words, volume and conditions of such credits may be administratively determined and ﬁnanced through public sector funds. Secondly, but less likely in developing countries, credits to the public sector may be subjected to the same conditions as the private sector. In contrast, it is not necessarily true that credits to the private sector, by deﬁnition contribute to economic growth. They may be misused through schemes such as connected lending.

Financial Development and Economic Growth in the MENA Countries

115

It is assumed that the closer this ratio to 1, the higher is the operational efﬁciency of the ﬁnancial system.13 Data are obtained from WDI. To see whether these variables have an effect on the growth performance of the developing countries a cross-section study was conducted. The sample was chosen on the basis of data availability and consists of 97 developing countries. In this study, a set of control variables (such as per capita income, the inﬂation rate, the savings rate, and monetary growth) are used along with ﬁnancial development indicators. In contrast, to answer the second question, intercept and slope dummies are introduced to see whether being a MENA country14 makes a difference concerning the direct effects of the ﬁnancial system. Since the aim of the exercise is to look at the long-run effects of ﬁnancial development on growth, the averages of the relevant variables for the 1980–2000 period are used. Countries without sufﬁcient data are eliminated. However, among the countries that remained in the sample, some have still missing observations for certain variables. In the equation reported,15 the control variables are per capita income (GDP/ N), inﬂation (INF) and the savings rate (S/GDP). The expected signs of the coefﬁcients for RM and RA are positive, whereas for OE it is negative. In contrast, statistical signiﬁcance of the slope and interaction dummy variables will give an indication whether the MENA countries differ from other developing countries in the relation between ﬁnancial development and economic growth. The regression equation was estimated by the OLS and the ﬁndings are reported in the Table 5. The results reported in the table indicate that the coefﬁcients of the three ﬁnancial development indicators are statistically signiﬁcant with the expected signs. Increasing the resource mobilization and resource allocation capacities of the ﬁnancial systems and enhancing the operational efﬁciency of the ﬁnancial system (measured by the decrease in interest margin) positively affect growth performances of developing countries. In contrast, the coefﬁcients of all dummy variables that are introduced to distinguish the MENA countries are statistically insigniﬁcant. In other words, the ﬁndings do not point out a qualitative difference in the relation between ﬁnancial development and economic growth in the MENA countries when compared with other developing countries.

13

Interest margin may be quite misleading in high inﬂation countries where banks have open foreign exchange positions. As the effects of exchange rate movements are accounted for separately from interest payments, if a bank ﬁnances itself with foreign exchange deposits and extends loans in domestic currency, the above ratio automatically rises. 14 The MENA countries included in the sample are Algeria, Bahrain, Egypt, Jordan, Kuwait, Morocco, Syria, and Tunisia. 15 Different sets of control variables did not change the qualitative nature of the ﬁndings. In general ﬁnancial indicators are statistically signiﬁcant with the expected signs, but the control variables were not.

116

Table 5.

Hasan Ersel and Magda Kandil

The Effect of Financial Indicators on the Rate of Growth in Selected Developing Countries

Parameter Constant Per capita GDP Inﬂation Savings rate Resource mobilization Resource allocation Operational efﬁciency MENA dummy (intercept) MENA dummy resource mobilization MENA dummy resource allocation MENA dummy operational efﬁciency

Coefﬁcients

T-Stat

Signiﬁcance

0.2451 0.00056 0.0051 0.0203 0.03588 0.2613 0.0072 0.2598 0.0560 0.2590 0.0089

2.371 0.7148 0.4079 1.3810 4.808 2.5218 2.003 0.6001 1.3842 0.5776 0.5039

0.01996 0.4767 0.6843 0.1709 0.000006 0.0135 0.0482 0.5500 0.1698 0.5651 0.6156

Adjusted R2 ¼ 0.3340; F(10, 86) ¼ 5.8144 (Signiﬁcance level 0.000001); DW statistic: 1.925. The null hypothesis that the coefﬁcients RMF, RAF, and OE are zero was also rejected, F(3,86) ¼ 11.23796 (Signiﬁcance level 0.000003).

3. Indirect contribution of financial markets to growth: The effectiveness of economic policy Financial development, through broadening the spectrum of ﬁnancial markets as well as inducing institutional developments is expected to create a better environment for monetary policy to be effective. Under normal assumptions, such a development is expected to have a positive effect on the growth potential of the economy. Before examining the role and effectiveness of the monetary policy a logical issue needs to be addressed. Consider an economy with only an omnipotent central bank. Obviously, in such an economy both markets and institutions will be quite different from what is envisaged in ﬁnancial development programs. Therefore, the information concerning the monetary aggregates may not shed sufﬁcient light on the ﬁnancial development level of the country in question. In contrast, the distinction made between direct and indirect contributions of a ﬁnancial system to economic growth implies that the ﬁnancial intermediaries are expected to operate efﬁciently in carrying out two different functions. The direct contribution of the ﬁnancial system to growth requires ﬁnancial intermediaries to be capable of taking decisions concerning ﬁnancial resource mobilization and its efﬁcient allocation. However, to carryout their second function, these institutions are expected to act as agents of economic policy makers, notably of central banks. It is extremely difﬁcult, if not impossible, to design an institution that can undertake these two rather diverse functions simultaneously in a perfect way. It can be argued that the former function is best served in a competitive environment accompanied by strong supervision. However, for the second function, assuming that internal informational and organizational problems can be

Financial Development and Economic Growth in the MENA Countries

117

solved, a gigantic central bank that controls both resource mobilization and resource allocation functions may suit better and serve the purpose. Such an institutional design may enhance efﬁciency of implementation of policy decisions for two reasons. First, it will eliminate the communication problems between the principal and the agents. Second, it may also lead to lower costs of operation when compared with an institutional setting based on principal (the central bank) and agents (banks) relation.16 These considerations highlight the importance of the ﬁnancial system architecture and the recent ﬁnancial reform efforts. Financial markets do not operate in a vacuum, but in an institutional environment. However, institutions are not neutral and they inﬂuence the working and the efﬁciency of the markets. Institutions are also human-made artifacts and change from one society to the other, owing to differences in cultural heritage, the legal system, and political factors. Therefore, it is neither realistic nor desirable to expect to design a uniform ﬁnancial architecture applicable to all economies. Instead, the focus should be on delineating the functions to be carried out by the ﬁnancial system and the tradeoffs that feasible structures impose on. As can be seen from the example given above, aggregating commercial banking and central banking under a single institution, even if it enhances ﬁnancial systems’ indirect contribution to growth, may hamper its direct contribution by eliminating competition among ﬁnancial intermediaries. Financial liberalization programs, one way or another, aim at enhancing a competitive environment for the ﬁnancial system to develop. Therefore, the expected outcome of these programs is an increase in the role of ﬁnancial intermediaries, in practice, notably banks. This implies a reduction in the relative size of monetary authorities, i.e., central banks. One simple, albeit naı¨ ve, way of measuring the relative size of central banks vis-a`-vis the ﬁnancial system is to compare reserve money with broad money (money+quasi money). Reserve money can be considered as a proxy for the size of the central bank whereas broad money is a proxy for the ﬁnancial system as a whole. Notice that central banks can increase reserve money, ceteris paribus, by increasing currency issued or by raising the reserve requirement ratio. In either case, the central bank’s role in resource mobilization and allocation increases. In Table 6, the ratios of reserve money to broad money are given as averages for 1977–1981 and the 1997–2001 periods for the MENA countries and a set of middle-income countries from other regions of the world. The periods are chosen to represent, vaguely, the initial periods of ﬁnancial liberalization and its ﬁnal stage. The ﬁve-year averages are taken to smooth the effect of outlying years.

16

Agents may have their own preferences and institution-speciﬁc constraints that affect their behavior. Under certain circumstances, their private concerns and/or constraints may dominate their behavior and the desired policy outcome may not be realized. Therefore, devising and implementing a mechanism that relies on the performance of intermediaries may not sound attractive for central banks.

Hasan Ersel and Magda Kandil

118

Table 6. Country

Reserve Money/Broad Money 1977–1981 Average

1997–2001 Average

MENA countries Algeria Egypt Irana Jordan Kuwait Lebanon Morocco Saudi Arabia Syria Tunisia

43.64 52.40 45.28 61.62 19.02 25.48 36.28 57.18 66.78 24.98

33.04 32.48 39.84 35.34 6.26 20.80 27.34 21.36 47.36 20.10

MENA average

43.27

28.39

Middle-income countries Argentina Brazil Chile India Korea Malaysia Mexico Phillippines Thailand Turkey Venezuela

34.38 30.62 43.36 37.00 29.90 47.22 60.72 53.54 23.24 55.14 31.40

19.84 22.46 79.56 26.06 8.32 25.72 21.18 19.96 12.68 17.38 40.20

Middle-income countries average

40.59

26.67

Source: International Financial Statistics Yearbook, 2002, Washington, DC: IMF, pp. 92–95. a Data for 1978 is not available for Iran.

It is clear from the table that, this ratio exhibited a marked decline for all the countries (except Chile) between the two periods. It should also be pointed out that despite such a decline, the averages for both groups of countries remain high when compared with those of the ﬁnancially advanced economies.17 In contrast, MENA countries still exhibit a higher ratio when compared to other middle-income countries for the 1997–2001 period.18

17

In 2001, this ratio was 13.4% for Japan, 9.5% for the USA, 6.2% for Canada, and 2.9% for the UK. Notice that the average of this ratio for 1997–2001 period is at the level of ﬁnancially advanced countries for Kuwait and Korea. 18 This result is strengthened if the outliers (Syria, Kuwait, Chile, and Korea) are excluded. In this case, the MENA countries’ average increases to 30.85% and the average of middle-income countries declines to 22.83% for the 1997–2001 period.

Financial Development and Economic Growth in the MENA Countries

119

This result can be interpreted in the following way: Financial development, indeed, led to an increase in the relative importance of the ﬁnancial intermediaries with respect to that of the central banks of the emerging economies. These countries made notable progress toward enhancing the role of ﬁnancial intermediaries. This tendency is relatively weaker for the MENA countries (Kuwait seems to be an exception within the sample). However, for the MENA countries, it can be assumed that the monetary policy is increasingly conducted in an environment where central banks take into account the behavior of the ﬁnancial intermediaries and ﬁnancial market developments. In the light of this observation, the following exercise is undertaken to see whether the MENA countries’ ﬁnancial development level enables them to conduct monetary policy in an effective manner or not. 3.1. Monetary policy design The ﬁrst step of the empirical investigation is to determine how rational agents perceive that the design of monetary policy varies in response to policy makers’ observations of economic conditions.19 It is assumed that the monetary authority’s reaction function in period t is dependent on variables realized in period (t-1) as follows: Dmoneyt ¼ a0 þ a1 Dirt1 þ a2 Dpt1 þ a3 Dyt1 þ a4 Dgovt1 þ a5 dumt1 þ Zm ð1Þ Here, Dmoneyt is the growth of the money supply at time t and a0 is a constant. Agents expect the monetary authority to determine the growth of the money supply at time t in response to their observations at time t-1 of the growth in international reserves, Dir, price inﬂation, Dp, real output growth, Dy, and the growth of government spending, Dgov.20 To account for possible structural change in the design of monetary policy, the empirical model accounts for a dummy variable, dum, that takes zero values prior to 1973 and 1 thereafter.21 The term Zm is a random residual with a zero mean and a constant variance. The growth of international reserves is likely to create monetary expansion. In contrast, a reduction in international reserves prompts the monetary policy authority to slow down monetary growth and raise the domestic interest rate to attract foreign ﬂows. The response of the monetary authority to price inﬂation is likely to vary depending on the policy objective. An increase in price inﬂation stimulates an

19

For some discussion on the views governing the design of monetary policy, see Hoskins (1993). Empirical investigations of the reaction function for monetary policy in developing countries include Porzecanski (1979) and Joyce (1991). 20 This list is not intended to be inclusive. Instead, the list aims at identifying major determinants of monetary policy subject to data availability. 21 1973 marks the energy price hike following the Arab–Israeli conﬂict.

120

Hasan Ersel and Magda Kandil

increase in money demand. An accommodating monetary policy is likely to vary monetary growth positively with price inﬂation. In contrast, a stabilizing monetary policy may slow down the growth of the money supply to counter inﬂationary pressure and achieve price stability. An increase in real income stimulates an increase in money demand. An accommodating monetary policy increases monetary growth in anticipation of higher real growth. In contrast, a stabilizing monetary policy attempts to counter ﬂuctuations in real output growth around its trend. Consequently, a recessionary decline in real output growth prompts the monetary authority to increase the money supply to stimulate growth back to its natural rate level. The growth of government spending may determine the growth of the money supply. In many developing countries, the Central Bank is totally controlled by the government. If the monetary authority is obligated to monetize the budget deﬁcit, an increase in the growth of government spending is likely to stimulate an accommodating growth in the money supply. The empirical model in (1) assumes that private rational agents hypothesize a rule with a feedback from past movements in the economy to the money supply. The central bank accounts for ﬂuctuations in the economy in designing its policy of monetary growth. The empirical investigation measures the effects of variables in model (1) on the growth of the money supply. Assuming rationality, agents are expected to observe variables at time (t-1) that may affect the design of monetary policy. If economic agents are able to utilize this information in wage and price negotiations, anticipated monetary policy is likely to prove inﬂationary without affecting the real economy. In contrast, if economic agents lack rationality and/ or the ability to utilize available information effectively, changes in monetary growth, although anticipated, determine real output growth. This is because agents are unable to observe or react to available information about future changes in monetary growth. Unanticipated changes in the money supply may be induced by variables that do not enter agents’ information set at time t-1 or in response to random shocks at time t. Given agents’ inability to anticipate surprises in monetary growth, they are unable to incorporate the information about these surprises in wage and price negotiations. Accordingly, unanticipated changes in monetary growth at time t shifts the demand curve along the economy’s short-run supply curve, affecting both real output growth and price inﬂation.22 Testing the relevance of available information to the effects of monetary policy will have direct bearing on the efﬁciency of ﬁnancial systems. More

22

These implications are shared by standard macroeconomic models that assume rationality to decompose demand shifts into anticipated and unanticipated changes. These include imperfect information models of the variety of Lucas (1973, 1975) and New Keynesian sticky-wage models (see, e.g., Fischer, 1977) or sticky-price models (see, e.g., Ball et al., 1988).

Financial Development and Economic Growth in the MENA Countries

121

efﬁcient ﬁnancial systems will increase agents’ ability to utilize available information and infer the direction of anticipated monetary policy. Hence, anticipated monetary policy will determine the long-run path of inﬂation that accommodates real output growth. The empirical model in (1) aims at two objectives. The ﬁrst one is to shed some light on the design of monetary policy by approximating the monetary policy reaction function based on economic conditions at time t-1. Second, the empirical estimates in model (1) will be used to decompose the growth of the money supply into two components: anticipated and unanticipated growth. The anticipated component is the predicted value from equation (1) that captures the change in monetary growth in response to variables realized at time t-1. The residual from equation (1) is the unanticipated component of monetary growth. Having achieved this decomposition, the investigation will then evaluate the real and inﬂationary effects of each component of monetary policy. The empirical models of real output growth and price inﬂation account for demand and supply shifts as follows: Dyt ¼ b0y þ b1y Dmoneyt þ b2y Dgovt þ b3y Dext þ b4y Doilpt þ b5y dumt þ Zy ð2Þ Dpt ¼ b0p þ b1p Dmoneyt þ b2p Dgovt þ b3p Dext þ b4p Doilpt þ b5p dumt þ Zp ð3Þ Here, b0y and b0p are constants. The trend components of the logarithm of real output, y, and the price level, p, are modeled as a stochastic function of time following the suggestions of Nelson and Plosser (1982). This requires estimating the models in ﬁrst-difference form where D(.) is the ﬁrst-difference operator. Sources of demand shifts increase real output growth and price inﬂation. Accordingly, an increase in monetary growth is likely to determine real output growth and price inﬂation with magnitudes, b1y and b1p . Similarly, the growth of government spending, Dgov, is likely to stimulate aggregate demand in two directions: An increase in government purchases of goods and services and an increase in subsidies to ﬁnance private spending. Accordingly, the growth of government spending determines real output growth and price inﬂation with magnitudes, b2y and b2p . An increase in the domestic currency price of Special Drawing Rights (SDRs) (a devaluation of the domestic currency) may affect aggregate demand as follows.23 Devaluation policy is likely to increase exports and curb imports and, therefore, stimulates aggregate demand. Further, devaluation increases real expenditures on imported variable inputs. This, in turn, induces ﬁrms to demand more money and decreases investment demand. In

23

The effects of devaluation in less developed countries are expected to vary from the standard implications for advanced countries. For more details see Porter and Ranney (1982).

122

Hasan Ersel and Magda Kandil

addition, devaluation is likely to determine aggregate supply by affecting the cost of the output produced. The change in the domestic currency price of Special Drawing Rights (SDRs), Dex, affects, therefore, real output growth and price inﬂation with magnitudes b3y and b3p . In addition, changes in the oil price are likely to affect the cost of the output produced in oil-importing countries and stimulate aggregate demand in oil-exporting countries. Subsequent effects of the change in the oil price on real output growth and price inﬂation are measured by b4y and b4p . To account for possible structural break, the dummy variable dumt is included in the empirical models for real output growth and price inﬂation. To differentiate between the effects of anticipated and unanticipated monetary policy, monetary growth in equations (2) and (3) is replaced as follows: Dyt ¼ b0y þ bay EDmt þ buy Dmst þ b2y Dgovt þ b3y Dext þ b4y Doilpt þ b5y dumt þ Zy

ð4Þ Dpt ¼ b0p þ bap EDmt þ bup Dmst þ b2p Dgovt þ b3y Dext þ b4p Doilpt þ b5p dumt þ Zp

ð5Þ Here, EDm and Dms are the anticipated and unanticipated components of monetary growth according to the monetary policy reaction function.24 The parameters ba and bu measure the effects of the anticipated and unanticipated components of monetary policy on real output growth and price inﬂation. By construction, anticipated monetary policy, if signiﬁcant, may have long-lasting effects on dependent variables. In contrast, random unanticipated changes in monetary policy are transitory in nature. Hence, their signiﬁcant effects induce short-term ﬂuctuations in the dependent variables.25 The long-run neutrality of monetary policy is dependent on the effect of anticipated growth of the money supply on real output. If rational agents are able to anticipate and react to monetary policy, anticipated monetary growth is absorbed fully in price without determining real output, i.e., money is neutral. In contrast, prices do not adjust fully to unanticipated growth in the money supply in the short run. Accordingly, monetary surprises are distributed between real output growth and price inﬂation in the short run. The allocation is dependent on the ﬂexibility to adjust prices and capacity constraints in the short run.

24

By construction, Edm is a function of lagged values of output growth and price inﬂation. This eliminates the need to include the lagged dependent variables for possible persistence in empirical models. 25 Tests of cointegration indicate that the non-stationary dependent variable is not jointly cointegrated with non-stationary independent variables. Hence, the empirical models are estimated without an error correction term.

Financial Development and Economic Growth in the MENA Countries

123

3.2. Time-series results The empirical models (1), (4), and (5) are estimated jointly for each of the countries under investigation to highlight determinants of monetary policy and the allocation of monetary growth between real output growth and price inﬂation. Based on data availability, the sample of countries under investigation includes Algeria, Egypt, Jordan, Kuwait, Morocco, Saudi Arabia, Syria, and Tunisia. The sample period varies between 1957 and 1997 across countries based on data availability. Table A.1 (in the appendix) presents the results of estimating the empirical model that approximates the monetary policy reaction function. Monetary growth appears to increase in response to the change in international reserves. This relationship is statistically signiﬁcant in Algeria, Saudi Arabia, Syria, and Tunisia. In these cases, the monetary authority is more inclined to pursue an expansionary monetary policy in response to the increased holdings of international reserves. In contrast, the depletion of international reserves prompts the monetary authority to contract the money supply in an effort to raise the interest rate and attract foreign funds. For other countries, changes in the money supply do not show any systematic correlation with international reserves.26 The response of monetary policy to price inﬂation varies across countries. Evidence of accommodating monetary policy is consistent with the positive and statistically signiﬁcant signs for Jordan, Kuwait, and Saudi Arabia. Monetary policy attempts to counter inﬂationary pressures in the remaining countries with signiﬁcant coefﬁcients in Egypt and Tunisia. The coefﬁcients measuring the response of monetary growth to real output growth is generally insigniﬁcant, except in Jordan and Saudi Arabia. The positive response of monetary growth to output growth in Jordan and Saudi Arabia conﬁrms the accommodating nature of monetary policy in these countries. The positive response of the money supply to the increase in government spending provides evidence on monetizing the budget deﬁcit. This response is statistically signiﬁcant in Algeria and Tunisia. That is, the growth of the money supply is used to ﬁnance higher government spending in these countries. Finally, the statistical signiﬁcance of the dummy variable in the money reaction function is evident for Egypt, Morocco, Syria, and Tunisia. This evidence is consistent with a structural shift that leads to a signiﬁcant increase in the average growth of the money supply after the oil price shock in 1973. That is, the monetary policy appears to have accommodated velocity shocks attributed to the inﬂationary pressure in response to the oil price shock of 1973.

26

This may be indicative of the systematic policy of sterilization in these countries. The paper’s evidence is robust with respect to a modiﬁcation that measures the design of monetary policy in terms of domestic credit rather than the money supply.

124

Hasan Ersel and Magda Kandil

Overall, the evidence indicates variation in the design of monetary policy in response to various determinants of monetary policy. Although central banks in most developing countries were originally established to ﬁnance government budget deﬁcits, efforts are now devoted to make them more independent. Statistically signiﬁcant coefﬁcients in the reaction function indicate that central banks of the respective countries are still not independent. Next, the real and inﬂationary effects of monetary growth in the countries under investigation are examined. The monetary policy reaction function is used to decompose monetary growth into two components. The predicted value measures the component of monetary policy that varies in response to economic conditions. Other random ﬂuctuations in monetary growth are captured in the residual. The results of estimating the empirical models (4) and (5) present the real and inﬂationary effects of anticipated and unanticipated monetary growth. These results are summarized in Table A.2 (in the appendix). Anticipated changes in the money supply are statistically signiﬁcant in determining real output growth in Algeria, Jordan, and Tunisia. That is, anticipated changes in the money supply are capable of inducing non-neutral lasting effects on output growth over time. The non-neutral effects indicate failure of the ﬁnancial system to transmit signals concerning the direction of monetary policy in light of available information. Other institutional rigidities (e.g., wage and price controls) may have further prevented nominal adjustment in the face of monetary shifts, providing a channel for an adjustment in real activity.27 Similarly, random unanticipated ﬂuctuations in the money supply have statistically signiﬁcant effects on real output growth in Algeria, Jordan, and Tunisia. In the remaining countries, monetary policy is not statistically signiﬁcant in determining real output growth. The inﬂationary effect of anticipated monetary policy dominates that of unanticipated monetary policy in Jordan, Kuwait, and Saudi Arabia. The inﬂationary effect of unanticipated monetary policy appears signiﬁcant only in Jordan. In the case of Algeria and Syria, the negative coefﬁcients highlight price rigidity in response to anticipated and/or unanticipated monetary growth. This may be attributed to price control policies that prevent inﬂation from rising in response to expansionary monetary policy. Accordingly, price inﬂation is decreasing despite growth in the money supply. From the time-series evidence, there appears to be a tradeoff between the real and inﬂationary effects of monetary policy. Indeed, the correlation coefﬁcient between the inﬂationary and real effects is 0.25 for anticipated monetary policy and 0.37 for unanticipated monetary policy, across countries.

27

Along the same line, the literature citing institutional rigidity in industrial countries has provided evidence that anticipated monetary growth determines real activity in the US (see Mishkin (1983)).

Financial Development and Economic Growth in the MENA Countries

125

3.3. Policy design and the effects of monetary growth The evidence of this section will seek to evaluate the importance of the design of monetary policy to the real and inﬂationary effects of monetary growth across the countries under investigation. Table A.3 (in the Appendix) presents correlation coefﬁcients between parameters of the monetary policy reaction function and the real and inﬂationary effects of monetary growth. These correlations highlight the importance of variation in the design of monetary policy to the real and inﬂationary effects of monetary growth across countries.28 In line 1, an increase in the response of monetary growth to international reserves decreases the real effects of monetary shifts. This is evident by the negative correlation coefﬁcients, 0.17 and 0.19, between the response of monetary policy to international reserves and the real effects of anticipated and unanticipated monetary policy across countries. In contrast, an increase in the response of monetary growth to international reserves appears somewhat inﬂationary. This is evident by the positive correlation coefﬁcients, 0.045 and 0.24, between the response of monetary policy to international reserves and the inﬂationary effects of anticipated and unanticipated monetary policy across countries. In line 2, the policy response to price inﬂation appears important to the real effects of anticipated and unanticipated monetary policy, as evident by the negative correlation coefﬁcients, 0.45 and 0.58. That is, where monetary policy is accommodating price inﬂation, the real effect of this policy is smaller. In contrast, where monetary policy aims at price stability, the real effect of this policy is larger. Consistently, the inﬂationary effects of anticipated and unanticipated monetary growth vary positively with the response of monetary policy to price inﬂation with correlation coefﬁcients 0.27 and 0.57, respectively. That is, where monetary policy is accommodating price inﬂation, the inﬂationary effect of monetary policy is larger across countries. Similarly, disinﬂationary monetary policy decreases the inﬂationary effect of monetary policy across countries. According to the evidence in line 3, the response of monetary policy to output growth does not appear to differentiate the real effect of monetary policy. This is evident by the small correlation coefﬁcients, 0.059 and 0.11, between the response of monetary policy to output growth and the real effects of anticipated and unanticipated monetary growth. In contrast, an increase in the

28

Cross-country correlations employ parameter estimates from the time-series regressions, multiplied by the inverse of the standard deviation in the time-series regressions. That is, parameters that are highly variable are weighted less heavily in the cross-country analysis. It should be emphasized, however, that cross-country correlations remain illustrative regarding variation in growth across countries. Country-speciﬁc factors are likely to differentiate these correlations. Nonetheless, crosscountry correlations aim to identify whether a general pattern dominates this correlation despite variations across countries. For a detailed discussion of limitations of cross-country growth regressions, see, for example, Lee et al. (1997).

126

Hasan Ersel and Magda Kandil

response of monetary policy to output growth raises the inﬂationary effects of anticipated and unanticipated monetary policy, as evident by the positive correlation coefﬁcients, 0.40 and 0.84, across countries. In line 4, accommodating the growth in government spending does not appear to have large correlations with the real effects of anticipated and unanticipated monetary growth, as evident by the correlation coefﬁcients, 0.11 and 0.14, respectively. Similarly, monetizing the increase in government spending does not appear to be an important factor that increases price inﬂation. This is evident by the small correlation coefﬁcients, 0.02 and 0.16, between the response of monetary policy to the growth of government spending and the inﬂationary effects of anticipated and unanticipated monetary growth. Overall, cross-section correlations highlight the importance of the inﬂationary component in the design of monetary policy. Across countries, an increase in monetary growth in response to economic conditions appears particularly important in highlighting the inﬂationary effect of monetary policy. The above results do not indicate that central banks enjoy a high degree of autonomy in the countries under investigation. Anticipated changes in monetary growth appear to be an important component in the design of monetary policy that affects real output growth and price inﬂation. Through this component, monetary policy is capable of inducing long-lasting non-neutral effects that differentiate trend output growth across countries over time. While agents may be able to anticipate the direction of monetary policy based on realized economic conditions, the institutional structure (underdevelopment of the ﬁnancial system) and rigidities in labor and output markets prevent agents from utilizing this information effectively to counter the effects of anticipated monetary policy on the real economy. Consequently, changes in monetary policy, though anticipated, affect trend real output growth. Monetary policy appears, therefore, to be an important component of economic growth in the economies under consideration. It remains, however, necessary to relax supply-side constraints that limit capacity and increase the inﬂationary pressure of expansionary monetary policy. Relaxing these constraints is highly dependent on a developed ﬁnancial system that ensures a smooth ﬂow of credit to sustain output growth. 4. Summary and policy recommendations Countries in the MENA region exhibit considerable differences among themselves with respect to their level and path of ﬁnancial development as well as in the nature of the relationship between ﬁnancial development and economic growth. However, without making a sweeping generalization for the region as a whole, the following, admittedly weak, conclusions can be derived from the ﬁndings: (1) The evidence suggests that the three dimensions of the direct effects of ﬁnancial development on growth can be detected for the MENA countries.

Financial Development and Economic Growth in the MENA Countries

127

In other words, the ﬁnancial development of countries in the region has a positive effect on their growth potential. (2) As is shown in Section 3 of the paper, ﬁnancial development created the necessary environment for the central banks of the MENA countries to conduct monetary policy. (3) The MENA countries do not exhibit qualitative differences compared with other developing countries with respect to the relation between ﬁnancial development and economic growth. In the light of this evidence, the contribution of ﬁnancial development to the growth performance of the MENA countries was positive, albeit quite limited. (4) These ﬁndings are not inconsistent with the view that the MENA countries should further develop their ﬁnancial systems as well as take other necessary measures to increase the beneﬁts of ﬁnancial development. Indeed, many MENA countries need to further their efforts to create a better institutional and regulatory environment for their ﬁnancial systems to operate in a sound and efﬁcient way. In this connection, the following points may be underlined: (i) Competition. The creation of a competitive environment backed by strong supervision is considered the only way to create an efﬁcient ﬁnancial system. Although this concept lies at the core of all ﬁnancial reform programs, not many of them can be considered as successful in this regard. In creating a competitive environment, the ﬁrst issue to be dealt with is the ownership structure of the ﬁnancial intermediaries. As indicated in ERF (2002, p. 32) state-owned banks’ dominance is widespread in the region. Second, these banks are among the ‘‘handful of large local banks control most of the assets’’ in many MENA countries (such as Algeria, Egypt, and Syria).29 Under such circumstances, promoting competition in banking requires both encouraging free entry of private banks and the restructuring of state-owned banks with an ultimate objective of privatization. However, the unfair competition that stems from the privileged status of enjoying soft budget constraints of the state-owned banks is a major obstacle for the private (particularly foreign) banks that may wish to enter. In contrast, the experiences of the developing countries indicate that, in general, policy makers hesitate to make these decisions or implement them courageously, unless they face serious problems in their banking systems. The rather complex set of interrelations between the ﬁnancial and corporate sectors, the relatively high levels of non-performing loans,30 discouraged the

29

In the Gulf Cooperation Council (GCC) countries that have world-scale banks and a relatively stronger banking system, state-owned banks are less frequent, although the public sector involvement through equity participation is widespread. 30 According to the ERF (2000, p. 43) the ratio of non-performing loans to total loans, climbed up to 50% for Algeria in 1999 and was estimated to be 20–40% for Egypt. For Tunisia this ratio was 23% for 2003.

128

Hasan Ersel and Magda Kandil

governments to be active on this front, owing to unpredictable negative effects of such decisions on the corporate sector (private or public). In the light of these considerations, it can be argued that the success of the MENA countries in achieving competitive ﬁnancial systems depends signiﬁcantly on the success in restructuring state-owned banks with an ultimate objective to privatize these banks. In the case of GCC countries, where stateowned banks are not as signiﬁcant as in the other MENA countries, privatization entails reduction in the government’s share in equity. (ii) Legal environment. It goes without saying that respecting property rights is a sine qua non for market mechanism to operate. However, it should also be accompanied with measures to encourage/enforce transacting parties to honor their contracts. Therefore, operational feasibility of the market mechanism rests on the existence of an institutional structure that contains the frequency of unfulﬁlled commitments below some threshold point. Otherwise, rational agents may refrain from engaging in transactions in such a market. This phenomenon is extremely important for the ﬁnancial markets where almost all transactions are based on commitments to be fulﬁlled in the future. Therefore, from the social point of view, designing mechanisms to enforce the implementation of ﬁnancial contracts is very much required.31 The legal systems of the MENA countries vary considerably. Nevertheless, there is also a strong urge to amend the existing legislation to facilitate ﬁnancial sector development. However, the existing problems concerning the implementation of the relevant legislation (notably, competition, banking, and bankruptcy laws) indicate the need for furthering the efforts to enhance the capability and effectiveness of the judiciary.32

31

In this context a widely discussed issue is the role of the legal system in ﬁnancial development. Starting from the early 1990s, researchers (both economists and scholars of law) began to examine the relative advantages (and disadvantages) of alternative legal rules in facilitating the establishment of an appropriate environment for the sound and efﬁcient functioning of the ﬁnancial markets. In undertaking such a study, researchers La Porta et al., 1998; La Porta and Lopez de Silanes, 1998a, 1998b) conﬁned their attention to compare the relative merits of the two main traditions; namely, common law (Anglo/American) and civil law (Continental European). Nevertheless, a legal system can only draw a framework for related parties to act accordingly. Whether they actually abide by it depends on the performance of the judicial system. The differences in the capabilities of the judicial systems in applying the legislation to concrete cases, along with the enforceability of court orders, seem to be more important than the differences in the legal systems in establishing a suitable environment for the ﬁnancial markets to operate efﬁciently. 32 It is noteworthy that ﬁnancial development not only increases the volume of transactions but also broadens their spectrum by introducing new and more sophisticated type of contracts. The ability of the legal system to support such a development rests on the availability of well-trained lawyers to understand the nature of such contracts and their intricacies. In many emerging countries, ﬁnancial reforms were launched without endowing the judiciary system with such qualities. The slowness or inability of the judiciary to deal with issues pertaining to ﬁnance, inevitably, constrain (or even hinder as was seen in the 2001 Turkish ﬁnancial crisis) ﬁnancial systems’ contribution to economic growth.

Financial Development and Economic Growth in the MENA Countries

129

(iii) Institutional development. Since a major distinguishing characteristic of a ﬁnancial system is its institutional architecture, ﬁnancial development programs place a heavy emphasis on institutional design. One can distinguish two problems in this context. The ﬁrst problem is to identify the type of intermediaries that can perform satisfactorily in a particular environment. The other is to enhance the efﬁciency and reliability of a particular institution by reforming its mode of operation. A ﬁnancial system may be based on the dominance of one type of ﬁnancial institutions or on their heterogeneity. The German-type universal banking can be considered as an example for the former, whereas the US-type specialized banking is an example of the latter. It is also well known that these systems lead to different ﬁnancial development patterns.33 Heterogeneity of the ﬁnancial institutions does not stem only from the legal environment. The changes in the mode of competition may induce banks to act as if they were in a specialized banking environment. Banks seem to concentrate on those activities where they seem to have a comparative advantage. The heterogeneity of the ﬁnancial institutions will become increasingly more in the agenda of the MENA countries. First of all, the ﬁnancial systems of the MENA countries, at varying degrees, allow both Islamic and classical banks to operate simultaneously. In contrast, many MENA countries (for example, Egypt) encourage non-bank ﬁnancial intermediaries to operate as part of their capital market development programs. Therefore, these countries already have partially competing ﬁnancial institutions and, therefore, face the challenge of achieving the best division of labor among them. Second, based on other countries’ experience,34 it can be argued that the, existing banking structure is expected to become increasingly more heterogeneous, if and when, the MENA countries liberalize their capital accounts and ﬁnancial services. The ﬁnancial reforms in the MENA countries, although led to the creation of new ﬁnancial institutions (such as securities ﬁrms), in general did not affect the dominant role of universal banking. However, as one can expect,

33

Different ﬁnancial development patterns also affect the overall performance of the economy. See for example, Allen and Gale (1995) who compare the US and the German ﬁnancial systems in this respect. 34 For example, the Turkish ﬁnancial liberalization program was based on enhancing competition without changing the universal nature of the banking system by creating partially overlapping competitive institutions, Atiyas and Ersel (1994). However, liberalization of the capital account of the country led to a major change in the structure and mode of operation of her ﬁnancial system, both by broadening the spectrum of markets and through enhancing competition. These changes led banks to diverge in their focus on ﬁnancial markets and eventually they ended up with different structural characteristics and modes of behavior. For an empirical study of the structural differences among the Turkish banks see Aleskerov et al. (2001).

130

Hasan Ersel and Magda Kandil

the experience with universal banking exhibits some variation among the countries of the region.35 The second problem concerning institution development is at the micro level. Changes in environment induce or require corresponding transformations in the mode of operation of the institutions. This is particularly important for ﬁnancial intermediaries (banks), as they rely relatively much more on establishing credibility and sustaining reputation to do their business. In other words, the institutions need to be designed in such a way as to enable share-holders to compete successfully in a market system. These concerns increased interest in corporate governance in banking. In recent years, especially after the 1997 crisis, in line with developments in the world, efforts to restructure the authority and control structures of exiting banks in line with the guidelines of corporate governance are intensiﬁed in the MENA region.36 (iv) Regulation and supervision. MENA countries were no exception to the general trend in launching reforms to strengthen their regulatory and supervisory authorities’ capabilities following the universal blueprint. In practice, there is not much dispute over the general concepts of prudential regulation. Therefore, the focus is more on ascertaining the effectiveness of supervision. In this context, delineating factors that limit the supervisory authorities from exercising their power becomes an important issue. It is possible to distinguish two sets of factors that may hinder the supervisory authority from functioning properly. The ﬁrst factor is the information processing and decision-making capability of such institutions. The second factor is the autonomy of the supervisory authority. In most instances the most effective challengers of the autonomy of the supervisory authorities are governments. Governments’ negative attitudes toward the autonomy of the supervisory authorities may stem from two different considerations. The ﬁrst is the politicians’ traditional unwillingness to share authority in policy design, which can be cured by developing democratic institutions. The other source of government’s resentment to the existence of autonomous supervisory authorities is practical considerations. When state-owned banks play an important role as policy tools and/or when domestic borrowing is crucially important, the supervisory authority’s decisions may directly affect governments. Under such conditions, a conﬂict between the government and the supervisory authority may lead to a de facto violation of the supervisory

35

For example in Tunisia, the ﬁnancial reform program, strengthened the universal banks by expanding the spectrum of their activities (Boughrara, 2002). In contrast, ﬁnancial reforms in Egypt, although did not change the universal nature of her banking system, kept ‘‘a limited distinction between commercial and other banks’’, Bahaa Eldin and Mohieldin (1998, p. 121). 36 There is an increasing literature on corporate governance in banking in Arab countries. See, for example, El-Kharouf, (2000); Suleiman (2000); Mohieldin et al. (2001); El Said (2002); and El-Kady (2003).

Financial Development and Economic Growth in the MENA Countries

131

authority’s autonomy and eventually its credibility in the eyes of the other players. It is clear that such a conﬂict is avoided if state-owned banks are privatized and if the domestic debt is not excessive. The interplay between the government and the state-owned banks highlights the importance of the interdependencies between the ﬁnancial and public sector reforms. The excessive involvement of governments of MENA countries in their ﬁnancial systems seems to be a crucially important factor behind their modest success in creating strong autonomous supervisory authorities. The above factors deserve a careful study aimed at enhancing the contribution of the ﬁnancial sector to the effectiveness of monetary policy and the promotion of steady economic growth in the MENA countries. Acknowledgments The authors gratefully acknowledge the comments and help of Alpay Filiztekin. The views in this paper are those of the authors and should not be interpreted as those of the afﬁliated institutions. References Aleskerov, F., H. Ersel and M. Mercan (2001), ‘‘Structural dissimilarities in Turkish banks: 1988–1999’’, Bogazici Journal, Vol. 15(1), pp. 57–69. Allen, F. and D. Gale (1995), ‘‘A welfare comparison of intermediaries in Germany and the US’’, European Economic Review, Vol. 39(2), pp. 179–209. Atiyas, I. and H. Ersel (1994), ‘‘The impact of ﬁnancial reform: the Turkish experience’’, pp. 103–139 in: G. CaprioJr., I. Atiyas and J.A. Hanson, editors, Financial Reform-Theory and Experience, Cambridge: Cambridge University Press. Bahaa Eldin, Z.A. and M. Mohieldin (1998), ‘‘On prudential regulation in Egypt’’, pp. 111–140 in: M. El-Erian and M. Mohieldin, editors, Financial Development in Emerging Markets: The Egyptian Experience, Cairo: ECES Publication. Ball, L., N.G. Mankiw and D. Romer (1988), ‘‘The new Keynesian economics and the output-inﬂation tradeoff’’, in: Brookings Papers on Economic Activity, Vol. 2, pp. 1–82. Boughrara, A. (2002), ‘‘The monetary policy of the Central Bank of Tunisia: an assessment’’, Paper presented at ERF 9th Annual Conference, October 26–28, Sharjah, UAE. Creane, S., R. Goyal, A. Mushﬁq Mobarak and R. Sab (2003), Financial Development in the Middle East and North Africa, Washington, DC: IMF Publication. El-Kady, M.F. (2003), ‘‘Sound corporate governance for banks’’, Paper presented to The International-Arab Banking Summit for 2003, June 25–26, Montreal, Canada.

132

Hasan Ersel and Magda Kandil

El-Kharouf, F.W. (2000), ‘‘Strategy, corporate governance and the future of the Arab banking industry’’, The Arab Bank Review, Vol. 2(2), October, pp. 30–39. El Said (2002), ‘‘Aspects of corporate governance in the banking system’’, Paper presented at the conference on Institutional and Policy Changes Facing the Egyptian Economy, May 26–27, Cairo, Egypt. ERF (2000), Economic Trends in the MENA Region, Cairo: Economic Research Forum Publication. ERF (2002), Economic Trends in the MENA Region, 2002, Cairo: Economic Research Forum Publication. Fischer, S. (1977), ‘‘Long-term contracts, rational expectations, and the optimal money supply rule’’, Journal of Political Economy, Vol. 85, February 1977, pp. 191–205. Hoskins, W.L. (1993), ‘‘Views on monetary policy’’, Review, Federal Reserve Bank of St. Louis, March/April, pp. 43–55. Joyce, J.P. (1991), ‘‘An examination of the objectives of monetary policy in four developing economies’’, World Development, Vol. 19(6), pp. 705–709. La Porta, R. and F. Lopez de Silanes (1998a), ‘‘Banks, markets and structure: implications and determinants’’, pp. 29–57 in: S.J. Burki and G.E. Perry, editors, Beyond the Washington Consensus: Institutions Matter, Washington, D.C.: World Bank. La Porta, R. and F. Lopez de Silanes (1998b), ‘‘Capital markets and legal institutions’’, pp. 99–117 in: S.J. Burki and G.E. Perry, editors, Beyond the Washington Consensus: Institutions Matter, Washington, D.C.: World Bank. La Porta, R., F. Lopez de Silanes and A. Scheifer (1998), ‘‘Law and ﬁnance’’, Journal of Political Economy, Vol. 106(6), pp. 1113–1155. Lee, K., M.H. Pesaran and R. Smith (1997), ‘‘Growth and convergence in a multi-country empirical stochastic Solow model’’, Journal of Applied Econometrics, Vol. 12, pp. 357–392. Lucas, R. (1973), ‘‘Some international evidence on output-inﬂation tradeoffs’’, The American Economic Review, Vol. 63(3), pp. 326–334. Lucas, R. (1975), ‘‘An equilibrium model of business cycle’’, Journal of Political Economy, Vol. 83, pp. 1113–1144. Mishkin, F. (1983), A Rational Expectations Approach to Macroeconometrics: Testing Policy Ineffectiveness and Efficient Market Models, Chicago: University of Chicago Press. Mohieldin M., N. Abulata and M. Lasheen (2001), ‘‘Corporate Government Country Assessment-Arap Republic of Egypt’’, Report on Observance of Standards and Codes, The World Bank, September, www.worldbank.org/ifa/ rosc_cg.html. Nelson, C. and C. Plosser (1982), ‘‘Trends and random walks in macroeconomic time series’’, Journal of Monetary Economics, Vol. 10, pp. 139–162. Porter, R. and S. Ranney (1982), ‘‘An eclectic model of recent LDC macroeconomic policy analyses’’, World Development, Vol. 10(9), pp. 751–765.

Financial Development and Economic Growth in the MENA Countries

133

Porzecanski, A.C. (1979), ‘‘Pattern of monetary policy in Latin America’’, Journal of Money, Credit, and Banking, Vol. 11(4), pp. 427–437. Suleiman, N.M. (2000), ‘‘Corporate governance in Islamic banks’’, Arab Gateway, Vol. 16(December), pp. 98–116. Appendix Table A.1. Country Algeria 68–86 Egypt 54–90 Jordan 68–90 Kuwait 68–90 Morocco 63–90 Saudi Arabia 66–90 Syria 62–89 Tunisia 67–90

Estimates of the monetary policy reaction function

Constant

Dirt1

Dpt1

0.13* (2.369) 0.22* (3.24) 0.069 (1.68) 0.020 (0.39) 0.08* (3.98) 0.10 (3.45) 0.12 (4.43) 0.11* (4.92)

0.058* (1.329) 0.015 (0.16) 0.021 (0.55) 0.025 (0.27) 0.0053 (0.29) 0.13* (2.83) 0.028** (1.54) 0.068* (2.62)

0.46 (0.62) 1.46** (1.35) 0.68* (1.97) 1.59** (1.32) 0.23 (0.72) 0.064* (6.26) 0.30 (1.11) 1.14* (2.61)

Dyt1

Dgovt1

0.086 0.63* (0.33) (1.73) 0.58 0.31 (0.74) (0.69) 0.39* 0.089 (1.75) (0.60) 0.083 0.28 (0.73) (0.79) 0.14 0.081 (0.63) (0.83) 0.14* 0.043 (2.07) (0.60) 0.044 0.14 (0.26) (0.80) 0.078 0.35** (0.43) (1.64)

dum1

R2

0.033 (0.57) 0.24* (2.19) 0.018 (0.42) 0.04 (0.08) 0.07* (2.70) 0.0004 (0.013) 0.10* (2.62) 0.067* (2.88)

0.39 – 0.69 – 0.46 – 0.54 – 0.34 – 0.93 – 0.40 – 0.49 –

Notes: Empirical Model: Dmoneyt ¼ a0 þ a1 Dirt1 þ a2 Dpt1 þ a3 Dyt1 þ a4 Dgovt1 þ a5 dumt1 þ Zm ; t-ratios are in parentheses, *and **denote statistical signiﬁcance at the 5 and 10% levels, respectively.

134

Table A.2.

Estimates of the output and price equations

Country

Variable

Constant

Dem

Dms

Dgov

Dex

Doilpt

dumt

R2

Algeria 68-86

Dy

0.057** (1.52) 0.026 (0.79) 0.11 (0.60) 0.14* (2.43) 0.11* (1.87) 0.016 (0.65) 0.0078 (0.82) 0.012 (0.99) 0.043 (0.51) 0.018 (0.40) 0.12 (1.15) 0.051** (1.59)

0.56 (2.59) 0.058 (0.30) 0.44 (0.60) 0.18 (0.68) 0.82* (2.50) 0.54* (3.76) 0.62 (0.92) 0.16* (1.93) 1.11 (1.13) 0.44 (0.84) 0.081 (0.26) 0.52* (5.19)

0.64* (4.94) 0.21* (1.85) 0.46 (0.67) 0.41 (1.15) 0.52* (1.74) 0.25* (1.89) 0.31 (0.57) 0.071 (1.088) 0.23 (0.92) 0.056 (0.42) 0.26 (0.23) 0.24 (0.69)

0.12 (0.70) 0.16 (1.11) 0.68 (0.51) 0.58* (1.95) 0.0028 (0.018) 0.028 (0.42) 0.19 (0.39) 0.13* (2.13) 0.04 (0.36) 0.013 (0.23) 0.058 (0.20) 0.013 (0.14)

0.11 (0.75) 0.069 (0.54) 0.21 (0.62) 0.25 (1.19) 1.037 (1.19) 0.26 (0.67) 0.53 (0.57) 0.051 (0.44) 0.11 (0.57) 0.082 (0.80) 0.055 (0.082) 0.027 (0.13)

0.22* (9.68) 0.03** (1.52) 0.053 (0.32) 0.027 (0.36) 0.066** (1.68) 0.069* (3.98) 0.62* (5.32) 0.027* (1.88) 0.027 (0.78) 0.057* (3.11) 0.56* (4.13) 0.045 (1.053)

0.014 (0.56) 0.036** (1.57) 0.063 (0.18) 0.14* (3.33) 0.030 (0.57) 0.04* (1.72) 0.04 (0.08) 0.04 (0.08) 0.066 (1.08) 0.029 (0.88) 0.074 (0.74) 0.0062 (0.20)

0.94

Dp Dy Dp Jordan 68-90

Dy Dp

Kuwait 68-90

Dy Dp

Morocco 63-90

Dy Dp

S.Arabia 66-90

Dy Dp

0.50 0.95 0.58 0.84 0.79 0.84 0.21 0.74 0.68 0.81

Hasan Ersel and Magda Kandil

Egypt 54-90

0.62

Syria 62-89

Dy Dp Dy Dp

0.92 (1.071) 1.61* (2.12) 0.36** (1.58) 0.018 (0.15)

0.094 (0.22) 0.21 (0.56) 0.41* (2.098) 0.091 (0.92)

0.22 (1.002) 0.093 (0.48) 0.17 (1.12) 0.19* (2.51)

0.60* (1.87) 0.83* (2.94) 0.11 (0.84) 0.076 (1.17)

Notes: Empirical Models: Dyt ¼ b0y þ b1y Dmoneyt þ b2y Dgovt þ b3y Dext þ b4y Doilpt þ b5y dumt þ Zy Dpt ¼ b0p þ b1p Dmoneyt þ b2p Dgovt þ b3p Dext þ b4p Doilpt þ b5p dumt þ Zp t-ratios are in parentheses *and

**

denote statistical signiﬁcance at the 5 and 10% levels, respectively.

0.14* (1.70) 0.057 (0.77) 0.081* (2.86) 0.0082 (0.57)

0.10** (1.46) 0.22* (3.64) 0.014 (0.76) 0.027* (2.90)

0.60 0.62 0.67 0.67

Financial Development and Economic Growth in the MENA Countries

Tunisia 67-90

0.11 (0.94) 0.23* (2.24) 0.037 (1.15) 0.020 (1.24)

135

Hasan Ersel and Magda Kandil

136

Table A.3. Correlation coefficients between parameters of the policy reaction function and the real and inflationary effects of monetary policy Parameters of the Policy Reaction Function (1) a1 (2) a2 (3) a3 (4) a4

Output Response to Anticipated Monetary Growth 0.17 (0.68) 0.45 (0.26) 0.059 (0.89) 0.11 (0.80)

Output Response to Monetary Shocks 0.19 (0.65) 0.58 (0.14) 0.11 (0.79) 0.14 (0.74)

Price Response to Anticipated Monetary Growth 0.045 (0.92) 0.27 (0.52) 0.40 (0.33) 0.02 (0.96)

Price Response to Monetary Shocks 0.24 (0.57) 0.57 (0.14) 0.84 (0.0092) 0.16 (0.70)

Notes: Empirical Models: Domneyt ¼ a0 þ a1 Dirt1 þ a2 Dpt1 þ a3 Dyt1 þ a4 Dgovt1 þ a5 dumt1 þ Zm Dyt ¼ b0y þ b1y Dmoneyt þ b2y Dgovt þ b3y Dext þ b4y Doilpt þ b5y dumt þ Zy Dpt ¼ b0p þ b1p Dmoneyt þ b2p Dgovt þ b3p Dext þ b4p Doilpt þ b5p dumt þ Zp Reported statistics measure correlations coefﬁcients between parameters of the monetary policy reaction function, ai i ¼ 1, y, 4 and the real and inﬂationary effects of anticipated and unanticipated monetary growth. The probability of zero correlation is in parentheses.

CHAPTER 5

Labor Markets and Economic Growth in the MENA Region$ Christopher A. Pissarides and Marie Ange Ve´ganzone`s-Varoudakis Abstract The labor market plays an important role in economic development through its impact on the acquisition and deployment of skills. This paper argues that countries in the Middle East and North Africa (MENA) region failed to deploy human capital efficiently despite high levels of education because of a large public sector, which has distorted incentives, and because of excessive regulation in the private sector. The education system is geared to the needs of the public sector, so the acquired skills are inappropriate for growth-enhancing activities. Excessive regulation of the private sector further removes the incentives for employers to recruit and train good workers. As a result, the MENA countries found it difficult to adapt to new conditions in the 1990s and their rate of productivity growth fell to very low levels. The group as a whole failed to keep up with countries that used to be at a comparable level of development such as East and South-East Asia.

Keyword: Middle East and North Africa economies, public sector employment, employment and growth, education and growth JEL classifications: J45, E24, I20, O53

$ A ﬁrst version of this paper was written by Christopher Pissarides in 1999 for the Growth Development Project of the World Bank as a ‘‘thematic’’ paper for the MENA region. The current version is fully revised and updated, following comments from anonymous referees.

CONTRIBUTIONS TO ECONOMIC ANALYSIS VOLUME 278 ISSN: 0573-8555 DOI:10.1016/S0573-8555(06)78005-1

r 2007 ELSEVIER B.V. ALL RIGHTS RESERVED

138

Christopher A. Pissarides and Marie Ange Ve´ganzone`s-Varoudakis

1. Introduction: the growth context The labor market is given a key role in the literature on economic growth. This may not appear at ﬁrst to be the case, as many models do not refer explicitly to the structure of labor markets. But on deeper examination of the most popular recent models, it is invariably found to be the case. A variety of ‘‘engines of growth’’ are usually discussed in the growth literatures, which are directly or indirectly related to human capital and the implementation of new ideas. The engine of growth in the baseline Solow model is unspeciﬁed labor-augmenting technology; what can labor-augmenting technology be if it is unrelated to human capital? In other models it is the ability to introduce new products, namely the knowledge on how to do new things. In addition, in others it is explicitly ‘‘human capital’’ without going into the specifics of what human capital does to output growth and how. The conclusion reached from a reading of the recent growth literature is that if we are to understand growth and development, we need to understand the creation and deployment of human capital. Human capital is created and put into use in labor markets. The structure of the labor market is therefore critical for the quantity and quality of human capital that is created and for the uses to which it is put. The structure of the market will determine, for example, how much human capital is put into growthenhancing activities and how much into other activities such as redistribution. It will also determine what types of human capital will be required in different environments. Yet, despite the wide appeal of the recent growth literature and the large number of economists who have been attracted to it, not many labor economists have switched from their traditional preoccupations to the study of growth. Research in growth has become the domain of macroeconomists whose data on labor markets amounts to two or three aggregate series – usually for employment, schooling and participation rates. As a result, progress in the integration of labor market institutions with aggregate growth has been slow. In fact, it has not progressed much beyond the insights that generated the initial interest in growth theory. Looking at what data macroeconomists have on labor markets, and what propositions have been put forward by growth theorists for the link between labor market outcomes and growth, it becomes obvious that not much progress can be made within the current cross-country research agenda.1 Deeper country research is needed that pays attention to the institutional structure of the country in question and to the links between human capital, the institutional structure and the growth outcomes. In light of this, research on growth and the labor market in the MENA region has to begin with an examination of the labor market institutions in the

1

Similar conclusions were reached by Topel (1999) in his survey of research on labor markets and growth.

Labor Markets and Economic Growth in the MENA Region

139

MENA countries that are likely to inﬂuence growth outcomes. Unfortunately, past literature provides little guidance in this respect. There is virtually no literature explaining which labor market institutions are likely to be good for growth and which are likely to be bad. Of course, economists have views on the matter, but there is no consensus founded on solid empirical research. For example, the thorough and comprehensive books on growth by Barro and Sala-iMartin (1995) and Aghion and Howitt (1998) do not mention a single labor market institution that might inﬂuence growth. But in many variants of their models they do devote space to what they consider the key to growth, human capital. The same can be said for the recent survey on labor markets and economic growth by Topel (1999). The survey is devoted to human capital and growth, but not to labor market institutions. Of course, this reﬂects the current state of the literature, not an omission by the authors of those works.2 In the case of the MENA region, although some recent studies analyze quite extensively the functioning of the labor market, there is very little discussion on the connections between the labor market and growth.3 Country research on growth can make a real contribution to our understanding of growth if it can produce new data that can identify labor market institutions that are conducive to the creation and deployment of good-quality human capital and institutions that are wasteful of human resources. The purpose of this ‘‘thematic’’ paper is to point out the links between human capital and growth that need empirical veriﬁcation and discuss some summary data for the MENA region that might help point directions for research on individual countries. The role that human capital occupies in growth theory and its likely empirical contribution in the MENA region is ﬁrst discussed. It is shown that despite rapid growth in human capital resources, it is unlikely that human capital has contributed significantly to growth in this region. The challenge facing research in this region is to ﬁnd the reasons for the apparent wasteful use of human capital. We examine the sectoral composition of employment and the institutional structure that governs wage determination and employment and make some suggestions about the role of each and the directions for future research. 2. Human capital and economic growth 2.1. Human capital definitions Human capital is created by formal education and formal training as well as by informal learning mechanisms. Each time someone develops the ability to do

2

More recently, there has been increased emphasis on the role of institutions in growth and their empirical measurement. See, for example, Acemoglu et al. (2005). 3 See, in particular, World Bank (2004) for a study of the functioning of labor markets in the MENA region and Agenor et al. (2003) for a more quantitative approach to the study of labor market institutions and reforms and outcomes.

140

Christopher A. Pissarides and Marie Ange Ve´ganzone`s-Varoudakis

something new, he or she increases his or her human capital. Of course, measuring human capital in its full dimension is an impossible task. For this reason, the literature usually conﬁnes itself to measuring the years of schooling in the working population and using the outcome as a proxy for all human capital in the country. As a ﬁrst step in the study of labor markets and growth we therefore need a series for the stock of human capital (years of schooling) in the labor force or the population of working age. This can be done by employing an inventory method to enrollment data, provided one or more points of reference are available for the stock from labor force surveys. The stock is augmented by the years of schooling of school leavers and depreciates by the education of those leaving the labor force. Data for the stock of human capital for the population of working age are also useful in making inferences about the likely changes in the supply of labor. A country with large amounts of human capital outside the labor force (e.g., with many educated women who are not participating) is more likely to experience changes in its participation rates than another. Data for 5-year periods during 1960–1999 are available (Barro and Lee, 2000) and these data could provide a starting point for the research on the connection between human capital and growth in MENA. Table 1 reproduces some summary statistics for a selection of MENA countries, as well as by region. The main feature of these statistics is the steady growth of education in all the MENA countries of our sample throughout the period. In the 1960s, MENA educational attainment was one of the lowest in the world with an average of 1.6 years of schooling for each adult over the age of 15 years. This low attainment was comparable to the low attainment in the countries of Sub-Saharan Africa and South Asia. By 1999, however, the MENA region had closed the educational gap with the more advanced developing economies. With an average of 5.8 years of schooling, MENA is far ahead of South Asia and Sub-Saharan Africa, and less than one year behind Latin America and East Asia. 2.2. Growth theory What speciﬁc role does growth theory give to human capital? Modern growth theory appears to give, at one and the same time, a large role to human capital but does not say enough about it. In some models, such as the Lucas (1998) model, human capital drives growth. Everything else – capital, output – adjusts endogenously to the accumulation of human capital. In other models, such as the imitation model of Rivera-Batiz and Romer (1991), it drives the imitation technology which ultimately determines the speed of convergence to the technological frontier. This model is of particular relevance to developing countries, especially those in proximity to more advanced trading partners. The MENA countries fall into this category in relation to the European Union. In the R&D models, such as the variant estimated by Benhabib and Spiegel (1994), human capital is the factor that is engaged in R&D and so its productivity determines the rate of growth. This model, however, is less relevant to MENA, as R&D activity is concentrated in a small number of advanced countries.

Labor Markets and Economic Growth in the MENA Region

Table 1.

141

Average number of schooling years, total population over 15 years old

Country

1960

1980

1999

Algeria Bahrain Egypt Iran Iraq Jordan Kuwait Syria Tunisia

1.0 1.0

2.7 3.6 2.3 2.8 2.7 4.3 4.3 3.6 2.9

5.4 6.1 5.5 5.3 4.0 6.9 7.1 5.8 5.0

Region Sub Saharan Africa East Asia Eastern Europe Developed Economies Latin America MENA South Asia

0.8 0.3 2.3 2.6 1.4 0.6 1960

1980

1999

1.7 3.4 4.5 6.7 3.5 1.6 1.5

2.5 5.3 6.3 8.3 5.1 3.7 2.9

3.5 6.7 7.2 9.5 6.2 5.8 4.2

Source: Barro and Lee (2000) and authors’ calculation from Barro and Lee (2000) for regional averages.

But whereas human capital is postulated to be the factor that plays these roles, it has not yet been possible to empirically test any model with sufﬁciently detailed data to arrive at a smaller set of possible links between human capital and growth. The macroeconomic models that have been estimated, mostly with crosscountry data, or with panel regressions making use usually of the Barro–Lee data set for human capital, cannot discriminate between these models.4 It can be argued that – as a ﬁrst step in our attempts to understand the connection between labor market structure and growth – research should concentrate on the relation between labor market structure and the contribution of human capital to GDP growth, without trying to discriminate between the different roles that have been attributed to human capital in the literature. Of course, if the research could discriminate between some roles it would be an added bonus. But the more urgent need (and the one that will be of more policy relevance) is to identify institutional structures that are good for the productive use of human capital and for encouraging more investments in it and structures that are poor in this respect.

4

The human capital by Nehru et al. (1993) have similar implications to those of the Barro–Lee data set where the two overlap.

142

Christopher A. Pissarides and Marie Ange Ve´ganzone`s-Varoudakis

To focus ideas, consider the neoclassical production function without human capital: y ¼ AK a Lb Human capital in some models inﬂuences the technological parameter A, through for example, R&D or imitation of more advanced countries’ technologies. Moreover, it augments the labor input L as a productive factor sometimes. In the ﬁrst class of models it inﬂuences, at least temporarily, the rate of growth of A, and hence of output. But in the second class of models it inﬂuences the level of output with the same coefﬁcient as the labor input. Cross-country regressions have come under criticism for a variety of reasons, but whatever their merits, they cannot distinguish between these two variants of the model – does the level of human capital inﬂuence the rate of growth of output or its level? In fact, hardly any robust results have been derived from these regressions about the contribution of human capital to growth, though when human capital does not show up as a significant inﬂuence in cross-country regressions most authors blame the data. Such is the conviction that human capital must be an important inﬂuence on growth. We will argue below that it is possible for human capital to have a high private rate of return but not contribute to growth, when the institutional structure of the labor market is such that ‘‘rent seeking’’ or other lessproductive activities yield a higher private return to the individual than do growth-enhancing activities. In some cases, the contribution of human capital to growth can be hampered by its low quality, or by its unsuitable nature, as when skill mismatches and market rigidities lead to the unemployment of qualiﬁed people. The fact that the stock of human capital has been trending up in most countries of the world but the rate of growth of output has not, gives an indication that if the level of human capital inﬂuences the rate of growth of output over certain periods of time, it is likely to be a temporary phenomenon. But if this were to lead to adoption of the other extreme view that the human capital stock inﬂuences output with the same coefﬁcient as labor, the role of human capital in production and the implied rate of return to human capital would be too small to be credible. The results of the cross-country empirical research are diverse, but they are not consistent with the hypothesis that human capital enters the production function with the same coefﬁcient as labor.5 The capital stock is usually treated as endogenous in growth models and driven by the savings rate in the economy. In the models which make human capital the engine of growth, the rate of growth of the capital stock eventually

5

See Topel (1999) for an evaluation and Barro and Sala-i-Martin (1995), Benhabib and Spiegel (1994) and Pritchett (1996) for empirical results and more discussion.

Labor Markets and Economic Growth in the MENA Region

143

converges to the rate of growth of human capital. In addition, the capital/labor ratio is also inﬂuenced by the features of the human capital production function. The prediction of growth models is that a country with more human capital will eventually have more physical capital as well. As a ﬁrst step, however, it might be worthwhile to explain the contribution of human capital to output growth net of the contribution of the capital stock, namely to the series log Ya log K, though ultimately a complete explanation of the contribution of human capital to growth will need to explain the dependence of investment on human capital.

2.3. Growth accounting Growth accounting exercises decompose the growth of output into growth owing to capital, employment and total factor productivity (TFP). Their information content is limited and the decompositions they arrive at should not be treated as research ﬁndings that should guide policy but as suggestive of further research. Yet occasionally growth accounting exercises come up with surprising (when viewed through the OECD lens, perhaps) facts and cause controversy, as the work of Young (1995) on growth accounting for South-East Asia did. For the MENA research on growth and labor, growth accounting exercises can shed light on the level and rate of growth of TFP in each country and their relation to human capital. More specifically, how can we account for the fact that the human capital stock has grown rapidly since 1960, yet output growth seems to have stagnated? Tables 2 and 3a show GDP growth rates for a selection of MENA countries and by region, gross and decomposed into growth in the capital stock, labor force and TFP. The share of capital in the TFP calculations is taken to be 0.4, although there are variations across countries and in deeper country research country-speciﬁc estimates could be given. The share of labor is correspondingly taken to be 0.6. In the absence of a time series for employment, we used a series for total labor force growth. TFP1 shows the results of the TFP calculations when the contribution of human capital is not netted out. The idea behind TFP1 is that human capital contributes to TFP growth; namely, that human capital is one of the factors that explains the path of TFP growth and not the level of output. TFP1 is calculated as the residual of the production function previously speciﬁed: log(TFP1) ¼ log(Y)0.4 log(K)0.6 log(L). In TFP2, however, we have treated human capital as a factor that improves the quality of the labor force and so we used data on human capital to augment the contribution of labor before deducting it from output growth. With a Cobb–Douglas production function this amounts to treating human capital as a factor of production. In this case, TFP2 is calculated as follows: log(TFP2) ¼ log(Y)0.4 log(K)0.6 log(L0 ) where L0 (the skilled labor) is the labor force (L) augmented by the number of years of schooling of the population over 15 years old (H), the rate of

144

Christopher A. Pissarides and Marie Ange Ve´ganzone`s-Varoudakis

Table 2. TFP calculations for a selection of MENA countries (average annual growth rates in %) Country Year

GDP

Capital stock

Labor force

TFP1

Skilled labor force

TFP2

Algeria 1970s 1980s 1990s

5.7 2.5 1.5

8.7 4.9 1.0

3.2 3.8 3.8

0.3 – 1.8 – 1.1

4.4 5.5 5.0

– 0.4 – 2.8 – 1.8

Egypt 1970s 1980s 1990s

8.0 4.9 4.3

9.8 8.8 3.4

2.1 2.5 2.9

2.8 – 0.1 1.2

4.5 4.5 4.2

1.4 – 1.3 0.5

Iran 1970s 1980s 1990s

0.7 3.8 4.0

12.8 2.6 1.3

3.0 3.0 2.3

– 6.2 1.0 2.1

4.3 4.2 3.7

– 7.0 0.3 1.2

Jordan 1970s 1980s 1990s

8.7 3.2 5.2

10.8 7.0 1.3

2.3 4.9 5.8

3.0 – 2.6 1.2

3.4 6.7 6.9

2.4 – 3.6 0.6

Morocco 1970s 1980s 1990s

5.6 3.6 2.2

8.8 4.9 3.5

3.2 2.6 2.5

3.0 – 2.6 1.2

4.1 3.4 3.4

– 0.4 – 0.4 – 1.2

Tunisia 1970s 1980s 1990s

7.4 3.7 4.8

7.0 4.6 3.3

3.6 2.7 2.9

2.4 0.3 1.7

5.1 3.7 4.0

1.5 – 0.4 1.0

Source: Authors’ calculations from World Bank data.

return to education being ﬁxed at 0.1.6 The results should be regarded as indicative of trends rather than accurate descriptions of reality, as data tend to be unreliable. There are more than one time series for GDP growth for these countries and they are not all consistent with each other, and the capital stock data sometimes exhibit implausible behavior. Tables 2 and 3a show that GDP growth experienced a sharp decrease in the 1980s, after the fall in oil prices. This has been the case in all the MENA countries in the sample (except Iran, Table 2) and in all regions (except South Asia, Table 3a). Both investment and employment decreased during the period,

6 Following Dasgupta et al. (2002), the exact way in which we augmented the labor input is L0 ¼ L exp(0.1 H) where H is the number of years of schooling.

Labor Markets and Economic Growth in the MENA Region

Table 3a. Region Year

145

TFP calculations by region (average annual growth rates in %) GDP

Capital stock

Labor force

TFP1

Skilled labor force

TFP2

Africa 1970s 1980s 1990s

3.7 2.2 2.7

5.5 3.1 2.0

2.5 2.7 2.6

0.0 – 0.7 0.3

3.1 3.4 3.2

– 0.3 – 1.1 0.0

East Asia 1970s 1980s 1990s

7.7 6.1 6.2

10.6 8.5 8.5

3.1 2.7 2.2

1.6 1.1 1.4

4.1 3.4 3.0

1.1 0.7 0.9

Eastern Europe 1970s 1980s 1990s

5.7 4.4 3.8

8.2 4.7 4.1

1.6 2.5 2.6

1.5 1.0 0.7

1.4 1.3 2.0

1.6 1.7 1.0

OECD 1970s

4.2

5.2

1.5

1.2

2.4

0.7

1980s 1990s

2.8 2.8

3.3 3.1

1.1 1.0

0.8 0.9

1.9 1.7

0.3 0.6

Latin America 1970s 1980s 1990s

4.7 1.0 3.2

6.4 3.1 3.3

2.9 2.8 2.7

0.4 – 1.9 0.3

3.6 3.5 3.2

0.0 – 2.3 0.0

MENA 1970s 1980s 1990s

6.1 3.8 3.7

9.6 6.1 2.3

2.8 3.3 3.3

0.6 – 0.6 0.8

5.0 5.4 4.3

– 0.7 – 1.9 0.2

South Asia 1970s 1980s 1990s

3.6 5.3 5.0

4.5 5.5 4.5

2.5 2.3 2.5

0.3 1.7 1.6

2.4 3.2 3.1

0.3 1.2 1.3

Source: Authors’ calculations from World Bank data.

especially the former. But these declines do not fully account for the decrease in GDP growth, so our calculations show a sharp fall in TFP growth which becomes negative in the majority of our MENA sample. In this regard, the MENA region did not perform better than Africa, although in terms of GDP and human capital growth the region as a whole outperformed Africa. In the 1990s, the MENA region experienced a small recovery of GDP and TFP growth, despite a further cut of investment projects (in the public sector in particular). The recovery, however, was not sufﬁcient to allow MENA to close the gap with the more advanced developing countries, with which the MENA region was catching up in the 1960s (East and South Asia in particular, Table 3a).

Christopher A. Pissarides and Marie Ange Ve´ganzone`s-Varoudakis

146

The independent TFP estimates of Nehru and Dhareshwar (1994) show a low TFP growth in all countries in the period 1960–1990 and, more strikingly, a negative overall TFP growth rate for the MENA region as a whole (see Table 3b). TFP growth in the MENA region compares poorly even with the rest of Africa. Our calculations are consistent with the Nehru and Dhareshwar estimates, but show that the negative results are even more striking when the contribution of human capital to TFP is deducted (TFP2 series, Table 3b). Because human capital accumulation was higher in the MENA region than in the rest of Africa, its performance net of human capital was even worse than in the earlier estimates that ignored it. This suggests a low contribution of human capital to growth in the region when compared with the rest of the world. The task faced by a researcher of economic growth in the MENA region is to explain why TFP growth in the region was so low, in view of the fact that other things equal, the low initial income should have returned the TFP growth rates to above the average of the world economy. The relatively high investments in human capital that took place in these countries since the 1960s should also have contributed to faster convergence dynamics, namely, faster TFP growth during the sample period. For research on labor markets and growth in particular, the task is to identify features of the labor markets of the MENA countries that contributed to the low return on human capital and the low overall TFP growth. Table 3b.

TFP calculations by regions (annual percentage changes)

Nehru and Dhareshwar (1994) Region Africa East Asia OECD Latin America MENA South Asia

1960–1990 ECM

FD

–0.8 0.5 0.5 –0.6 –1.2 0.3

–0.4 1.2 1.3 –0.1 –0.3 0.8

Authors’ Calculations Region Africa East Asia Eastern Europe OECD Latin America MENA MENA South Asia

1960–2000 (TFP1)

(TFP2)

0.0 1.3 1.5 1.3 0.1 0.5 1.1

–0.3 0.8 1.6 0.8 –0.3 –0.4 0.7

Note: Unweighted averages of annual country rates. ECM stands for ‘‘error-correction method’’ and FD for ‘‘ﬁrst-difference method.’’ Both sets of calculations ignore human capital.

Labor Markets and Economic Growth in the MENA Region

147

3. Labor market structure Human capital in the MENA region grew steadily throughout the period of low TFP growth (see Table 1). Therefore, even in the absence of a careful statistical analysis and despite the other factors such as economic reforms, macroeconomic and political instability or governance that played a role in slowing down TFP growth, it is obvious that human capital will not be able to contribute much to growth in country regressions. This may be due to the fact that human capital in MENA has suffered either from low quality or from a misallocation that diverted it from employment in growth-enhancing activities. A lot of it must have stood idle, engaged in rent-seeking or less-productive activities (not properly recorded in national income statistics such as the running of social services). An analysis of this issue, with a view to ﬁnding ways to improve the situation if misallocations are found, requires an examination of the labor market’s static efficiency. 3.1. Static efficiency Static efﬁciency investigates the allocation of labor across sectors of the economy. The concern in the present context is whether the allocation of skilled labor is the one most likely to maximize the country’s growth potential. Table 4a shows the allocation of labor across some broadly deﬁned sectors. Manufacturing occupies a smaller fraction of the labor force than in other industrializing countries, and this is compensated for in the MENA region by a bigger public sector (Table 4b). Tables 5a and 5b show that MENA’s government and public sector (excluding health and education in the case of Table 5b) employ a bigger fraction of the non-agricultural labor force than in any region outside Africa. Another set of more detailed statistics shows that in Egypt (Table 6), government employs more than half of all degree holders in the country, with public enterprises also employing a large fraction. Public sector employment of this magnitude clearly interferes with the static efﬁciency of the labor market and deeper country studies need to investigate carefully the uses to which the public sector puts this human capital. Recent studies show that low productivity is exacerbated in the public sector of countries in the MENA region by an increasing overstaffing (World Bank, 2004). In the early 1990s, the share of underutilized workers in the public sector ranged from 17% (Algeria) to 21% (Egypt) and to even more in the oilexporting countries. This share, despite its substantial size, has increased recently to 35% in Egypt and 40% in Jordan. Berthelemy et al. (1999) estimated the average loss of GDP growth owing to public sector employment by making the admittedly strong assumption that the fraction of human capital employed in the administrative public sector does not contribute at all to growth. Their estimate is shown in the last column of Table 5b. The loss in the MENA and Sub-Saharan Africa regions is bigger than elsewhere by a large margin.

Christopher A. Pissarides and Marie Ange Ve´ganzone`s-Varoudakis

148

Table 4a.

Sectoral distribution of employment (% of total employment)

Country

Agriculture

Manufacturing & other Industries

Services

12 1 30 23 16 6 1 44 28 22 14 8

30 54 21 31 18 25 24 21 25 34 34 33

58 43 49 45 66 69 74 33 46 44 52 59

Algeria (1995) Bahrain (1994) Egypt (2000) Iran (1996) Iraq (1990) Jordan (1993) Kuwait (1988) Morocco (1999) Syria (1991) Tunisia (2001) WB & G (2000) UAE (2000) Source: World Bank (2004).

Table 4b.

Public sector employment (% of total employment)

Country

(1)

(2)

Algeria (1990–1999) Bahrain (1991–2001) Egypt (1988–1998) Iran (1986) Jordan (1987–1996) Kuwait (1989–2000) Morocco (1991–1999) Oman (1991–1999) Quatar (1986) Saudi Arabia (1992–1999) UAE (late 1980s) Tunisia (1997–2001) Yemen (late 1980s)

58 68 27 31 45 42 12 76 37 70 31 24 16

30 80 38 36 75 8 79 82 21

Note: (1) and (2) relate respectively to the ﬁrst and the second year in the country list above. Source: Shaban et al. (2001) and World Bank (2004).

Table 5a.

Size of government in the 1990s

Region Sub-Saharan Africa Asia Eastern Europe OECD Latin America MENA Source: Schiavo-Campo et al., (2003).

Employment (% total)

Wages (% GDP)

6.2 6 16 17.5 9 17.6

6.3 4.5 3.9 4 5 9.8

Labor Markets and Economic Growth in the MENA Region

Table 5b.

Share of public sector in non-agricultural employment and human capital

Region Sub-Saharan Africa Asia OECD Latin America MENA

Table 6.

149

Share of Public Sector

Estimated Loss GDP Growth (1985–1995)

32.9 19.8 20.6 17.7 31.7

8.8 4.9 5.1 4.3 8.4

Educational attainment by employment sector in Egypt (1988)

Sector All sectors Government Public enterprise Private agricultural Private non-agricultural Total

Below Intermediate Intermediate and Above University and Above Total 76.9

14.6

8.5

7.0 5.3 57.6 29.7 100

47.2 15.8 9.6 24.5 100

55.6 14.7 2.6 23.0 100

100 16.9 7.6 45.9 28.4 100

Source: Shaban, et al. (1994), Table 11.

Another source of misallocation of skilled labor comes from unemployment of highly qualiﬁed people. Unemployment rates are generally high in the MENA region (Table 7a) but more importantly, the unemployment rates of people in the middle to the upper end of the educational distribution are even higher (Table 7b). The waste of human capital through the high unemployment rates must be a contributory factor to the low overall contribution of human capital to growth. Interestingly, despite the high unemployment rates, entrepreneurs in these countries regularly cite the lack of labor with suitable skills as an important constraint to hiring. The combination of high skilled unemployment and skill shortages at the industrial level provides support to those who have claimed that the education systems in the MENA region have mostly been geared to the needs of the public sector. The high wages and other job advantages (such as job security, worker protection and social allowances) offered by the pubic sector led to educated workers queuing for public sectors jobs, and so to the absence of pressure to reform the educational system according to the needs of industry. How did the MENA countries ﬁnd themselves in a situation of large public sectors and misallocation of their human capital resources? Country experiences differ and a full analysis requires an examination of the institutional structure of the countries in question and the historical context. For the region as a whole, however, there has been one big missed opportunity. Historically, the biggest inﬂuence in these countries has been the oil boom of the 1970s, which lasted up to about 1982, and which enriched the public sector and led to the large

Christopher A. Pissarides and Marie Ange Ve´ganzone`s-Varoudakis

150

Table 7a.

Unemployment rates

Country

% Labor Force

Algeria (2000) Bahrain (2001) Egypt (2000) Jordan (2000) Iran (2001) Kuwait (2003) Lebanon (1997) Morocco (2002) Oman (1996) Quatar (2002) Saudi Arabia (1999) Syria (2001) Tunisia (2001) UAE (1999) W-B & Gaza (2001) Yemen (1999)

29.8 13 9 14 13.8 2.6 8 22 12 12 7.5 11 15.4 2.5 25 11.5

Source: World Bank (2004).

Table 7b.

Unemployment rates (by educational level)

Country

None

Primary

Secondary

Tertiary

All

Algeria (1995) Egypt (1998) Jordan (1991) Morocco (1999) Oman (1996) Tunisia (1997)

9.6 4.1 8.2 9.4 5.6 10.2

30.9 5.7 8.7 26.3 13.4 20.8

30.9 22.4 25.8 32.4 24.8 15.4

68.4 9.7 21.5 37.6 2.8 6.4

27.9 11.4 14.4 15.6 10.8 15.7

Source: Shaban et al. (2001) and World Bank (2004).

education expansion. But the resources gained during the oil boom were used to expand and protect the public sector from market competition and not spent in a way that was favorable to growth. They contributed to the large expansion of public sector employment, to the misallocation of resources in the public educational systems and to the introduction of other institutional rigidities that the economies could afford (perhaps) when the oil revenues were abundant, but not when they dried up.7

7

See Pissarides (1993) for more discussion of the historical context and the relation between oil revenues and public sector expansion. More recent discussion about the role of reforms can be found in Dasgupta et al. (2002) and more discussion about the role of social contracts, political regimes and wars and their inﬂuence on the economy of the countries in the MENA region can be found in World Bank (2004).

Labor Markets and Economic Growth in the MENA Region

151

3.2. Institutional foundations Labor market institutions inﬂuence the allocation of resources. In the present context the question is whether human capital is employed in growth-enhancing activities or elsewhere. There is a risk, when writing about labor market institutions to drift too far from the growth context. The focus of our discussion here is on the institutions that are likely to inﬂuence the allocations of labor in growth versus non-growth enhancing activities, even if other institutions appear more important at ﬁrst sight. Two broad institutional structures appear most relevant for the allocation of resources and growth in the MENA region. First, wage setting institutions and their implications for relative wages across sectors. Do high public sector wages explain why the public sector is so large in the MENA region? Table 8 shows that public sector wages relative to the private sector ones in the MENA region are higher than anywhere else (in addition, the public sector offers more non-wage beneﬁts). Although this difference may partly reﬂect the higher educational background of the labor force working in the public sector, it also indicates the presence of distortions. If it reﬂected only the higher educational attainment of workers in the public sector there would have been no queues to enter the public sector and no higher unemployment of more educated workers. As it is, the public sector is obviously not competing with industry for qualiﬁed labor – it rather sets the agenda for wages and rations employment. The high wages and high non-wage beneﬁts that we already mentioned constitute an important incentive for qualiﬁed workers to enter the public sector, with the poor results for the contribution of human capital to overall growth that we have already noted. In addition to the waste of human capital in less-productive public sector employment, the large size of public employment in these countries inﬂates the public sector wage bill. The wage bill has to be met from the public budget and the debt and tax implications for private sector activity impose another large burden on the economy that works against growth. Second, the institutional framework for hiring and ﬁring employees and more generally the framework that regulates the employment relationship, including Table 8.

Central Government wages (early 1990s)

Region

Sub-Saharan Africa Asia ECA Latin America OECD MENA Overall Note: Table 3 of MNSED (1999).

Central Government Wage Bill (% of GDP)

Ratio of Public Sector to Private Sector Wages

6.7 4.7 3.7 4.9 4.5 9.8 5.4

1.0 0.8 0.7 0.9 0.9 1.3 0.8

152

Christopher A. Pissarides and Marie Ange Ve´ganzone`s-Varoudakis

Table 9a.

Labor market regulation in developing regions

Region

Composite Index

Sub-Saharan Africa East Asia Europe and Central Asia Latin America MENA South Asia

1.45 1.6 1.95 2.05 1.65 1.25

Source: Doing Business Database (2003).

Table 9b.

Labor market regulation in the MENA countries

Country

Composite Index

Algeria Egypt Jordan Iran Lebanon Morocco Syria Tunisia

1.5 1.85 1.55 1.9 1.2 1.35 1.3 1.7

Source: Doing Business Database (2003).

the legal framework for standards at work, minimum wages and trade union recognition and powers. This institutional framework is likely to be the result of government policy but workers’ organizations may have their own rules on hiring and ﬁring. In the MENA region, labor market regulations have historically been stringent and are still too tight compared with the other regions of the developing world, although not as high as the formerly planned economies or Latin America (Table 9a). The majority of the countries in the region are affected by this situation, which has introduced rigidities in the labor market (Table 9b). At the micro level rigidities of this kind lead to low productivity and removal of the incentive to innovate and start new businesses. At the macro level these negative consequences translate to slow growth, and to inertia in response to macroeconomic shocks, unemployment and misallocation of labor. More indepth analysis is required of the implications of the regulation of employment and business for growth than can be pursued here, and the recent availability of the World Bank database provides a basis on which country papers can build. 3.3. Market flexibility It is clear from the results obtained so far on TFP growth in the MENA countries (and elsewhere) that the way that we deﬁne and measure TFP gives a

Labor Markets and Economic Growth in the MENA Region

153

cyclical TFP series with downswings that can last for many years. This is a reﬂection of the well-known fact that labor productivity is procyclical. But unlike OECD economies, which are diversiﬁed and suffer from regular cyclical shocks of small intensity, developing countries have suffered mostly from more pronounced shocks associated with well-deﬁned one-off events such as the debt crisis in Latin America, the fall of oil prices in MENA and the ﬁnancial crisis in South-East Asia. The intensity and persistence of the resulting downswings in TFP growth are directly related to the ability of the economy to adjust to new long-term conditions with the minimum of waste. This is where the issue of market ﬂexibility becomes key to growth. Consistent with this view it has been shown that, where labor markets are more rigid, countries tend to experience deeper recessions before adjustment as well as slower recoveries (Forteza and Rama, 2001). Over the last 20 years, in the MENA region, the main economic shocks that seem to have affected TFP growth are the fall in the price of oil after 1982 and the Gulf War a decade later. Both these events reduced GDP growth for many years. The speed of response of the economy to the after-shock situation was generally slow, a fact that is at least partly due to the rigid institutional structure of the labor market. As with static efﬁciency, the institutional features that are likely to inﬂuence the dynamic adjustment are likely to be the wage setting institutions and how ﬂexible they are in allowing wages to respond to changes in market conditions; the hiring and ﬁring restrictions that govern the speed of labor turnover and the ease with which workers can migrate and change sector of employment. Of course, the economy sooner or later adjusts to a new steady state. The question in the present context is how fast it adjusts and whether policy can help it adjust faster and with less waste. In the MENA region the adjustment to the oil shocks seems to have been slower than the adjustment to crises in other regions; for example, both Latin American and South-East Asian countries came out of their respective debt and ﬁnancial crises faster than did the MENA countries out of their oil crisis. Although there has been some downward adjustment of real wages in the 1990s, the dominant role of government as employer has slowed down the adjustment. It requires a lot more research at the country level, however, to determine whether the labor market structure was responsible for the slow adjustment. In particular whether one or more of the three institutional features of the preceding paragraph – wage setting, employment ﬂexibility and migration – played a role in slowing down the adjustment to the oil shocks. No general claims can be made for the region as a whole since the institutional structure is not well documented at the regional level. 4. Acquisition of skills The discussion so far focused on the allocation of human capital between alternative uses, some of which are growth-enhancing and some less so. But

154

Christopher A. Pissarides and Marie Ange Ve´ganzone`s-Varoudakis

equally important is the question of the acquisition of skills, and how effective are labor markets in generating a large amount of skills that are useful for growth. This requires ﬁrst a discussion of the educational and training system and how it is organized in each country. It includes the coverage and quality of the education system, who pays for education and what incentives there are for individuals to engage in training. Has human capital grown in some countries because the state ﬁnanced it and encouraged particular skills or because private initiative did it? Moreover, what can be said about the quality of the human capital stock? The free market usually signals the need for skills through the relative wage system, namely, through the private rate of return to education and training. Are labor markets in the MENA region effective in this signaling function? Do individuals respond to the signals? The answers to these questions touch on the dynamic efficiency of the labor market. Under dynamic efﬁciency the emphasis is shifted to the efﬁciency of the labor market in the provision of adequate resources for education and training and to the recruitment of enough people for these purposes. In MENA, these questions need special attention because of the distortionary labor market institutions that we have highlighted. The outcome appears to be low-quality education8 and skill mismatches due to the bias toward public sector needs. The distortions in relative wages that we have highlighted increase the private rate of return to skill without a corresponding increase in its social rate of return, and its implications for dynamic efﬁciency are an issue that needs further investigation at the country level. Statistics on rates of return to schooling by sector of activity – available for different countries – conﬁrm the attractiveness of the public sector. This is true at all levels of education, with the exception, recently, of Jordan and Yemen (Table 10).9 5. Conclusions Human capital occupies a central role in modern thinking about growth. Despite a large literature on the matter, however, there is a lot to be learned: there is no consensus on its role in growth and development, probably because this role varies across different institutional settings and national environments. The labor market is the place where human capital is created and deployed. This paper has argued that the study of the links between labor markets and growth should concentrate on a study of labor market inﬂuences on the quantity,

8

See for example, United Nations Development Program (2002, 2003), human development reports. It has not been possible to ﬁnd easily accessible data sources for the quality of the human capital stock for the region as a whole. Psacharopoulos (1985) mentions virtually no MENA countries in his sample. Data on quality on a country basis are reported in the updated paper by Barro and Lee (2000), but the regional coverage for MENA is not satisfactory.

9

Labor Markets and Economic Growth in the MENA Region

Table 10.

155

Rate of return to schooling (% per year) Egypt

Morocco

Jordan

Yemen

1988

1998

1991

1999

1997

1997

Primary Male (public) Male (private) Female (public) Female (private)

8.2 2.3 1.9 0.9

6.4 3.6 5.3 7.2

12.4 3 28.2 8.5

6.1 3.4 10.5 9.4

3.5 2 –3.9 14.7

2.7 2.7 5.1 8

Lower secondary Male (public) Male (private) Female (public) Female (private)

7 2.5 7.7 3.2

4.9 4.4 8.2 –11.2

10.7 6.4 22.3 13.9

8.2 6.3 13.4 10

2.9 5.5 5.2 9.8

2.7 2.7 3.7 7.4

Upper secondary (General) Male (public) Male (private) Female (public) Female (private)

8.6 6.3 8.6 3.8

8.8 7.3 9.7 –1.5

10.6 10.4 18.1 16.4

8.8 7.7 12.1 11

2.8 6 4.6 10.4

2.2 2.2 3.9 12.1

Upper secondary (vocational) Male (public) Male (private) Female (public) Female (private)

9.6 5.3 7.9 4.4

7.2 5 9.6 4.9

8.4 6.9 16.5 11.1

6.8 5.8 11.9 11.3

3.8 3.2 4.3 8.6

3.3 3.3 4.3 10.7

10.1 8.5 8.9 9.1

8.8 7.3 10.7 10.9

10.8 12.5 15 15.2

8.9 9.5 12.8 9.3

4.6 10.2 6.8 12.9

3.8 5.2 4.4 6.8

University Male (public) Male (private) Female (public) Female (private)

Source: Assaad (2002) and World Bank (2004).

quality and productivity of human capital. This requires an investigation at the level of individual countries of the institutions that inﬂuence the acquisition and employment of human capital and their effectiveness in enhancing growth. Our investigation of broad trends in the MENA region has revealed that there has been fast growth in the acquisition of skills through general education. But following the oil crises of the 1980s, the countries in the region have been unable to utilize their human resources to overcome the negative consequences of the crises for output and growth. In this respect, human capital in the MENA region has been less successful in contributing to growth than elsewhere, e.g., in East or South Asia. We argued that large and inﬂexible public sectors, wage inﬂexibility and excessive labor market regulation may be some reasons behind this failure. This opens up a number of issues about reform which require more and deeper research than can be given here on the role of labor market institutions at the country level.

156

Christopher A. Pissarides and Marie Ange Ve´ganzone`s-Varoudakis

References Acemoglu, D., S. Johnson and J. Robinson (2005), ‘‘Institutions as the fundamental cause of long-run growth’’, in: P. Aghion and S. Durlauf, editors, Handbook of Economics of Growth, Amsterdam: North-Holland forthcoming. Agenor, P.R., Nabli, M.K., Youssef, T.M. and Jensen, H.T., (2003), ‘‘Labor market reforms, growth and unemployment in labor-exporting countries in the Middle East and North Africa’’, World Bank (unpublished). Aghion, P. and P. Howitt (1998), Endogenous Growth Theory, Cambridge, MA: MIT Press. Assaad, R. (2002), ‘‘The transformation of the Egyptian labor market: 1988–98’’, in: R. Assaad, editor, The Egyptian Labor Market in an Era of Reform, Cairo: American University in Cairo Press. Barro, R. and Lee, J.-W., (2000), ‘‘International data on educational attainment: updates and implications’’, Working Paper No. 42, Center for International Development, Harvard University, Harvard. Barro, R.J. and X. Sala-i-Martin (1995), Economic Growth, New York: McGraw-Hill. Benhabib, J. and M.M. Spiegel (1994), ‘‘The role of human capital in economic development’’, Journal of Monetary Economic, Vol. 34, pp. 143–173. Berthelemy, J.-C., C.A. Pissarides and A. Varoudakis (1999), ‘‘Human capital and growth: the cost of rent seeking activities’’, in: T. De Ruyter van Steveninck, editor, Economic Growth and its Determinants, Dordrecht: Kluwer Academic Publishers. Dasgupta, D., Keller, J., Srinivasan, T.G., 2002. ‘‘Reform and elusive growth in the Middle-East – what has happened in the 1990s?’’, MENA Working Paper Series No. 25, June, The World Bank. Doing Business Database, 2003. Washington, DC: World Bank. Forteza, A. and M. Rama (2001), ‘‘Labor market rigidity and the success of economic reforms across more than 100 countries’’, Policy Research Working Paper, No. 2521 (January), World Bank, Washington, DC. Lucas, R.E. (1988), ‘‘On the mechanics of economic development’’, Journal of Monetary Economics, Vol. 22, pp. 3–42. MNSED (1999), ‘‘Workers in crisis in the MENA region: an analysis of labor market outcomes and prospects for the future’’, Unpublished paper, World Bank. Nehru, V., Dhareshwar, A., (1994), ‘‘New estimates of total factor productivity growth for developing and industrial countries’’, Policy Research Working Paper 1313, The World Bank, Washington, DC. Nehru, V., E. Swanson and A. Dubey (1993), ‘‘A new database on human capital stock: sources, methodology and results’’, Policy Research Working Paper No. 1124, International Economics Department, World Bank.

Labor Markets and Economic Growth in the MENA Region

157

Pissarides, C.A. (1993), Labor markets in the Middle East and North Africa’’, Middle East and North Africa Discussion Paper Series No. 5, The World Bank, Washington, DC. Pritchett, L. (1996), ‘‘Where has all the education gone?’’, World Bank Policy Research Paper No. 1581. Psacharopoulos, G. (1985), ‘‘Returns to education: a further international update and implications’’, Journal of Human Resources, Vol. 20, pp. 583–604. Rivera-Batiz, L.A. and P.M. Romer (1991), ‘‘Economic integration and endogenous growth’’, Quarterly Journal of Economics, Vol. 106, pp. 531–555. Schiavo-Campo, S., de Tommaso, G., Mukherjee, A., 2003. ‘‘An international statistical survey of government employment and wages,’’ World Bank Policy Research Working Paper, No. 1806, World Bank, Washington, DC. Shaban, R., R. Assaad and S. Al-Qudsi (1994), ‘‘Employment experience in the Middle-East and North Africa’’, Working Paper No.199401, Economic Research Forum, Cairo. Shaban, R.A., A.R. Assaad and S. Al-Qudsi (2001), ‘‘Employment experience in the Middle East and North Africa’’, in: D. Saleha-Eshfahani, editor, Labor and Human Capital in the Middle East, Reading, UK: Ithaca Press. Topel, R. (1999), ‘‘labor markets and economic growth’’, in: O.C. Ashenfelter and D. Card, editors, Handbook of Labor Economics, Amsterdam: NorthHolland. United Nations Development Program (2002, 2003), Arab Human Development Report, United Nations, New York. World Bank (2004), Unlocking the Employment Potential on the Middle East and North Africa. Toward a New Social Contract, Washington, DC. Young, A. (1995), ‘‘The tyranny of numbers: confronting the statistical realities of the East Asian growth experience’’, Quarterly Journal of Economics, Vol. 110, pp. 641–680.

This page intentionally left blank

CHAPTER 6

Microeconomics of Growth in MENA: The Role of Households$ Djavad Salehi-Isfahani Abstract This paper discusses whether households in the Middle East and North Africa (MENA) allocate their resources efficiently and in such a way as to promote growth. It focuses on the role of urban households because they form the majority and they are the main source of growth in human capital. I argue that an efficient and pro-growth allocation of household resources may not be feasible because of constraints that households face in their decisions to supply labor, and to accumulate human and physical capital. I identify two aspects of the environment in which MENA households operate as critical to conditioning their behavior: the large role of the state in the economy, which distorts the incentives in the education and labor markets, and social norms regarding gender, which influence the division of labor at home and in the economy. Patriarchal gender norms limit women’s participation outside the home, resulting in higher fertility and lower labor force participation of women in MENA compared to countries with similar income. The strong role of the state affects incentives in three key markets for credit, education, and labor. Powerful central governments have inhibited the development of modern financial markets by preventing the emergence of private banking and an independent judiciary, which is critical for the enforcement of financial contracts. Distorted financial markets affect household savings by keeping interest rates low and often negative, and thereby discourage accumulation of financial assets relative to unproductive assets such as land. Of greater importance is state intervention in

$ This paper was prepared for the Global Research Project, Global Development Network, and was ﬁrst presented in a GDN conference in Prague, June 2000. Corresponding author.

CONTRIBUTIONS TO ECONOMIC ANALYSIS VOLUME 278 ISSN: 0573-8555 DOI:10.1016/S0573-8555(06)78006-3

r 2007 ELSEVIER B.V. ALL RIGHTS RESERVED

160

Djavad Salehi-Isfahani

the markets for education and labor, which determine the amount and type of human capital the MENA households accumulate. I argue that the prevalence of public-sector employment and the regulation of private employment have increased private returns to formal schooling over and above their social return, while at the same time reduced private returns to other types of productive skills below their social return. As a result, households invest in an inefficient portfolio of human capital with dire consequences for long-run growth.

Keywords: economic growth, middle east, household behavior, human capital, labor markets JEL classifications: D13, J22, J24, O12, O15 1. Introduction As the longest standing unit of social organization, the family is also recognized as an important agent of economic growth. Households make important decisions that affect production and accumulation of physical and human capital. Increased recognition of this fact in economics has led to a surge in theoretical and empirical models of household behavior in the last two decades. Manski (2000) considers the increased emphasis on households as one of the major recent innovations in neoclassical economics. A recent microeconomic view of growth in Sub-Saharan Africa by Collier and Gunning (1999) has produced important insights.1 However, research on economic growth in the Middle East and North Africa (MENA) region has largely ignored the role of households, instead focusing on actions taken by governments. Historically, the region has been ruled by strong and dominant central governments, and often with negative effects on economic growth (Issawi, 1995). More recently, in the 20th century, socialist ideology and availability of oil revenues have helped maintain the image, if not the reality, of the all-powerful state. Nevertheless, we must understand the behavior of the micro units – households and ﬁrms – to reach a deeper understanding of aggregate outcomes, including the consequences of actions taken by powerful states. In this paper I examine the role of households in economic growth of MENA. I focus on the incentives and constraints faced by households that help or hinder their role in fostering growth. The constraints I consider are those imposed by the two institutions that are not peculiar to the region, but have greater inﬂuence in MENA than elsewhere: the importance of central

1

For a review of similar applications of microeconomics to economic growth around the world, see Guriev and Salehi-Isfahani (2003).

Microeconomics of Growth in MENA

161

governments in economic life and social norms regarding gender. I show how these institutions deﬁne the incentives and the constraints that shape household behavior pertaining to growth. The strong presence of the state operates primarily in the form of extensive interventions in the markets for education and labor that create large gaps between private and social returns to human capital, leading individuals and households to invest more in formal education and diplomas than productive human capital. Social attitudes toward gender roles affect household behavior primarily by affecting the allocation of women’s time between home production and market work. Women are discouraged from engaging in market work, thereby limiting the supply of labor and human capital to the economy and reducing incentives for investment in the education of girls. Limits on women’s access to education and the labor market in turn promote population growth, reduce the education of the next generation, and thereby retard economic growth. Throughout the paper I rely on the theory of household economics to guide the discussion, and on the rather limited body of empirical studies to relate the theory to the experience of individual countries in the Middle East. The most useful among the latter are those that use survey data. Learning from micro data is the most effective way to learn about household behavior in speciﬁc countries. Unfortunately, such data are only available for a handful of countries and researchers have only recently started to use them in micro econometric studies of households and markets in the Middle East. In the Appendix to this paper I provide a partial list of the micro data sets that exist for the MENA region, some of which are publicly available.

1.1. The role of urban households I limit my discussion to the role of urban households in part to keep the discussion more focused, but more importantly, because urban households have greater inﬂuence on human capital accumulation and modern growth. Rural and urban households perform many similar economic functions – procreation, education, and labor supply – but they differ in important ways. Historically, the arid climate and the feudal system of surplus extraction have created urban centers that are quite distinct from rural communities. Rural and urban areas remain geographically and economically quite distinct even today. Because rural households operate both as ﬁrms and families, a proper treatment of the role of rural households in economic growth would require the widening of our scope considerably and the discussion of the role of agriculture in economic growth. It is best that the role of agriculture is treated separately. Furthermore, social norms interact with family decisions differently in rural areas because in rural settings the distinction between private and public space is more blurred, so rural women face less conﬂict between market-related activity and housework than urban women.

Djavad Salehi-Isfahani

162

Table 1. Country

The importance of urban households in MENA Urban Population as % of Total

Labor Force in Agriculture (%)

1970

2002

1970

1990

Algeria Egypt Iran Iraq Jordan Kuwait Lebanon Libya Morocco Oman Saudi Arabia Syria Tunisia Turkey UAE Yemen, republic

39.5 42.2 41.9 56.2 50.5 77.8 59.4 45.3 34.5 11.4 48.7 43.3 44.5 38.4 57.2 13.3

58.3 42.8 65.4 67.5 79.0 96.2 90.3 88.2 56.7 77.0 87.1 52.1 66.8 87.6 25.3

47.4 51.8 43.8 47.1 27.8 1.8 19.8 28.9 57.6 56.9 64.2 50.2 41.9 70.7 8.8 70.4

26.1 40.3 38.8 16.1 15.3 1.2 7.3 10.9 44.7 44.5 19.2 33.2 28.1 53.1 7.8 61.0

MENA East Asia and Paciﬁc South Asia Sub-Saharan Africa LAC Middle income countries

41.5 18.5 18.7 18.7 57.4 33.2

58.0 38.2 28.0 33.1 76.4 52.6

49.8 76.4 71.1 78.5 41.0 61.1

34.8 68.9 63.4 67.5 25.5 52.8

Source: World Bank, (2004a).

Fortunately, focusing on urban households only is less of a limitation in MENA than in the other regions because, next to Latin America and the Caribbean (LAC), the MENA countries are the most urbanized in the developing world (see Table 1). The degree of urbanization is even higher if we control for income per capita. Figure 1 shows that in the late 1990s all the MENA countries were at or above the (linear) conditional mean rate of urbanization.2 For the greater part of the 20th century, the MENA countries have had a relatively high proportion of their population living in urban areas. In 1970 the share of the urban population in the MENA was 41%, compared to 19% each for South and East Asia; in 2002 the MENA share had increased to 58%, compared to 28% and 38% in South and East Asia, respectively. Except in Egypt and Yemen, in all the MENA countries the urban population exceeded the rural population (Table 1). Given that the average rural household is larger,

2

The regression in Figure 1 (as well as in other similar ﬁgures) includes observations on all low- and middle-income countries. All data are from World Bank Development Indicators, 2004.

Microeconomics of Growth in MENA

Figure 1.

163

Urbanization in the MENA countries

100 Lebanon Share of urban population

Saudi Ar 80

Oman Jordan

60 Syria Morocco

Turkey

Tunisia Iran Algeria

Egypt 40 Yemen

20 1000

3000

5000

7000

9000

11000

GDP per capita (PPP)

urban households must outnumber rural households in most MENA countries by a good margin. The share of the labor force in agriculture also lower in the MENA countries in 1990 (data for later years are sparsely available) compared to the other developing regions except Latin America and the Caribbean (Table 1), which further indicates the signiﬁcance of the economic role of urban households in MENA. Urban areas in MENA have historically been important centers of public administration, trade, and surplus extraction. Until the second half of the 20th century, those living in urban areas were not important as direct producers, but their positions as managers of the agricultural surplus placed them at the helm in capital accumulation and growth. Their decisions as landlords, tax farmers, and government bureaucrats affected how much was produced and collected in taxes and how the proceeds were used in the local economy as well as in international trade. The role of urban dwellers as managers of the agricultural surplus diminished as major land reform programs in the 1960s – notably, in Egypt, Iran, Iraq, and Syria – weakened the traditional land tenure systems and shifted the control of agricultural production to rural households. In subsequent decades, with industrialization and globalization, the signiﬁcance of human capital in production increased, and with it the role of urban households as its main producer. Finally, the dramatic increase in the ﬂow of foreign exchange to the region after the oil price revolution of 1973 further tilted the balance away from the rural to the urban households. Although all

164

Djavad Salehi-Isfahani

MENA countries did not beneﬁt to the same extent from the oil price increase, labor migration to oil-rich countries and the resulting remittances spread the foreign exchange increase widely across the region. Foreign exchange inﬂow enabled more import of food and thereby diminished the role of rural households as the agents of economic growth. Thus, in the modern period, although urban households have lost their traditional roles, as producers and suppliers of human capital they remain at the center of the growth process.

2. Economic and social environment Many aspects of the social and economic environment affect growth. My interest in this paper is in those aspects that constrain and distort household choices. I further narrow my attention to those that are exogenous to household choices (such as geography) because only these can be meaningfully said to limit choices. Some facets of the environment that are commonly believed to affect choice, such as low education, do not ﬁt the criteria because they are themselves the result of household actions. Besides geography, two key features of the MENA environment that I believe ﬁt the requirement of exogeneity are the strong role of the state in the economy and the social norms regarding gender. I argue that state interventions in the markets for education and labor are the greatest single source of distortion for household decisions. They distort incentives for investing in human capital and adversely affect growth. The state funds and directly provides much of the education and acts as intermediary in the relation between workers and employers. State interventions in the labor market reduce ﬂexibility in wage setting and turnover, which not only reduce incentives to work and cause misallocation of workers to jobs (static misallocation), but also result in inefﬁcient accumulation of human capital (dynamic misallocation).3 Social gender norms, too, affect growth by inﬂuencing static and dynamic allocations. Gender discrimination in the household and in the work place affects the allocation of time within the household by limiting women’s access to market work outside the home. Social gender norms that govern the division of labor in the household can be considered exogenous to household decisions, and therefore constrain those decisions, to the extent that they have developed in an earlier period to serve a purpose which no longer exists. In a modern economy, restricting women to private spaces may inhibit economic growth by encouraging high fertility, reducing female labor supply, and lowering returns to female education. In this sense, as a constraint on household decisions, norms act in the same way as public sector domination of employment which distorts returns to human capital.

3

Salehi-Isfahani and Murphy (2004).

Microeconomics of Growth in MENA

165

2.1. Natural endowments and climate The climate in MENA is arid and semi-arid. Scant and unreliable rainfall has forced settlers in most of the region to develop vast systems of irrigation based on rivers and underground aquifers. In 1997, about 31% of the MENA cropland was irrigated, compared to 3% in Sub-Saharan Africa, 21% in Latin America, and 38% in South Asia. The low proportion of irrigated land in Africa is one of the reasons why the African rural households face high risks, which Collier and Gunning (1999) argue has at the micro level caused poor economic growth. Irrigation is an important part of the response of farmers to the risky environment in the region. In some MENA countries where rainfed cultivation still contributes a large share of agricultural production, as in Syria, rural households must cope with a high degree of risk owing to variability of rainfall. In the more densely populated parts of the region, such as the banks of the Nile, irrigation is the only method of cultivation, and helps attenuate weather risks almost entirely. In the last half a century, another geographic factor, the region’s rich reserves of hydrocarbons have exerted an inﬂuence at par with the climate. With twothirds of the world’s oil and one-third of natural gas resources the economies of the region were to varying degrees affected by the rising price of crude oil in the second half of the 20th century. The oil wealth has greatly affected the course of economic growth by raising real wages faster than productivity and changing relative prices in favor of non-traded sectors through the well-known Dutch Disease phenomenon (Gelb, 1988).

2.2. Trade shocks The largest trade shocks to the region result from ﬂuctuations in the price of oil. A group of nine oil exporters in the region is directly affected by oil price ﬂuctuations, and the rest through worker remittances and direct aid (El-Erian et al., 1996). Oil prices quadrupled in 1973 and jumped again by a factor of three in 1979–1980, after which they started a gradual decline until 1986 when they collapsed, wiping out all but 10% of the gains made during the two previous price hikes. Since 1986 oil price ﬂuctuations have continued (high in 1990–1991, low in 1998, and high again in 2003–2005), resembling price ﬂuctuations for other primary commodities, with the obvious difference of the role played by political factors. There is no evidence available regarding how the variability in oil incomes enters household decision making. Governments, who are arguably better informed about oil price shocks than households, have so far failed to smooth over their own expenditures. Kuwait is the only country that systematically excludes a part of its oil revenues from current use by placing it in a fund for future generations. Several other countries, such as Iran, have oil stabilization funds that help smooth consumption over temporary price ﬂuctuations, but not over generations (Davis et al., 2001).

166

Djavad Salehi-Isfahani

2.3. Institutional environment 2.3.1. Role of state Central authorities have historically played a signiﬁcant role in the economic life of the MENA societies. The reasons for the strong role of the state have changed over time and differ from country to country (Anderson, 1987). In the past the state dominated economic life because of its role as the manager of water resources, extractor of surplus, and protector of agricultural communities.4 The role of Islam in the promotion of a patrimonial system may also have contributed to the rise of the state (Bill and Leiden, 1974). In the recent past, socialist ideology following independence (Iraq, Syria, and Tunisia), rise of nationalism (Egypt and Iran), and oil revenues that accrue to the state (Saudi Arabia, Iran, Kuwait, and other Gulf states), have helped the state continue its dominant role to date (Mahdavy, 1970, and Richards and Waterbury, 1996). Although the state has contributed to economic growth by providing infrastructure and other public goods, it has also inhibited micro units – ﬁrms and households – from playing a more positive role in economic growth. For households state intervention in the markets for capital and labor has been the main inhibitor. Speciﬁcally, as we see below, the rise of public sector employment and interventions that reduced the ﬂexibility of the labor market have distorted individual incentives for lending to ﬁrms and investments in human capital. There is a large literature in economics that shows labor market regulations affect employment and growth, but they do not directly relate to the role played by households. Lazear (1990) shows how regulation in the form of severance pay hurts employment; Besley and Burgess (2004) show how manufacturing growth in Indian states with stricter labor regulation has lagged behind those with less regulation; and Botero et al. (2003) show that more regulation reduces labor force participation. The large size of the public sector in MENA can be deduced from the share of public expenditures in GDP. In Figure 2, where this ratio is depicted for individual countries relative to a regression line for low and middle income countries, with the exception of Iran, Syria, Turkey, and Yemen, all MENA countries are above the regression line. The public sector in Iran is perhaps one of the largest in MENA, but the heavy underpricing of foreign exchange and energy products is responsible for the relatively small share of public expenditures reported in the World Bank data. Esfahani and Taheripour (2002) put the share of government expenditures closer to 50% of the GDP, which would make Iran an outlier in the other direction. State expenditures as

4

The terms ‘‘Oriental Despotism’’ and ‘‘Asiatic Mode of Production’’ have been used to describe the economic systems at the helm of which stood the state. For an application to 18th century Iran, see Ashraf (1970) and Abrahamian (1974).

Microeconomics of Growth in MENA

167

Figure 2. Size of government in the MENA countries Oman

Share of public expenditures in GDP

40

Lebanon Egypt Morocco Tunisia Jordan Algeria 30

Yemen

Syria Iran Turkey

20 1000

3000

5000

7000

9000

11000

GDP per capita (PPP)

percentage of GDP range from a low of about 25% in Turkey to 35% for the North African countries, to 50% in Kuwait and Iran, compared to 12% in East Asia and 25% in Latin America (Table 2). Not only are public expenditures high in relation to GDP, the share of wages and salaries of public employees in the GDP is also high relative to the other developing regions. Wages and salaries comprise about one-third of public expenditures in MENA compared to one-fourth in East Asia and Latin America (Table 2). The share of public sector wages and salaries in GDP, which ranges from 6.0% in Turkey to 15.4% in Jordan, is high compared to 2.6% for East Asia. Except for Turkey, the share of government wages and salaries in the GDP of MENA countries lies above the conditional mean represented by the regression line (Figure 3). The heavy burden of public employment is correlated with overall intervention in the labor market. To start with, the state is by far the largest employer. If we consider public employment (civil service and state enterprises), the extent of the impact of public employment on the labor markets becomes evident. In 1990, public employment as percentage of total employment ranged from 21% in Morocco to 34% in Egypt, 44% in Jordan, 57% in Algeria, and 85% among the nationals in Kuwait (Said, 2001; Shaban et al., 2001). Nearly one out of every three Arabs working outside agriculture is a public employee, compared to one out of ﬁve in the Organization for Economic Cooperation and

168

Djavad Salehi-Isfahani

Table 2. Size of government Country

Algeria Egypt Iran Iraq Jordan Kuwait Lebanon Libya Morocco Oman Saudi Arabia Syria Tunisia Turkey UAE Yemen, Republic MENA East Asia and Paciﬁc South Asia Sub-Saharan Africa LAC Middle income

Government Wages and Salaries as % of GDP

Public Expenditures as % of GDP

1977

1997

1977

1997

– 9.52 11.27 – – 6.53 – – 10.44 4.95

– 46.9 45.8 – 44.8 31.4 – –

7.30 9.52 5.98 – –

8.30 6.3 9.16 – 15.43 13.07 8.68 – 11.34 10.23 – – 10.33 6.08 4.15 11.41

40.2 52.6 49.4 33.3 20.9 8.0 –

31.8 34.2 23.2 – 35.0 41.5 37.9 – – – – 23.8 32.6 26.9 11.8 39.2

9.11 – 2.11 7.08 6.09 –

7.96 2.62 – – – –

38.09 – 12.7 23.0 19.1 –

27.89 11.6 17.7 – – –

Source: World Bank, (2004a).

Development (OECD) countries (Said, 2001). Growth in public employment has come from access to oil revenues (Iran, Kuwait, Saudi Arabia, and UAE), job guarantees (Egypt and Morocco), and the government acting as the policy of employer of last resort during recessions (Algeria, Jordan, and Tunisia). As the largest provider of formal employment and regulator of private sector employment, governments in MENA have radically affected the labor market and the economic environment. Public sector employment is characterized by low turnover and emphasis on formal education, especially high school and university degrees (Said, 2005). The share of state in the employment of the educated workers is even higher. In Iran, in 2001, 58% of men in the public sector had upper secondary education or more, compared to 20% in the private sector; for women, 75% in the public sector compared to 35% in the private sector (SalehiIsfahani, 2005). The state is thus in a position to inﬂuence the structure of rewards to education out of proportion to its share in total employment. By emphasizing diplomas at the expense of productive skills, the state helps reinforce a system of education based on memorization and testing rather than acquisition of productive skills.

Microeconomics of Growth in MENA

Figure 3.

Share of government wages and salaries in the MENA countries

15 Yemen Share of wages and salaries in GDP

169

Jordan

Morocco Tunisia

Oman

10 Iran Algeria Egypt

Lebanon Turkey

5 1000

3000

5000

7000

9000

11000

GDP per capita (PPP)

In addition to their role as the largest employer, most MENA states have further reduced the ﬂexibility of the labor market through regulation of pay and rules for job termination. According to the Heritage Foundation index of wage and price ﬂexibility, the MENA countries represented in the sample, with the exception of Jordan, Morocco, and Tunisia, had a labor market rigidity score above the median.5 In Iran, which has a high score of 4 (only four countries in the sample had a higher score), the government sets the pay scales for privatesector jobs based on formal schooling, and the 1990 Labor Law places the burden of proof on the employers who lay off low-productivity workers (SalehiIsfahani, 1999, 2000, 2002). In 2002, the Labor Law was amended to exempt ﬁrms with ﬁve or fewer workers. Egypt also has pay scales based on education and a system of compulsory arbitration between the employer and employees (Assaad, 1997; Said, 2001). According to Labor Law 137 (1981), all prospective employees must register at a district labor ofﬁce where government ofﬁcials evaluate skills and note ratings on an ‘‘employment certiﬁcate.’’ Upon being hired or terminated, workers must notify the labor ofﬁce of their change of status. In the event a ﬁrm is acquired by another, the obligation to continue employment falls to the new owner. Furthermore, the state has successfully used

5

Heritage Foundation, 2002 Index of Economic Freedom. Washington, DC and New York, NY: Heritage Foundation and Wall Street Journal.

170

Djavad Salehi-Isfahani

the hierarchical union organization under its control to inﬂuence the employment relation in private enterprises (Posusney, 1997). In most other MENA countries the government places restrictions on ﬁring of employees (Said, 2001). Egypt and Morocco adopted policies of job guarantees for educated workers in an effort to increase incentives for education (Assaad, 1997; Shaban, et al., 2001). Recognizing the distortions that these policies have on incentives in the markets for education and labor, both Egypt and Morocco have in the 1990s backed away from their employment guarantee obligations. In 1990, Morocco provided exemptions from the stringent employment regulations that allowed businesses to hire, for up to 18 months, skilled workers without restrictions on wages and beneﬁts or any obligation to retain them (Said, 2001). The effect of state regulation of the labor market on human capital accumulation is exacerbated by its direct role in its production of education. The educational system in MENA is under heavy inﬂuence from the state, both in setting the curriculum and in ﬁnancing of education. Except in Lebanon, where private schooling at all levels predominates, in all the MENA countries the state is the main provider of education. In 1992, only 7.2% of primary and 6.2% of secondary students in Arab countries were enrolled in private schools, compared to 11.7% and 41% for the upper middle-income economies (Barnett et al., 1998). In Iran, in the mid-1990s the same ﬁgures were 1 and 2% only (World Bank, 1997a). Psacharopoulos and Nguyen (1997) show that in Arab countries private spending was only 10% of the total spending on education compared to 50% for East Asia and the Paciﬁc. The trend is for more privatization of education in MENA, but the environment for human capital accumulation may not be signiﬁcantly affected by private provision of education as long as the labor markets remain inﬂexible. To the extent that educational institutions take their cues from the labor market, private schools may not behave very differently from public schools (SalehiIsfahani and Murphy, 2004). In Iran, private schools only outperform public schools in test taking to prepare them for entry into universities, not in offering a more varied curriculum. After all, for parents and students success is still deﬁned as passing the university entrance examinations and landing a job in a labor market that rewards diplomas rather than productive skills. 2.3.2. Social norms It is commonplace to speak of the Middle Eastern societies as patriarchal and characterize their gender relations as less equal compared to other developing regions6. The veil has come to symbolize separation, and to many, the subordination of women in social and economic life in the MENA societies.

6

There is a vast literature on this subject. For an excellent discussion of gender norms in the Middle East as it pertains to work and education, see World Bank (2004b).

Microeconomics of Growth in MENA

171

Although the origins of these gender relations can be traced to pre-Islamic Middle East (Nashat, 1999), Islamic laws and edicts may have made them more resilient to change long after their purpose had disappeared, hence the designation of social gender norms as a constraint on household actions in the modern Middle East. Although the importance of social norms in social outcomes is generally acknowledged, little is known about the precise way in which they inﬂuence micro decisions. Mason (1997) and McDonald (2000) divide gender relations into on family decisions into institutional stratiﬁcation at the macro level and gender roles at the micro (household) level. Patriarchal gender relations exist at both levels in the MENA countries, in personal attitudes toward female education and work as well as at the level of social institutions (Kazemi, 2000). The state has often played an important role in promoting and enforcing social norms. In several countries gender norms have been ‘‘codiﬁed in law, especially in the region’s personal status or family laws, such as those requiring women to obtain the permission of fathers or husbands to gain employment, to seek a loan, to start up a business, or to undertake any form of travel’’ (Moghadam, 1998). The ban against women driving in Saudi Arabia, lack of women’s suffrage in Kuwait prior to 2005, and the sexual segregation of men and women in buses, classrooms, and workplaces in Iran are means by which the state enforces social norms. In Iran, where women have made strong gains in education, now outnumbering men in universities, former president Rafsanjani has complained of political pressures on him (while president in the 1990s) to limit women’s access to the university.7 Patriarchal gender roles at the household level, reinforced by the social image of men as breadwinners, have been blamed for low female participation in the labor market and wage rigidity (Karshenas, 2001). The gap in gender relations between MENA and other regions of the world points to social norms as an important feature of the social environment in which micro units operate. The Gender Empowerment Measure (GEM) calculated by the United Nations, which measures the economic and social opportunities open to women relative to men, allows a comparison of gender relations across countries.8 Figure 4 depicts mean GEM by income per capita

7

‘‘They asked why women should study if they are not going to work. And even some radical representatives spoke from the tribune of the Majlis asking why should we give the seats in universities to a woman who when she ﬁnishes her education must go home and take care of children. I said that an educated mother without a job would be effective in the society because of the children that she will educate.’’ (Interview with M. H. Rafsanjani, author’s translation from Persian, Hamshahri 1/10/00, p. 15) 8 According to the Human Development Report, 1999, the Gender Empowerment Measure (GEM) ‘‘captures opportunities for women in selected economic and political areas. It examines whether women and men are able to actively participate in economic and political life and take part in decision making. It tracks the percentages of women in parliament, among administrators and managers and among professional and technical workers – and women’s earned income share as a percentage of men’s.

Djavad Salehi-Isfahani

172

Figure 4.

Gender empowerment measure for the MENA countries

Gender Empowerment Measure

0.6

0.5

0.4 Tunisia Syria Morocco Egypt

0.3

Turkey Iran

Algeria Jordan 0.2 1000

3000

5000

7000

9000

11000

GDP per capita (PPP)

and shows that all MENA countries lie below the regression line. Interestingly, GEM in MENA is even below that of much poorer Sub-Saharan African countries (ERF, 2000). Around the world women live longer than men. The ‘‘Disability Adjusted Life Expectancy’’ (DALE) index calculated by the World Health Organization (WHO) shows the MENA countries among a select group in which men fare better than women (Murray and Lopez, 2000). Indeed, the top ﬁve countries in this category are the MENA countries. As one observer put it, the lack of ‘‘a sex gap in Egypt and Iraq, or the existence of a gap in favor of men in Turkey and Iran, may reﬂect a relative devaluation of women’s wellbeing in those countries as well as large inequalities between the sexes in education, a variable that correlates highly with health.’’9 Gender norms are constraints on individual decisions, but they also change when enough individuals choose to defy them. Norms with respect to childbearing have gone through drastic change in many MENA countries, where fertility has declined precipitously in recent decades, notably in Iran where the slow pace of increase in women’s work outside the home, despite improvements in health, reproduction, and education, might lead one to suppose very rigid gender norms sparingly as explanation of behavior (Salehi-Isfahani,

9

Washington Post, June 12, 2000; p. A09.

Microeconomics of Growth in MENA

173

2005). Since our understanding of how speciﬁc social norms arise and disappear is far from adequate, we must use social norms sparingly as explanation of behavior. Ironically, the oil boom that increased incomes and hastened modernization in the region may have strengthened the traditional gender contract because the inﬂow of oil revenues removed the need for women to work outside the home (Moghadam, 1998). This observation ﬁts the pattern of gender norms across the region relatively well: norms have proved more resilient to change in oil exporting countries of Iran, Kuwait, and Saudi Arabia, where women have low labor force participation rates despite relatively high education, but shown less rigidity in Morocco, Tunisia, and Turkey where women have the highest activity rates with no apparent advantage in education or skills compared to women in oil-rich countries. 3. Household decisions Economic theory identiﬁes three household decisions with direct impact on economic growth: to save, to accumulate human capital, and to procreate. These decisions are interdependent. There is the well-known tradeoff between the quantity and quality of children, which derives from the scarcity of resources, mainly time, in the household (Becker, 1991). Children can be a substitute for other assets as means for old-age support. This interdependence implies that, in principle, all other household actions also indirectly affect growth. The allocation of household time between home production, leisure, and market work is the most important of such decisions. For example, rewards for market work for women inﬂuence household choices with regard to fertility and investment in human capital and thus indirectly affect growth. In this section I consider decisions with both direct and indirect impact on growth and two features of the environment that constrain these decisions – the interventionist state and social norms. 3.1. Household time allocation The most important asset of the household is the total time available to its members. The household divides its total endowment of time between leisure, work for home production, and market work. For urban households home production primarily consists of ‘child services’ (Hotz et al., 1997), which is itself a function of the number and quality of children. Modern economic growth is associated with a shift from home production of commodities to market work, especially for women, and from quantity to quality of children. The allocation of time in MENA households differs in signiﬁcant ways from households in other developing countries, and this is most noticeable in the gender allocation of tasks between married couples. Moghadam (1998) speaks of a speciﬁc ‘‘gender contract’’ which designates men as breadwinners and women as homemakers. The gender contract implies a low female labor force

174

Djavad Salehi-Isfahani

participation, at odds with which is the levels of fertility and education in the MENA societies. Whether low participation can be blamed on social norms is an empirical question for which we do not yet have a satisfactory answer. MENA gender norms inherited from the past can be said to constrain household choices – and thereby hinder growth – at present, if we can determine empirically that they prevent a more productive allocation of time between household members. I now examine how social norms affect time allocation in fertility and labor market participation decisions. 3.1.1. Fertility Despite substantial decline in fertility in several MENA countries, notably Egypt, Iran, Lebanon, Turkey, and Tunisia, the region as a whole has been slow in its demographic transition (Rashad and Khadr, 2002). In 2002, fertility rates in MENA, averaging 3.2 births per woman, were exceeded only by fertility in countries of Sub-Saharan Africa (World Bank, 2004a). Delay in the transition and persistence of high fertility in the Arabian Peninsula where, despite high incomes and high education, fertility remains around 6 births per woman, has given rise to the notion of Islamic or Arab fertility (Caldwell, 1986; Obermeyer, 1992). As late as 1977, MENA births averaged 6.3 per woman (Table 3), nearly as high as in Sub-Saharan Africa (6.6) with much lower per capita income, and twice that of East Asia and the Paciﬁc (3.3). Figure 5 shows that in 1977 the total fertility rate (TFR) in all but two MENA countries (Egypt and Turkey) was higher than indicated by the regression line which depicts mean TFR by income for all developing countries. Recently, fertility has declined in most countries of the region, lowering the TFR from 6.3 to 3.6 in the last 20 years (Table 3). In 1997 at least half of MENA countries were below the regression line (Figure 6). Two implications of lower fertility for MENA growth are important to note. First, the move from high to low fertility provides these countries with a onetime bonus derived from a favorable age structure. Economic historians have labeled the beneﬁts to growth from a more rapid labor force growth and a low dependency ratio (ratio of working to non-working population) that follow fertility transitions as a ‘‘demographic gift’’ (Bloom and Williamson, 1997) and a ‘‘window of opportunity’’ (Barlow, 1994). Salehi-Isfahani (2002, 2005), Tunali (1996), and Yousef (1998) show how the changing age structure can play a positive role in the MENA countries. Even though fertility has been on the decline in the last 20 years, the rate of growth of the labor force has remained high, averaging 3.2% during 1985–1995 and 2.9% during 1995–2002 (Table 3) and will likely remain above 2% for the next decade or so. Labor force growth rates could be even higher if female labor force participation rates increase. As noted earlier, although the effect of lower fertility and more female education on labor force participation of women in the formal economy is yet to fully materialize, it is only a matter of time before women begin to seek work outside the home.

Table 3. Population, labor force, and fertility Country

Population Growth (annual %)

Growth of Labor Force (annual %)

Fertility Rate, Total (Births per Woman)

1987–1996

1997–2002

1975–1984

1985–1994

1995–2002

1977

1997

2002

Algeria Egypt, Arab Republic Iran, Islamic Republic Iraq Jordan Kuwait Lebanon Libya Morocco Oman Saudi Arabia Syrian Arab Republic Tunisia Turkey UAE Yemen, Republic

3.1 2.5 3.5 3.3 3.8 5.1 1.3 4.3 2.2 4.9 5.4 3.3 2.7 2.3 8.7 3.6

2.4 2.2 2.2 3.0 4.6 0.6 2.0 2.1 1.9 4.1 3.6 3.2 1.9 2.0 5.2 4.1

1.6 1.9 1.2 2.1 3.0 3.4 1.4 2.0 1.7 2.6 2.7 2.5 1.2 1.7 4.9 2.9

3.2 2.2 3.0 2.8 2.1 7.5 -0.1 4.3 3.2 5.7 5.1 3.1 3.7 1.4 21.8 1.7

3.9 2.6 2.4 3.2 6.0 1.0 3.2 1.9 2.5 3.7 4.3 3.7 2.9 2.8 5.0 5.0

3.4 3.0 2.7 3.0 4.1 5.7 2.7 2.1 2.5 2.3 2.9 4.1 2.5 2.5 4.4 3.0

7.2 5.3 6.5 6.6 7.2 5.9 4.3 7.4 5.9 10.1 7.3 7.4 5.7 4.5 5.7 8

3.5 3.6 2.8 4.7 3.9 2.9 2.5 3.8 3.1 4.8 5.7 4 2.4 2.6 3.5 6.4

2.8 3.1 2 4.1 3.5 2.5 2.2 3.3 2.8 4 5.3 3.4 2.1 2.2 3 6

MENA East Asia and Paciﬁc South Asia Sub-Saharan Africa LAC Middle income

3.2 1.6 2.3 3.0 2.2 1.6

2.7 1.5 2.0 2.7 1.8 1.4

1.9 1.0 1.8 2.4 1.5 1.0

2.8 2.3 2.3 2.6 3.2 2.2

3.0 2.0 2.2 2.6 2.7 1.8

2.9 1.2 2.4 2.6 2.2 1.3

6.3 3.3 5.5 6.6 4.5 3.4

3.7 2.2 3.5 5.5 2.7 2.2

3.1 2.1 3.2 5.1 2.5 2.1

Microeconomics of Growth in MENA

1997–1986

Source: World Bank, (2004a). 175

Djavad Salehi-Isfahani

176

Figure 5.

Total fertility rate for the MENA countries, 1977 Oman

10

Total fertility rate

8

Syria Jordan Algeria

Saudi Ar Iran

Morocco 6

Tunisia Egypt Turkey

4

2 1000

3000

5000

7000

gdp77

Figure 6.

Total fertility rate for the MENA countries, 1997

Yemen 6

Saudi Ar

Total fertility rate

Oman Jordan Syria

4

Algeria Egypt Morocco Tunisia Iran Lebanon Turkey

2

0 1000

3000

5000

7000

GDP per capita (PPP)

9000

11000

Microeconomics of Growth in MENA

177

3.1.2. Participation of women in market work Reliable data on labor force participation of women in the region is lacking. A few countries report data based on household surveys (Egypt, Kuwait, Iran, Morocco, and Turkey). As in other developing countries, accounting for women’s work, especially in self-employment and in agriculture, is notoriously difﬁcult. In addition, many women prefer to declare their occupation as housework (or are identiﬁed as such by their husbands if they are the respondent) rather than declare themselves unemployed, even though they would readily accept an appropriate job. Variation in the deﬁnition of what is meant by ‘‘work’’ reduces comparability of estimates, but this is a less severe problem in urban areas where own account workers form a smaller proportion of the workforce than in rural areas. Comparison of urban women across the world reveals a much lower rate of participation for MENA. In 1990s female participation rates in urban Turkey was only 15% (Tunali and Baslevent, 2000), in Egypt about 16% (Assaad and El-Hamidi, 2002), in Jordan 12.1% (Flynn, 1999), and in Iran 8.6% (Statistical Center of Iran, 1998). These rates are all less than half the rates reported for East Asia. Even in the predominantly Muslim Malaysia female labor force participation has reached 40%. Table 4 presents a more complete comparison of the MENA female participation rates based on the share of women in the labor force as reported in World Bank (2004). The participation rate for the MENA women is the lowest among any region and is not changing as fast as one would expect given the rise in education and decrease in fertility. The sharpest contrast is with Sub-Saharan Africa where, despite higher fertility and lower education, women comprised 42.2% of the workforce in 2002, compared to 28.6% for MENA. The fact that Africa is more rural explains a part but not all the difference. The share of MENA women in the labor force appears low even taking into account their higher fertility. Although labor force participation, female education, and fertility are obviously interrelated, it is still informative to compare labor force participation rates conditional on education and fertility. Figures 7a and b present the regression for the share of women in labor force and depict the relative positions of MENA and Sub-Saharan African countries. In this regression, the shares of women in African countries appear well ‘explained’ by their high fertility and low enrollment rate in secondary school, whereas in the MENA countries they appear as outliers. Finally, changes in participation over time in MENA suggest a slow response to falling fertility. During 1977–2002, a period when fertility dropped by more than half (Table 3), the share of women in the labor force increased very modestly, from 23.4% to 28.6% (Table 4). Given the declining trend in fertility and rising female education in the MENA region, future increase in female labor participation has the potential to contribute to the MENA growth. If the proportion of women who work outside the home were to gradually increase to the level in East Asia,

Djavad Salehi-Isfahani

178

Table 4. Women in MENA labor force Country

Labor Force, Female (% of Total Labor Force) 1977

1997

2002

Algeria Egypt, Arab Republic Iran, Islamic Republic Iraq Jordan Kuwait Lebanon Libya Morocco Oman Saudi Arabia Syrian Arab Republic Tunisia Turkey UAE Yemen, Republic

21.1 26.3 20.0 17.0 14.3 11.6 21.4 17.9 32.9 6.2 6.8 23.3 27.3 36.2 4.8 30.9

25.7 29.4 25.4 18.7 22.6 31.2 28.8 21.7 34.6 15.1 14.2 26.2 30.9 36.7 13.8 27.9

29.0 31.0 28.4 20.4 25.6 32.1 30.1 24.0 34.9 18.9 17.7 27.6 32.1 38.1 15.9 28.3

MENA East Asia and Paciﬁc South Asia Sub-Saharan Africa LAC Middle income

23.5 42.1 33.8 42.0 26.4 40.2

26.6 44.4 33.0 41.9 34.1 42.0

28.6 44.5 33.6 42.0 35.2 42.2

Source: World Bank, (2004a).

three times as many women would be participating in the labor market. This increase in the rate of growth of the labor force has the potential to generate economic growth. These observations indicate that, given their levels of education and fertility, the low rate of market work of the MENA women is an anomaly, but they do not explain it. Social norms are frequently advanced as an explanation (see, for example, El-Sanabary, 1993), but convincing evidence that they do in fact act as a constraint is still lacking. An alternative hypothesis for the low participation of women in MENA is the importance of oil income (World Bank, 2004b). Oil income permits the average MENA family to enjoy a higher standard of living given their education and productivity, thereby reducing the need for women to work. Non-wage family income is well known as a negative inﬂuence on the labor force participation of women. The question is whether oil income at the national level acts on household behavior the same way as unearned income. The answer depends on how oil income is distributed. To the extent that households receive transfers from the government, their real income rises without changing the price of leisure. Direct transfers to households, as in

Microeconomics of Growth in MENA

179

Figure 7(a). Share of women in labor force, conditional on GDP per capita, total fertility, and female school enrollment in Sub-Saharan Africa; (b) Share of women in labor force, conditional on GDP per capita, total fertility, and female school enrollment in MENA

FLF|(TFR, fem_sec_enroll)

20

FLF|(TFR, fem_sec_enroll)

Predicted FLF

10

0

-10

-20 -5000

0

(a)

5000 10000 GDP|(TFR, fem_sec_enroll)

15000

20 FLF|(TFR, fem_sec_enroll)

FLF|(TFR, fem_sec_enroll)

10

0

-10

Tunisia Morocco Egypt Turkey Iran Algeria Syria Yemen Jordan

Kuwait

Bahrain

Iraq

-20

S. Arabia

-5000 (b)

Predicted FLF

0

UAE

5000 10000 GDP|(TFR, fem_sec_enroll)

15000

Kuwait, and subsidies for consumer goods, as in all oil-exporting countries, fall into this category. But transfers that occur through the labor market, e.g. higher pay for public sector jobs, do not because they raise the price of leisure which would increase labor supply. Anecdotal evidence suggests that the income effect on women’s labor supply may be negative. For example, we read that, coping with hard times, ‘‘reluctantly, men allow their daughters to work until marriage or allow their wives to go on working until the mythical day when they can

180

Djavad Salehi-Isfahani

afford to forgo a second salary,’’ (Al-Sayyid-Marsot, 1989, p. 121). But systematic evidence for this view is lacking. Assaad and El-Hamidi (2002) ﬁnd evidence from micro data from Egypt which indicates the opposite – that female wage workers are more likely to be found in households with a male wage worker and one with higher earnings. The search for cultural reasons for differences in the labor force participation of women with different ethnic backgrounds in the US has produced mixed results. Ortiz and Cooney (1984) ﬁnd that the inﬂuence of Hispanic culture on the rate of participation of women works mainly through education, fertility, and language. Reimers (1985) conﬁrms this ﬁnding for Hispanic women, but concludes that the difference between US-born whites on the one hand and Asian and blacks on the other is due to ‘‘direct cultural effects on the parameters of the labor force participation function.’’ In contrast, decline in the female labor force participation during industrialization has been explained by social norms that prevent women from accepting blue-collar jobs (Mammen and Paxson, 2000). Goldin (1995) argues that in the US blue-collar jobs are shunned because they stigmatize the husbands as unable to care for their wives, whereas women holding white-collar jobs do not face such a situation. Gender norms may inﬂuence labor supply of married women through the relative power of men and women in the household. The literature on the division of labor within the household suggests that the relative bargaining position of men and women inﬂuences the labor force participation of women, but the direction of inﬂuence is not obvious. More power to women may raise or lower their likelihood of market work. Grossbard-Schechtman (1993) presents evidence of circumstances in which women increased their participation in response to a decrease in their share of the gains from marriage, this itself being caused by a decline in the ratio of males to females. It is equally plausible to imagine, as indeed the prevailing wisdom on obstacles to women’s participation goes, that increased female power within marriage would lead to higher participation. This version appears to be more true in MENA where female labor force participation declines with marriage (World Bank, 2004b). In Egypt, Assaad and El-Hamidi (2002) show that women’s market work drops sharply with marriage rather than with childbearing. In Iran, married women have the lowest participation rates, followed by single and widowed women. In Kuwait, Shah and Al-Qudsi (1990) ﬁnd that the labor force participation of single women aged 25–39 years is more than twice that of married women (60% compared to 30). Where the rise in education has increased labor force participation rates, social norms may have forced occupational segregation. In MENA, where they have evolved to allow women to work outside the home, gender norms continue to restrict choice by deﬁning what jobs are appropriate for women. However, although job segregation appears to discourage women from participation, it does not appear to be out of line in comparison to other regions (World Bank, 2004b, p. 93). As in most other regions of the world, educated women in MENA predominate in teaching, nursing, and clerical work.

Microeconomics of Growth in MENA

181

3.2. Human capital Historically, human capital has played a signiﬁcant role in the economic growth of the MENA region. Harnessing water resources for agriculture and developing the calendar to assist cultivation presupposed the development of a scientiﬁc community in urban centers. When the environment has been conducive to human capital accumulation, the region has prospered. Although institutions of formal education existed for periods of time in the major cities of the region, the urban nuclear family seems to have played a signiﬁcant role in the intergenerational transfer of human capital. To this day, education in the MENA region enjoys a high social status, which is reﬂected in a high level of public commitment. Public expenditures on education are higher than in most developing countries (World Bank, 1997b) and average years of schooling and enrollments are commensurate with the region’s income.10 Yet the region’s experience of economic growth does not seem to have beneﬁted from its growth of education (Pritchett, 1999). Thus, the central question which is posed by the MENA experience is why the productivity of human capital has been so low (Pissarides and veganzones-varoudakis, this volume). In this section I focus on the key dynamic decisions of MENA households that affect growth through the accumulation of human capital. A general characterization of the behavior of the average urban household in MENA is that it invests little in physical capital, a respectable amount in human capital, and has a relatively high fertility. The key dynamic decisions inﬂuenced by the MENA external environment are those that relate to fertility and the accumulation of human capital. My emphasis is on how the external environment, as deﬁned by the preeminent role of the state in the labor market, has inﬂuenced human capital accumulation. I argue that the relatively high level of investments in education in MENA results in a low accumulation of productive human capital because excessive job protection distorts incentives in favor of testable knowledge and away from less observable types of human capital, such as creativity. The importance attached to diplomas as distinct from learning productive skills is the results of labor market institutions which reward formal schooling, but not productive skills that only employers are able to observe. In short, private returns to measurable skills exceed their social returns, while the opposite is true for employer-determined abilities. 3.2.1. Health and education Although governments share with households the responsibility for provision of health and education of children, they cannot fully substitute for the role of households. Decisions taken by households to provide for the health and

10

For a review of education in Arab countries also see, Arab Human Development Report, 2002 (UNDP, 2002, Chapter 4).

Djavad Salehi-Isfahani

182

Table 5. Country

Health indicators

Life Expectancy at Birth, Female (Years)

Life Expectancy at Birth, Male (Years)

Mortality Rate, Infant (per 1,000 Live Births)

1977

2002

1977

2002

1977

2002

Algeria Egypt, Arab Republic Iran, Islamic Republic Iraq Jordan Kuwait Lebanon Libya Morocco Oman Saudi Arabia Syrian Arab Republic Tunisia Turkey UAE Yemen, Republic

58.5 55.3 57.0 62.3 – 71.7 67.0 59.3 57.5 56.1 59.9 61.9 61.5 62.5 68.9 48.5

72.1 70.5 70.3 63.9 73.6 79.0 72.6 74.9 70.4 75.6 74.9 72.7 74.6 72.5 76.8 58.1

56.5 52.9 56.2 60.5 – 67.5 63.1 56.0 54.1 53.8 57.6 58.3 59.6 58.0 64.7 45.0

69.4 67.3 68.3 61.4 70.4 74.9 69.0 69.9 66.4 72.6 71.4 68.0 70.8 67.5 74.0 56.8

112.0 131.0 100.0 84.0 42.0 34.0 48.0 107.0 110.0 51.0 75.0 67.0 84.0 120.0 70.0 158.0

39.0 33.0 34.0 102.0 27.0 9.0 28.0 16.0 39.0 11.0 23.0 23.0 21.0 35.0 8.0 83.0

MENA East Asia and Paciﬁc South Asia Sub-Saharan Africa LAC Middle income

57.5 63.9 51.9 48.4 66.0 66.0

70.1 71.3 63.8 46.6 74.0 72.3

55.4 61.8 52.6 45.1 60.9 62.7

67.2 67.7 62.2 45.1 67.6 67.5

107.0 58.7 130.5 119.7 67.5 63.5

43.7 32.4 67.9 103.1 28.3 30.2

Source: World Bank, (2004a).

education of children during the early years are of critical importance for their human capital as adults (Young, 1995). Furthermore, while governments can be effective in the provision of health and education services, parents must be willing to take advantage of those services. Health outcomes in MENA appear to be commensurate with the economic standing of the region (Table 5). Infant mortality declined by more than 50% between 1977 and 2002, in line with fertility decline. Life expectancy and mortality reﬂect better health outcomes in MENA than in South Asia and Africa, but lower than in East Asia and Latin America. Parental investment in children is affected by the incentive structure they face. The most critical signals they receive are the rates of return to investment in various types of human capital which the labor market and the educational system generate. We can assume that decisions to invest in children are mostly driven by considerations for child welfare and old-age security for the parents. Depending on the level of old-age security that is socially provided, parents may depend more or less on their children as means for old-age support. Parents may also wish to increase the welfare of their children by investments that are embodied in

Microeconomics of Growth in MENA

Table 6. Country

183

School enrollment rates

Primary

Secondary

Tertiary

1980

1996

1980

1996

1980

1996

Algeria Egypt Iran Iraq Jordan Kuwait Lebanon Libya Morocco Oman Saudi Arabia Syria Tunisia Turkey UAE Yemen, republic

80.9 – – 98.6 – 84.5 – – 61.6 42.6 48.6 89.5 82.2 – 73.6 –

94.1 93.0 89.8 76 – – 76.1 – 73.8 68.7 61.4 91.2 97.6 – – 52.0

30.5 – – 46.8 – – – 62.3 20.3 9.9 21.3 39.3 22.9 – – –

56.2 67.8 68.8 – – – – – – – 42.4 38.1 – – – –

5.9 16.1 – 8.7 26.6 11.3 30.1 7.8 5.9 – 7.1 16.9 4.8 5.4 3.1 4.1

13.4 22.6 17.1 10.9 – 26.7 27.1 20.0 11.3 6.4 16.3 15.1 13.7 18.2 11.9 4.2

MENA East Asia and Paciﬁc South Asia Sub-Saharan Africa LAC

78.2 93.5 – – 85.5

86.8 101.3 – – 91.2

– – – – 28.7

61.2 – – –

10.8 3.3 4.5 1.4 14.1

15.9 8.1 6.3 3.4 18.6

Source: World Bank, (2004a).

them – human capital – rather than by bequest. The tradeoff between these two depends on the performance of other asset markets and inheritance laws. Lack of conﬁdence in asset markets and inheritance laws that conﬂict with the wish of parents – for example, Islamic laws that allow daughters to inherit only one-half as much as boys – increase the willingness of parents to invest in human capital. The pattern of increase in education in MENA offers a clue to the productivity puzzle noted earlier. While literacy rates are not high in comparison with other developing countries, enrollments in high school and universities are very impressive. In 1997 female and male literacy rates of 46% and 71%, respectively, in Arab countries were marginally better than 49% and 65% in Africa, and much worse than 86% and 88% in Latin America and 75% and 91% in East Asia (United Nations, 1999). In contrast, in 1996 MENA university enrollment rates were on average twice as high as East Asia and Paciﬁc (Table 6). Emphasis on university education is also evident in the share of public education budget that goes to higher education, ranging from a low of 50% in Algeria to as high as 99% in Lebanon (United Nations, 1999). Rewards to education above the basic level – upper and post secondary – appear to be high, in keeping with the reward structure in the public sector-dominated labor market.

Djavad Salehi-Isfahani

184

Table 7. Country

Gender and education

Female–Male Differential in Literacy

% of Female Students in Secondary Education

% of Female Students in Higher Education

1997

1996

1996

Algeria Egypt Iran Jordan Kuwait Lebanon Libya Morocco Oman Saudi Arabia Syria Tunisia Turkey UAE Yemen, republic

25.0 24.3 15.0 10.4 5.6 12.9 25.8 26.6 21.9 18.6 30.0 22.2 18.5 2.8 43.3

47.9 45.3 46.3 47.2 49.6 – – – 48.6 45.8 46.3 – – 49.9 –

– – 45.0 46.9 61.6 49.2 – – 46.0 46.5 – 44.6 – – 12.5

MENA East Asia and Paciﬁc South Asia Sub-Saharan Africa LAC

22.9 13.9 26.9 16.1 2.2

41.3 – – – –

46.3 33.2 36.3 – –

Source: World Bank, (2004a).

The same anomaly appears in the gender gap in education: the gap in literacy is higher than in all other regions except South Asia, but much narrower in secondary and higher education (Table 7). The proportion of women in secondary education in the MENA countries is between 45% and 50%, comparable to 48% in South Korea. In higher education the proportion is surprisingly high, exceeding 40% for many countries, surpassed only by Latin American countries. In Kuwait, Iran, and Saudi Arabia women equal or outnumber men in universities. In MENA, through intensive testing, university education is rationed according to student aptitude, willingness to work hard and to memorize, and to a lesser extent by parental willingness to pay, compared to Europe where they are ‘‘rationed not by price or aptitude, but by achievement in the core subjects studied in secondary schools’’ (Bishop, 1996, pp. 120–121). MENA states enforce uniform tests at all levels of education, but their control over the university entrance examinations is probably their most effective tool in determining incentives in all levels of education. In Iran, private schools work harder than public schools to teach for the ‘‘big test’’ and parents spend large

Microeconomics of Growth in MENA

185

sums on private tutoring (Salehi-Isfahani, 2002). In Egypt, parents invest heavily in private tutoring to help prepare their children for university entrance examinations (World Bank, 1999). In 1994, 64% of urban and 51% of rural primary school-age children had received supplementary tutoring (Bray and Kwok, 2003, p. 2). In 1997, household expenditures on supplementary tutoring in preparatory, primary, and secondary levels accounted for 1.6% of the GDP (World Bank, 2002, Figure 15), which is more than one-quarter of all public expenditures in education. Private tutoring is also an important industry in Turkey, costing parents about 1.4% of the GDP (Tansel and Bircan, 2004). As noted earlier, despite high levels of enrollments in high school and university education, the impact of education on growth in MENA is not noticeable. Pritchett (1999) argues that the MENA region has enjoyed ‘‘the fastest expansion of schooling of any region, including East Asia,’’ but has experienced one of the slowest growth rates in the world. Pritchett points to the low quality of education in MENA as a possible explanation for the low impact of education on growth. But the problem may not be so much with the quality of education than lack of a link between what students learn and what productive jobs require. Schools place too much emphasis on repetition and memorization, which may help with entry into the university, but not with productivity on the job (World Bank, 1999). Finishing high schools and entering the university has a huge premium in most MENA countries because it increases the probability of getting a job and of higher pay (see references below to returns to education). Highly competitive exams for entering university induce parents and students to invest heavily in skills that increase the chance of passing the university entrance examinations, as evidenced by the huge expenditures on private tutoring just noted. The test-taking approach to education not only shortchanges students on the skills they need for productive employment, but also creates huge losses in efﬁciency as the bar is continuously raised to admit a limited number of students into the highly desirable places in universities. One outcome of this competitive system is that while top students from MENA perform very well in international competitions, students who take part in other international tests (such as in the Third International Mathematics and Science Study) that measure learning by the average student do rather poorly (World Bank, 1997a). 3.2.2. The role of the labor market The returns to investment in human capital are in large part realized in the labor market. Wide-ranging state interventions have reduced the ﬂexibility of MENA labor markets (Pissarides and Veganzones-Varoudakis, this volume). Evidence for labor market rigidity was presented earlier in Section 2. Two aspects of this inﬂexibility are particularly important from the point of view of human capital accumulation. First, is the importance of formal education and diplomas for selection into public and private sector employment, and, second, the inability of employers to reward or lay off workers according to productivity (SalehiIsfahani and Murphy, 2004). Employers, public or private, place too much

186

Djavad Salehi-Isfahani

emphasis on formal degrees and little on individual characteristics that they deem as productive. As noted earlier, some countries even offer job guarantees for high school and university graduates, as in Egypt and Morocco. Emphasis on formal schooling and diplomas on the part of employers, private or public, is a logical response to an inﬂexible labor market (Salehi-Isfahani and Murphy, 2004). Individual characteristics that are not easily measured by testing, but can be observed after a period of employment are not very useful to employers who cannot lay off workers once they have hired them. As a result, they are likely to place greater value on ex ante signals of human capital, such as years of schooling, test scores, and class and school rank. Naturally, in the MENA countries workers invest in the type of human capital that can land them a good job rather than those that help them hold on to a job once it has been acquired. While theory helps us understand the potential distortionary effects of labor market interventions on incentives, the actual effects can only be assessed empirically. Studies of private returns to education are the most widely used for this purpose. Unfortunately, only a few studies of returns to education exist for MENA (see Tansel, 1994; Assaad, 1997; Wahba, 2001). In general, they show the convex (rising) rate of return that one would expect from an education system in a distorted labor market. There is also some evidence of the effect of public sector dominance of employment. Assaad (1997) studies the returns to public and private jobs in Egypt. He ﬁnds that public sector hiring practices have a substantial impact on the labor market because, despite erosion in pay, they still offer an advantage over private sector jobs. The overall premium caused by compensation and job tenure is high enough to cause large unemployment queues for public sector jobs. 3.3. Savings Like households elsewhere, the MENA households save for a variety of reasons, to improve a business, for old age or a dowry, or just as insurance against loss of income. The behavior of urban households that operate small enterprises, and therefore engage in both saving and investing, are more appropriately addressed in a paper on ﬁrm behavior. The majority of urban households that do not own a business and are the focus of this paper resort to ﬁnancial or other assets as a means to save. Existing data do not permit disaggregation of national savings by origin (households, ﬁrms, and the government). Income and expenditure data reported in household expenditure surveys are not reliable for estimating personal savings.11 The high-saver countries in MENA are exclusively large oil exporters (Table 8). Non-oil exporters’ propensity to save is less than half of that of the

11

Expenditure and integrated household surveys in Iran typically report expenditures in excess of income, sometimes by 30%. The blame usually goes to the unreliability of the income data.

Microeconomics of Growth in MENA

Table 8. Country

187

Savings and interest rates

Average Real Interest Rates (%)

Gross Saving/GDP (%)

1968–1977

1978–1987

1988–1997

1977

1997

Algeria Egypt Iran Jordan Kuwait Lebanon Libya Morocco Oman Saudi Arabia Syria Tunisia Turkey UAE Yemen, republic

– 4.6 – – – – 3.1 – – – – – – – –

– 4.5 – – 1.3 6.3 1.8 1.0 1.4 – – 4.1 2.4 – –

– 5.9 – 2.9 1.6 15.9 1.5 0.5 2.6 – – 2.5 – – –

35.7 18.5 37.0 9.7 51.9 – 50.0 11.6 44.9 54.9 13.0 22.1 13.3 69.9 –

34.5 13.0 34.1 5.5 25.2 16.7 – 15.2 26.7 34.6 19.0 24.2 19.3 27.4 12.8

MENA East Asia and Paciﬁc South Asia Sub-Saharan Africa Middle income

– – – – –

– – – – –

– – – – –

36.6 28.0 18.1 23.4 25.7

25.5 37.7 18.2 16.7 25.7

Source: World Bank, (2004a).

East Asian countries. Private sector investment in MENA is below other countries, about 10% in the 1990s compared to 18% for developing countries and 22% for Asian countries (IMF, 1996). No data are available on the part of this investment that is ﬁnanced by household savings as opposed to retained earnings of ﬁrms. In developed countries household savings is a signiﬁcant source of ﬁnance for ﬁrms, but not so in developing countries, where weak ﬁnancial intermediation cannot channel household savings toward investors. Policies to liberalize the ﬁnancial markets and deepen ﬁnancial intermediation intend to raise the ability of small savers, such as households, to contribute to growth. Several MENA countries, mainly in North Africa, have attempted ﬁnancial reform as part of their structural adjustment, but studies that can show if household savings have increased as a result do not exist. Economic theory does not give a clear indication of the impact of interest rates on personal savings (Deaton, 1997). Neither of the main theories of why households save – life cycle, permanent income, or consumption smoothing – predicts unambiguously that savings should increase with interest rates. What we do know is that a well-developed system of ﬁnancial intermediation is associated with a more effective use of household savings. It is generally agreed that with low and negative real rates of interest it is difﬁcult to raise long-term ﬁnance from personal sources. Table 8 shows that for those countries that report

188

Djavad Salehi-Isfahani

the real interest rate in World Bank (2004), it has increased over time and is positive. In Iran, until recently, real interest rates averaged negative, and as recently as 2000 were about 5% points below zero (Jalali-Naini, 1997). Easterly (1999) notes a similar rise in the rate of interest for developing countries as a whole, showing that real rates, which were negative before 1980, were on average positive afterward. The ﬁnancial environment in which MENA households generally operate not only suffers from lack of depth, they are also insecure due to the heavy inﬂuence of the public sector and a weakly developed legal system overseeing ﬁnancial contracts. As a result, households may prefer to place their savings in unproductive assets such as gold and land rather than in the ﬁnancial system. The risks of nationalization and loss of value owing to inﬂation, even in countries undergoing reform, deter savers from depending on the ﬁnancial system for long-term savings. 4. Conclusions The question posed in this paper is whether households in MENA allocate their resources efﬁciently and in a way as to promote growth. To answer this question, I focused on the role of urban households, because they form the majority of households and are the main source of growth in human capital and therefore of modern economic growth. I argued that an efﬁcient allocation of household resources which would maximize growth may not be feasible because of constraints that households face in their decisions to supply labor, and human and physical capital. I identiﬁed two aspects of the environment in which MENA households operate as key to conditioning their behavior: the large role played by the state in the education and labor markets and the social norms regarding gender. I argued that this environment has affected an important static decision of the households, namely the division of labor within the household, resulting in high fertility and low labor force participation of women. Given the declining trend in fertility and rising female education, increase in female labor force participation has the potential to raise MENA growth rates. If the proportion of women who work outside home were to gradually increase to the level in East Asia, three times as many women would be participating in the labor market, resulting in a lasting positive impact on economic growth. I then identiﬁed the implications of the same environment for the dynamic aspects of household decisions – human capital accumulation and savings. Although little evidence exists regarding the magnitude of household savings in MENA, we know that their effectiveness for growth has been compromised by lack of conﬁdence on the part of the households in the ﬁnancial markets, which is itself the result of arbitrary actions by governments and the absence of the rule of law. I argued that the impact of state interventions and social norms on household actions were much greater in the accumulation of human capital than in the physical capital. The role of the public sector in MENA in education and the

Microeconomics of Growth in MENA

189

labor market has created a system of incentives in which households strive hard to accumulate formal education, but not enough in productive human capital. There are several implications of the analysis of this thematic paper for future research. First, studies focused on individual countries should examine the extent to which characteristics of the environment identiﬁed here – the large role of the state and social norms – apply in speciﬁc cases. As far as the role of the state is concerned, a description of employment policies and labor market regulations is a good place to start. For social norms, indicators of restrictions placed on women in schools, the workplace, and public space should be developed to deﬁne the environment in which the households make decisions regarding fertility, female education, and labor force participation. If these constraints are judged as relevant and adequate descriptions of the environment in which households operate, their effect on household behavior should be the next item on the research agenda. Micro studies of household behavior can be very useful in understanding the impact of the environment on the static and dynamic decisions of the households. In particular, they are indispensable in disentangling the effects of individual preferences and social norms as constraints on household behavior. For example, it is important to distinguish gender discrimination owing to parental preferences from that which is a rational household response to the gender discrimination in the labor market. Evidence on the static decisions could come from aggregate data on fertility and labor force participation of women, but ideally, one would need micro data to link individual characteristics, such as income and education, to fertility and labor force participation decisions. By controlling for observed individual characteristics, micro studies can provide a closer link between social norms and demographic and labor market outcomes. For decisions to save and to accumulate human capital, the focus should be on the institutions of markets for credit, education, and labor. With respect to physical capital accumulation, the efﬁciency with which credit institutions can channel household savings toward productive investments should be determined. Do banks and other mechanisms of ﬁnancial intermediation provide households with good alternatives to investments in unproductive assets such as land? Micro studies of consumption smoothing over the life cycle and intergenerational transfers through inheritance and dowries can provide valuable information on the household motives to save. With respect to human capital, which is produced within households as well as in schools, two questions should guide country studies. First, to what extent is the productivity of human capital affected by state interventions in the labor market? Second, to what extent household investments in the education of boys and girls are driven by the gender inequalities owing to social norms? Studies of returns to education which compare returns to formal schooling in private versus public sector, large establishments regulated by labor laws versus those unregulated, can reveal how state interventions in the labor market affect household choices in the amount and type of human capital to accumulate.

190

Djavad Salehi-Isfahani

Acknowledgments I am grateful to Francois Bourguinon, Angus Deaton, Valery Makarov, Heba Handoussa, Hashem Pesaran, Lyn Squire, and two anonymous referees for helpful comments. They have no responsibility for my views or errors.

References Abrahamian, E. (1974), ‘‘Oriental despotism: the case of Qajar Iran’’, International Journal of Middle East Studies, Vol. 5(1), pp. 3–31. Ashraf, A. (1970), ‘‘Historical obstacles to the development of a bourgeoisie in Iran’’, pp. 308–332 in: M.A. Cook, editor, Studies in the Economic History of the Middle East from the Rise of Islam to the Present Day, Oxford: Oxford University Press. Al-Sayyid-Marsot, A. (1989), ‘‘Women and social change’’, in: G. Sabbagh, editor, The Modern Economic and Social History of the Middle East in its World Context, Cambrigde, UK: Cambridge University Press. Anderson, L. (1987), ‘‘The state in the Middle-East and North Africa’’, The Journal of Comparative Politics, Vol. 1(3), pp. 1–18. Assaad, R. (1997), ‘‘The effects of public sector hiring and compensation policies on the Egyptian labor market’’, World Bank Economic Review, Vol. 11(1), pp. 85–118. Assaad, R. and F. El-Hamidi (2002), ‘‘Female labor supply in Egypt: participation and hours of work’’, in: I. Sirageldin, editor, Population Challenges in the Middle East and North Africa: Towards the Twentieth First Century, London: I.B. Tauris and Cairo: American University of Cairo Press. Barlow, R. (1994), ‘‘Population Growth and Economic Growth: some More Correlations’’, Population and Development Review, Vol. 20(1), pp. 153–165. Barnett, S., S. Eken, M. Lockheed, and W. V. Eegen (1998), ‘‘Effective ﬁnancing of education: role of public and private sectors’’, Paper presented at the 1998 Annual Seminar of the Arab Monetary Fund and the Arab Fund for Economic and Social Development, February 21–22, 1998, Abu Dhabi. Becker, G.S. (1991), A Treatise on the Family, second ed., Cambridge, MA: Harvard University Press. Besley, T. and R. Burgess (2004), ‘‘Can labor regulation hinder economic performance? Evidence from India’’, The Quarterly Journal of Economics, Vol. 119(1), pp. 91–134. Bill, J.A. and C. Leiden (1974), Politics in the Middle East, Boston: Little, Brown and Company. Bishop, J.H. (1996), ‘‘Incentives to study and the organization of secondary instruction’’, pp. 99–160 in: W.E. Becker and W. Baumol, editors, Assessing Educational Practices, Cambridge MA: MIT Press.

Microeconomics of Growth in MENA

191

Bloom, D. and J. Williamson (1997), ‘‘Demographic transitions and economic miracles in emerging Asia’’, World Bank Economic Review, Vol. 12(3), pp. 419–455. Botero, J., S. Djankov, R.L. Porta, F.L. de Silane, and A. Shleifer (2003), ‘‘The regulation of labor’’, NBER Working Paper No. 9756, National Bureau of Economic Research. Bray, M. and P. Kwok (2003), ‘‘Demand for private supplementary tutoring: conceptual considerations and socio-economic patterns’’, Economics of Education Review, Vol. 22, pp. 611–620. Caldwell, J. (1986), ‘‘Routes to low mortality in poor countries’’, Population and Development Review, Vol. 12(2), pp. 171–220. Collier, P. and J.W. Gunning (1999), ‘‘Explaining African economic performance’’, Journal of Economic Literature, Vol. 37(1), pp. 64–111. Davis, J., R. Ossowski, J. Daniel, and S. Barnett (2001), ‘‘Stabilization and savings funds for nonrenewable resources: experience and ﬁscal policy implications’’, International Monetary Fund Occasional Paper No. 205, IMF: Washington DC. Deaton, A. (1997), Analysis of Household Data, Baltimore: The Johns Hopkins University Press. Easterly, W. (1999), ‘‘The lost decades: explaining developing countries’ stagnation 1980–1998’’, World Bank, mimeo. El-Erian, M., S. Eken, S. Fennel, and J. Chaurfour (1996), Growth and stability in the Middle East and North Africa, International Monetary Fund: Washington DC. El-Sanabary, N. (1993), ‘‘Middle East and North Africa’’, pp. 136–174 in: E.M. King and M.A. Hill, editors, Women’s Education in Developing Countries: Barriers, Benefits and Policies, Baltimore and London: The Johns Hopkins Press. ERF (2000), MENA Indicators 2000, Cairo: Economic Research Forum. Esfahani, H. and F. Taheripour (2002), ‘‘Hidden public expenditures and the economy in Iran’’, International Journal of Middle Eastern Studies, Vol. 34(3), pp. 691–718. Flynn, D. K. (1999), ‘‘Reconceptualizing the female labor force in Jordan: accounting for the informal sector’’, Paper presented in the ERF Sixth Annual Conference, October 28–31, 1999, Cairo. Gelb, A. (1988), Oil Windfalls: Blessing or Curse?, Oxford, UK: Oxford University Press. Goldin, C. (1995), ‘‘The U-shaped female labor force function in economic development and history’’, in: T.P. Schultz, editor, Investment in Women’s Human Capital, Chicago: University of Chicago Press. Grossbard-Schechtman, A. (1993), On the Economics of Marriage: A Theory of Marriage, Labor and Divorce, Boulder, CO: Westview. Guriev, S. and D. Salehi-Isfahani (2003), ‘‘Microeconomics determinants of growth around the world’’, in: G. McMahon and L. Squire, editors,

192

Djavad Salehi-Isfahani

Explaining Growth: A Global Research Project, IEA Conference Vol. No. 150, London: Palgrave, MacMillan. Jalali-Naini, S.A.R. (1997), ‘‘A look at inﬂation and monetary, foreign exchange, and credit policies’’, in: M. Nili, editor, The Iranian Economy, Tehran: Institute for Research in Planning and Development (in Persian). IMF (1996), Building on Progress, Reforms and Growth in the Middle East and North Africa, Middle East Department. Issawi, C. (1995), The Middle East Economy: Decline and Recovery, Princeton: Markus Wiener Publishers. Karshenas, M. (2001), ‘‘Economic liberalization, competitiveness and women’s employment in the Middle East and North Africa’’, in: D. Salehi-Isfahani, editor, Labor and Human Capital in the Middle East: Studies of Labor Markets and Household Behavior, Reading, UK: Ithaca Press. Kazemi, F. (2000), ‘‘Gender, Islam and politics’’, Social Research, Vol. 67(2), pp. 453–474. Lazear, E.P. (1990), ‘‘Job security provisions and employment’’, Quarterly Journal of Economics, Vol. 105(3), pp. 699–726. Mahdavy, H. (1970), ‘‘Patterns and problems of economic development in rentier states: the case of Iran’’, in: M. Cook, editor, Studies in the Economic History of the Middle-East, Oxford, UK: Oxford University Press. Mammen, K. and C. Paxson (2000), ‘‘Women’s work and economic development’’, Journal of Economic Perspective, Vol. 14(4), pp. 141–164. Manski, C. F. (2000) ‘‘Economic analysis of social interactions,’’ Journal of Economic Perspectives, Vol. 14(3) (Summer), pp. 115–136. Mason, K.O. (1997), ‘‘Gender and demographic change: what do we know?’’, pp. 158–182 in: G.W. Jones, R.M. Douglas, J.C. Caldwell and R.M. D’Souza, editors, The Continuing Demographic Transition, Oxford: Clarendon Press. McDonald, P. (2000), ‘‘Gender equity in theories of fertility’’, Population and Development Review, Vol. 26(3), pp. 427–439. Moghadam, V.M. (1998), Women, Work, and Economic Reform in the Middle East and North Africa, Boulder, CO: Lynne Reinner Publishers. Nashat, G. (1999), ‘‘Women in the Middle East, 8000 B.C.E.-C.E. 1800’’, pp. 5–72 in: G. Nashat and J.E. Tucker, editors, Women in the Middle East and North Africa, Bloomington and Indianapolis: Indiana University Press. Obermeyer, C.M. (1992), ‘‘Islam, women and politics’’, Population and Development Review, Vol. 18(1), pp. 33–57. Ortiz, V. and R. Cooney (1984), ‘‘Sex-role attitudes and labor force participation among young Hispanic females and non-Hispanic White females’’, Social Science Quarterly, Vol. 65(June), pp. 392–400. Pissarides, C., and Veganzones-Varoudakis (this volume), ‘‘Labor markets and economic growth in the MENA Region’’, Explaining Growth in the Middle East. J. B. Nugent and M. H. Pesaran, editors, London: Elsevier.

Microeconomics of Growth in MENA

193

Posusney, M.P. (1997), Labor and the State in Egypt, New York: Columbia University Press. Pritchett, L. (1999), Has education had a growth payoff in the MENA region? MENA Working Paper Series, No. 18, Washington DC: World Bank. Psacharopoulos, G. and X. Nguyen (1997), ‘‘The role of government and the private sector in fighting poverty, Washington D. C.: World Bank Publications. Rashad, H., and Khadr, Z. (2002). ‘‘The demography of the Arab region: new challenges and opportunities’’, in: Sirageldin, I. editor, Human Capital: Population Economics in the Middle East. London: I.B. Tauris, 2002. Reimers, C.W. (1985), ‘‘Cultural differences in labor force participation among married women’’, American Economic Review Proceedings, Vol. 198575(2), pp. 251–255. Richards, A. and J. Waterbury (1996), A Political Economy of the Middle East, 2nd ed., Boulder, Colorado: Westview Press. Said, M. (2001), ‘‘Public sector employment and labor markets in Arab countries: recent developments and policy implications’’, in: D. SalehiIsfahani, editor, Labor and Human Capital in the Middle East: Studies of Labor Markets and Household Behavior, Reading, UK: Ithaca Press. Salehi-Isfahani, D. (1999), ‘‘Labor and the challenge of restructuring in Iran’’, Middle East Report, Vol. 28(210), pp. 34–37. Salehi-Isfahani, D. (2000), ‘‘Demographic factors in Iran’s economic development’’, Social Research, Vol. 67(2), pp. 599–620. Salehi-Isfahani, D. (2002), ‘‘Population, human capital, and economic growth in Iran’’, in: I. Sirageldin, editor, Human Capital: Population Economics in the Middle East, London: I.B. Tauris and Cairo: American University of Cairo Press. Salehi-Isfahani, D. (2005), ‘‘Human resources in Iran: potentials and challenges’’, Iranian Studies, Vol. 38(1), pp. 117–147. Salehi-Isfahani, D. and R. D. Murphy (2004), ‘‘Labor market ﬂexibility and investment in human capital’’, Working paper, Department of Economics, Virginia Tech. Shaban, R., R. Assaad and S. Al-Qudsi (2001), ‘‘Employment experience in the Middle East and North Africa’’, in: D. Salehi-Isfahani, editor, Labor and Human Capital in the Middle East: Studies of Labor Markets and Household Behavior, Reading, UK: Ithaca Press. Shah, N. M. and S. S. Al-Qudsi (1990), ‘‘Female work roles in traditional oil economy: Kuwait’’, in: Vol. 6, 213–246 I. Sirageldin, editor, Research in Human Capital and Development. Statistical Center of Iran (1998), Employment and Unemployment of Households, Tehran, Iran. Tansel, A. (1994), ‘‘Wage–employment, earnings and returns to schooling for men and women in Turkey’’, Economics of Education Review, Vol. 13(4), pp. 305–320.

194

Djavad Salehi-Isfahani

Tansel, A. and F. Bircan (2004), ‘‘Private tutoring expenditures in Turkey’’, IZA Working Paper No. 1255, Bonn, Germany. URL: http://ssrn.com/ abstract=560283. Tunali, I. (1996), ‘‘Labor market implications of the demographic window of opportunity’’, The Forum, Vol. 3(4), p. 3. ERF: Cairo. Tunali, I. and Baslevent, C., (2000), ‘‘Estimation of female labor supply parameters when self-employment is an option’’, Mimeo. United Nations (1999), Human Development Report CD-Rom, New York: United Nations. UNDP (2002), Arab Human Development Report 2002, New York: United Nations Development Program. Wahba, J. (2001), ‘‘Returns to education and regional earnings differentials in Egypt’’, in: D. Salehi-Isfahani, editor, Labor and Human Capital in the Middle East: Studies of Labor Markets and Household Behavior, Reading, UK: Ithaca Press. World Bank (1997a). ‘‘Education in MENA: beneﬁts and growth payoffs now and then’’, Mimeo World Bank (1997b). ‘‘Spending on education in selected MENA countries’’, Mimeo, Washington DC World Bank (1999), Education in the Middle East & North Africa: A Strategy towards Learning for Development, Human Development Sector, MENA region. World Bank (2002), Arab Republic of Egypt: Education Sector Review – progress and Priorities for the Future, Washington, DC: The World Bank. World Bank (2004a), World Development Indicators CD-ROM, Washington DC: The World Bank. World Bank (2004b), Gender and Development in the Middle East and North Africa Women in the Public Sphere, MENA Development Report, Washington, DC: The World Bank. Young, M.E. (1995), Early Childhood Development: Investing in the Future, Washington DC: The World Bank. Yousef, T. (1998), ‘‘Demography, capital dependency, and growth in MENA’’, ERF Working Paper No. 9801, ERF: Cairo

PART II: Country-Speciﬁc Studies

This page intentionally left blank

CHAPTER 7

Sources of Economic Growth and Technical Progress in Egypt: An Aggregate Perspective Hanaa Kheir-El-Din and Tarek Abdelfattah Moursi Abstract Egypt’s economic growth record has been uneven during the period 1960–1998. The mean rate of growth was consistently high with an average of about 6 percent per annum. Recurrent fluctuations in the annual growth rate of real output were observed, associated with frequent changes in the political environment and in socioeconomic conditions. In this paper, the traditional total factor productivity (TFP) methodology is used to analyze sources of aggregate economic growth in Egypt. Two sources of growth in output are identified: changes in physical inputs (capital and labor) and disembodied technical innovations measured by growth in TFP. Unlike other studies postulating arbitrary estimates for factor shares liable to misrepresent actual TFP changes, a simple integrated approach is used. It is based on the Kalman filter algorithm (Hamilton, 1994), which allows for simultaneous estimation of the parameters of the aggregate production technology and of TFP within a multi-equation system framework. Based on these estimates, an aggregate growth accounting analysis for the role of physical inputs and of technical progress in economic growth is presented. TFP estimates are further used to provide a detailed chronological characterization of the structural pattern of economic growth. Then, employing regression analysis, the correlation between technical progress and a set of selected policy-related macroeconomic variables is examined. This helps in identifying stylized facts that are useful for pinpointing specific strategies to accelerate growth. The paper concludes with some reflections on the sources of economic growth in Egypt during the period of study.

Corresponding author. CONTRIBUTIONS TO ECONOMIC ANALYSIS VOLUME 278 ISSN: 0573-8555 DOI:10.1016/S0573-8555(06)78007-5

r 2007 ELSEVIER B.V. ALL RIGHTS RESERVED

198

Hanaa Kheir-El-Din and Tarek Abdelfattah Moursi

Keywords: economic development, technological change, economic growth, aggregate productivity JEL classifications: O10, O30, O40 1. Introduction Egypt’s economic growth record has been uneven from the early 1960s until the late 1990s. During the period 1960–1998, the mean rate of growth was consistently high with an average of about 6 percent per annum. Nevertheless, that period exhibited recurrent ﬂuctuations in the annual growth rate of real output because of frequent changes in the political environment and in socioeconomic conditions. Early in 1960 and further in 1961, a big nationalization wave brought the public sector in control of around 90 percent of the total investment, in what became known as the ‘‘Socialist Revolution.’’ In the First Five-Year Comprehensive Plan (1960–1965), the public sector dominated all nonagricultural activities. There was also signiﬁcant expansion in the provision of free education and health services and in securing guaranteed jobs for all ‘‘secondary technical’’ schools and university graduates. The main lines of the socialist economic policy orientation prevailed until the next major political turning point: the Infitah (or opening-up) of President Sadat in 1974. Under Infitah, starting in the late 1970s through the 1980s, there were new policies attempting to encourage private sector initiatives to develop and upgrade the infrastructure, to expand new urban and industrial clusters in the desert and to control population expansion. However, Egypt’s institutional and management setup remained unchanged. Consequently, a number of these policies proved to be too expensive and required revision. Moreover, macroeconomic imbalances, which were partly brought about by the high costs of implementing these policies, threatened the stability of the economy. The year 1991 witnessed a key turning point in Egypt’s modern economic history with initiation of an economic reform and structural adjustment program (ERSAP). The program aimed at removing macroeconomic imbalances and promoting economic efﬁciency. This was envisaged through market-oriented strategies based on elimination of price distortions, on relieving government budget from hefty consumption subsidies, on foreign trade deregulation, on intensifying the role of private sector in economic activity, on ﬁnancial and capital market reform and on encouraging foreign trade openness and improving Egypt’s potentials for more solid foreign trade relations especially with the West. While the program was successful in reducing both internal and external imbalances, its impact on economic growth during 1991–1998 has been disappointing for the government and for the Egyptian people, as it could not raise the average growth rate back to the pre-1991 levels, despite modest recovery in the level of economic

Sources of Economic Growth and Technical Progress in Egypt

199

activity by the beginning of 1992/93.1 The pivotal factor underlying this recovery appeared to hinge on increasing domestic supply of typically scarce foreign currency as a result of external debt forgiveness in return for Egypt’s political role in the Gulf War. That is, the success of the reform measures and apparent stability of the macroeconomic framework were not sufﬁcient to generate high sustainable levels of growth. This is not at all surprising. It has been known for quite some time that macroeconomic stabilization and reform are only necessary, but not sufﬁcient prerequisites for growth (Fischer, 1993). Recent years have witnessed extensive endeavors to ﬁnd ways to salvage the Egyptian reform program by reviving the previous high rates of growth. Various attempts have focused on identifying different aggregate variables and policy instruments that are conducive to the creation of an enabling macroeconomic environment, which in turn would stimulate economic growth. In this paper, we use the traditional total factor productivity (TFP) methodology to analyze sources of aggregate economic growth in Egypt during the period from 1960 to 1998. We identify two sources of growth in output: one that is attributed to changes in physical inputs (capital and labor) and the other owing to disembodied technical innovations measured by growth in TFP. The conventional TFP approach presupposes existence of prior information on factor shares in income. The information is typically available from one of two sources: detailed national accounts surveys or reliable studies that provide accurate estimates for the parameters of an aggregate production function (APF). Because of unavailability of regular time series data, both sources are inaccessible in the Egyptian case for the period that is examined. Alternatively, unlike other studies, we ﬁnd ourselves reluctant to postulate arbitrary estimates for factor shares, which are liable to misrepresent the actual TFP changes. Instead, we introduce a simple integrated approach based on utilizing the Kalman ﬁlter (KF) algorithm (Hamilton, 1994). Our approach allows for simultaneous estimation of the parameters of the aggregate production technology and of TFP within a multiequation system framework. The main objective of the paper is to use the TFP estimates, derived from that system, to provide a detailed chronological characterization of the structural pattern of economic growth during 1960–1998. Then, employing regression analysis, we examine the correlation between technical progress and a set of selected policyrelated macroeconomic variables. Understanding the relation between technical progress and these variables should be helpful in identifying stylized facts that may be useful for pinpointing speciﬁc strategies to accelerate growth. The paper begins in Section 2 with a presentation of Kalman-ﬁltered estimates for TFP and for the parameters of the APF. Based on these estimates, Section 3 presents an aggregate growth accounting analysis for the role of

1

The average real rate of growth declined to about 4.6 percent per year during 1991–1998.

Hanaa Kheir-El-Din and Tarek Abdelfattah Moursi

200

physical inputs and of technical progress in economic growth. The relation between technical progress and each of the selected macroeconomic indicators is explored in Section 4 with the help of regression analysis. Section 5 concludes with some reﬂections on the sources of economic growth in Egypt from 1960 to 1998. The paper includes two appendices. In Appendix A, we provide a short review of the technique that is developed for the joint multi-equation system estimation of TFP and the parameters of the APF by Kalman ﬁltering. It also includes a comparison between the system results vis-a`-vis analogous estimates from a standard single-equation regression and from a model based on an autoregressive moving average (ARMA) in the errors speciﬁcation. Appendix B documents the sources of data used in the study.

2. Total factor productivity estimates (1960–1998) The standard neoclassical model presumes two potential sources of growth in the aggregate amount of output. They can be identiﬁed by decomposing the growth rate of real output into two parts.2 The ﬁrst is that which can be explained by the growth in the quantity of physical inputs used in production. The second is a residual that is left after accounting for the growth in inputs. Under the assumption of competitive markets, this residual growth is typically taken to represent gains in output that can be explained by improvements in the level of technological efﬁciency with which physical inputs are used. Hence, the residual growth, dubbed TFP growth (TFPG), measures disembodied factor-neutral technical progress that emanates from production process innovations that do not necessarily require input quality improvements and/or from improvements in the quality of inputs because of mutations of unknown and known origin in the production process associated with exogenous impulses. Such impulses may include changes in factors affecting the motivation of workers and the productivity of capital and changes in levels of education, health and standard of living of members of the society. The residual growth would also contain measurement and any other unknown statistical errors in the output or input data. Putting it formally, we assume that output (Y) can be adequately approximated by a two-factor linearly homogenous Cobb–Douglas APF with Hicksneutral technical progress, which describes the relation between output and factor inputs – capital stock (K) and labor force (L) – represented in physical units. Under the assumption of competitive equilibrium and marginal changes in output and factor inputs, the APF can be written in the form Y t ¼ At K at L1a t

2

The decomposition process is commonly known as growth accounting.

ð1Þ

Sources of Economic Growth and Technical Progress in Egypt

201

where the coefﬁcient a, 0oao1, is the share of capital in income and At denotes TFP. The exponents of capital and labor in the production function are set to add up to unity in compliance with the constant returns to scale (CRS) assumption. By dividing by L, taking natural logs and differentiating equation (1) totally with respect to time, we can express the production function in its intensive (per-worker) form y_ t ¼ a k_t þ a_ t

ð2Þ

_ measure the logarithmic growth rates of such that the lower case letters, y_ and k, output (dy) and of capital (dk) per-worker (where y ¼ ln(Y/L) and k ¼ ln(K/L)) and the variable a_ is an unobserved index of technical progress (TFPG) or dA=A where a ¼ ln(A). In standard applied aggregate growth accounting studies for individual countries, it is critical to establish a priori information on the size of the elasticity of output with respect to capital, a, before estimating TFP residually from a production function.3 Information on the size of a is often postulated on the basis of estimates for capital’s share of income obtained from different empirical studies and international evidence from developed as well as developing countries (Collins and Bosworth, 1996; Harrison, 1996; Maddison, 1987). Given the selected value of a, an estimate of TFPG in period t can be numerically computed as a residual from equation (2), after rearranging terms, conditional on observed (logarithmic) changes in the time series {yt} and {kt} over some sample period t ¼ 1, y, T, by using the formula a_ t ¼ y_ t ak_t

ð20 Þ

When the residually calculated TFPG estimates are averaged over the sample period, the mean gives a cumulated estimate of the effect of shifts in the production function over the relevant sample horizon that are not explained by the growth rates of inputs. Apart from high probability of being biased (Chen, 1997), the estimates of TFPG derived from equation (20 ) are also sensitive to the arbitrarily predetermined size of the capital share. Alternatively, recent studies conﬁrm that the TFP estimates are also highly sensitive to the chosen magnitude of a (Dowling and Summers, 1998). Naturally, such ﬁndings cast doubt on the efﬁciency and robustness of estimates of technical progress derived using the conventional growth accounting methodology.

3

Regression models have been used occasionally, as an alternative to growth accounting exercises, to estimates TFP. In such models, TFP estimates are based on the size of the estimated constant term and/or time trend from an APF regression. Such estimates are usually subject to several statistical limitations including sensitivity to selected sample period and multicollinearity between explanatory variables (Dowling and Summers, 1998).

202

Hanaa Kheir-El-Din and Tarek Abdelfattah Moursi

We think that these limitations may be overcome by simultaneous estimation of TFP and the parameters of the underlying APF (Solow, 1957). In this way, subjective determination of the value of a is completely circumvented. More speciﬁcally, the innovation of this paper is estimating jointly both unobserved TFP and the share of capital income (a) using observed output and capital perworker time series. The estimation is based on a state-space representation of the multi-equation system composed of the APF, equation (1) and of a process explaining the temporal dynamics of TFP. Sequential estimates of at for the period 1960–1998 are derived from the state-space model by means of the KF algorithm. We present a review of this technique in Appendix A. TFP is estimated with annual time-series data for Egypt on output and capital per-worker during 1960–1998. The series yt and kt are obtained by dividing real GDP and physical capital stock, throughout the selected sample period, by the number of employed workers. This number is determined by adjusting the total labor force by the unemployment rate in each year. Appendix B lists the sources of data on real GDP, real capital stock, total labor force, unemployment rate and on the other variables that are used in the rest of the study. The successive KF estimates for (logged) TFP measured up to a scalar from 1960–1998 are portrayed in Figure 1. Each estimated value for at represents the shift factor of the APF in a given year.4 Following Solow (1957), the estimates of at can be employed to construct an APF that is represented by drawing a scatter plot of yt-at against kt, with at valid only up to a scalar measure. The APF is sketched in Figure 2. The diagram discloses a remarkably smooth and tight ﬁt for the estimated functional relation. Figure 2 reveals two striking characteristics of the APF for Egypt during the period 1960–1998. First, the plotted function exhibits visible evidence of diminishing marginal returns to capital, k, which is marked by an apparent concave curvature. Second, a set of observations in the plot tend to lie distinctly higher, parallel and toward the right of the points in the main scatter such that values of capital per capita (k) around 6 seem to match with more than one value for y-a. In addition, it appears that these maverick points encompass observations pertaining to the last 8 years in our sample (1991–1998). It is perhaps important to note that the displacement of the observations for 1991–1998 cannot be attributed to an upward shift in the APF, as any upward shift in that function is already totally taken into account by means of the variable, At. Solow (1957) provides three possible views on the sources of such displacement, a couple of which seem relevant here. One explanation is related to the possibility of existence of measurement errors in the time series involved in

4 In general, the level of TFP per se, as opposed to TFPG, is not analytically useful since at is determined arbitrarily. Hence, our interest in the TFP level series is conﬁned to examining changes in its trend, to identifying disturbances and upturn and downturn points in the course of its historical evolution and, naturally, to computing TFPG.

Sources of Economic Growth and Technical Progress in Egypt

Figure 1.

203

Kalman Filter estimates for TFP (a) (1960–1998)

1

Period III

0.9 0.8

Period II

0.7 Period I

0.5

Assassination of Sadat Sadat journey to Jerusalem and repercussions of Peace Treaty

0.4 0.3 1973 war 0.2 0.1

Recovery with Infitah

1967 war and repercussions of war in Yemen

0 60

65

70

Figure 2.

75

80 Year

85

90

95

Aggregate production function

0.3

0.25

0.2

y-a

a

0.6

Stabilization program and Egyptian support in Gulf war Mid-1980 s slowdown

0.15

0.1

0.05

1

2

3

4

k

5

6

7

204

Hanaa Kheir-El-Din and Tarek Abdelfattah Moursi

estimation, especially those with respect to physical capital. Undoubtedly, such errors are highly probable due to the crudeness of the methodology used in the construction of physical capital stock (Nehru and Dhareshwar, 1994) and/or to problems generated in the process of aggregation of physical inputs. Aside from errors in measurement, another explanation pertains to overestimation of the productivity increase because of the unusual decline or underestimation of physical capital input services per-worker. The continued contraction in aggregate investment expenditure that started a few years before the mid-1980s through the early 1990s – partly in response to the decrease in government investment in public sector companies in view of rising ﬁscal deﬁcits – led to a decline in the stock of capital. Moreover, two years during the period 1991–1998 (speciﬁcally 1993 and 1996) witnessed marked increases in the level of employment to popularize newly instated governments.5 This resulted in a sharp fall in the capital-labor ratio, k. Yet, ostensibly limited labor efﬁciency may cause a fall in k to deceptively represent a rise in the productivity of capital. Additionally, speeding up the pace of the privatization program in the early 1990s eventually led to an increase in the supply of shares available on the market (EFG-Hermes, 1997). Hence, stock prices started tumbling in mid-October 1994 and continued to fall until June 1996. Throughout most of this period, many trading companies were considerably undervalued, with the stocks of several privatized companies trading at prices below their initial public offering price. However, cheap deals and falling prices at the time were still not sufﬁcient to induce the public to increase their demand for stocks. It is likely that underestimation of capital inputs, resulting from undervaluation of stock prices, led to overestimation of productivity increases, as reﬂected in the displacement of the 1991–1998 observations in the production function scatter diagram (Figure 2). 3. Role of input changes and technical progress in economic growth Following the mainstream literature, we assume two sources for growth in output; the ﬁrst is associated with growth in physical inputs and the second with technical progress. In this section, the contribution of physical inputs and technical progress to the process of economic growth in Egypt from 1960–1998 is examined within an aggregate growth accounting framework. 3.1. Role of physical inputs During the period 1960–1998, the annual rate of real GDP growth averaged 5.90 percent. In that period, the growth of GDP was characterized by considerable volatility and frequent ﬂuctuations around its average with values ranging

5

In 1993 and 1996, new governments came into ofﬁce. Dr. Sidki’s government was reinstated in 1993 and in 1996, government responsibilities were transferred to a new Prime Minister, Dr. El Ganzouri.

Sources of Economic Growth and Technical Progress in Egypt

205

between a minimum of 0.25 percent (1967) and a maximum of 15.44 percent (1976), with a standard deviation of 3.20 percent. Analysis of the impacts of changes in TFP and in physical inputs on output growth over a relatively long time horizon requires cautious periodization that permits identiﬁcation of broadly homogenous subperiods during which the main drivers of growth are rather stable. The high frequency and volatility of changes in real GDP generate extreme unevenness in the pattern of economic performance that precludes the possibility of relying on aggregate growth to dichotomize the entire period, 1960–1998, meaningfully into homogenous subperiods. Alternatively, important interactions among technical progress, aggregate growth and sociopolitical development processes make it reasonable to rely on TFP to provide a benchmark for periodization. Despite its smooth positive trend, the TFP graph displays two main breaks in aggregate productivity in 1977 and 1991 (Figure 1).6 They divide the entire period into three principal subperiods: 1960–1976, 1977–1990 and 1991–1998. While each of the three subperiods witnessed some turning points, the behavioral pattern of TFP seems uniform within each one of them. We adopt this periodization for our growth accounting study. Table 1 and Figures 3–5 display the behavior of output (y), labor and capital (k) in the three subperiods. With few exceptions, Table 1 as well as Figure 4 show that average employment growth during the whole period 1960–1998 is noticeably stable, with only marginal increase from one subperiod to the next.7 Excluding the extreme values of 1976 as well as 1977, the average employment growth is even more stagnant with a relatively small standard deviation of less than 1.1 percent.8 The other peaks and troughs in the employment growth series are relatively infrequent, mainly arising because of unanticipated shocks (e.g. the 1967 war) or changes in the political environment (e.g. those causing the employment increases of 1993 and 1996). There are several reasons underlying steady employment growth rates in spite of rapid population expansion and high fertility rates. One reason is due to

6

The year 1977 designates the beginning of a period of generous government social and investment spending following the bread riots of January 17–18, 1977, whereas 1991 marks the beginning for the implementation of ERSAP. 7 The average rate of growth of employment during 1960–1998 is about 2.26 percent (with values ranging between 3.34 percent in 1976 and 7.19 percent in the following year) with a modest standard deviation of 1.61 percent. 8 Employment rates for 1976 and 1986 are obtained from the Population Census (PC). The rest of the data in the series come from the Labor Force Sample Survey (LFSS) (Moursi and Kamel, 1989). PC and LFSS data are not comparable. To overcome the incompatibility without losing pertinent information embodied in the PC data for the two census years, we adopt the following procedure. Adaptive exponential smoothing bias minimizing forecasts for census years are estimated using the combined LFSS/PC unemployment series. We use the forecasted estimates for census years, 1976 and 1986, in lieu of census ﬁgures from which the magnitude of the forecast differs signiﬁcantly from the corresponding actual PC data.

Hanaa Kheir-El-Din and Tarek Abdelfattah Moursi

206

Table 1.

Annual and average growth rates of output and inputs in selected subperiodsa

Average (trend) growth (percent) of Output Labor Capital Output per-worker Capital per-worker Annual average growth (percent) of Output Labor Capital Output per-worker Capital per-worker a

Period I (1961–1976)

Period II (1977–1989)

Period III (1990–1998)

4.88 1.96 5.01 2.92 3.05

6.08 2.07 10.08 4.02 8.01

4.55 2.86 2.38 1.69 0.49

5.74 1.92 5.95 3.81 4.03

6.40 2.45 10.06 3.95 7.61

4.57 2.61 2.80 1.96 0.19

All averages are calculated as a simple arithmetic mean.

Figure 3.

Real output (GDP) per-worker growth (1960–1998)

Sources of Economic Growth and Technical Progress in Egypt

Figure 4.

207

Labor growth (1960–1998)

underestimation of employment ﬁgures, which include civilian labor only and omit military force employment and consequently its ﬂuctuations. In addition, the extension of the compulsory schooling program – though largely unenforced – caused a decline in the number of children who are employed. The decrease of children participation in the labor market has recurrently slowed down employment growth especially during the 1970s and the early 1980s. In the 1990s, Egypt’s commitment to structural reform had a signiﬁcant effect on employment growth. Though the private sector was given the chance to play a more dominant role in the country’s development program, its share in employment increased only marginally because of a limited supply of skilled human resources. Slow employment growth in that period was aggravated by ﬁscal budget problems that almost completely hampered the government’s historical commitment to hire all fresh graduates in context of what is known as the employment drive. Two conclusions may be drawn in light of the above discussion. First, the stability of average employment growth leads to signiﬁcant covariability between output and output per-worker and between capital and capital-labor ratio growth rates. Hence, abstracting from productivity ﬂuctuations, the growth of physical capital would be sufﬁcient to characterize the behavioral structure of real output changes. Second, stability of employment growth accompanied by

208

Hanaa Kheir-El-Din and Tarek Abdelfattah Moursi

Figure 5.

Capital per-worker growth (1960–1998)

restrained human capital development and restricted labor skills – arising from an inefﬁcient formal and tertiary educational system – limit the role of labor participation in the growth process. Unlike labor, with the exception of the interwar years (1967–1973), there is considerable correlation between capital and output growth. Figure 6 portrays the procyclic behavior of physical capital growth. Moreover, we use TRAMO/ SEATS (Go´mez and Maravall, 1997) to detect outliers in the capital and the output series and in their respective growth rates. We ﬁnd that both level series are not contaminated by additive outliers (AO), transitory change (TC) or level shifts (LS). Statistical tests for the growth rates reject the presence of AO or LS. Yet, both the capital and output growth series are subjected to TC (starting in 1975 and 1976, respectively), which is characterized by a slowly decaying effect that could be detected until 1998. Not much is really known about the structure, vintage and productivity of capital stock at either the aggregate or sectoral levels in Egypt.9 With this deﬁciency

9

The only exception probably is the agriculture sector where land is the major capital asset. The quality and productivity of agricultural land in different areas is well known and documented.

Sources of Economic Growth and Technical Progress in Egypt

Figure 6.

209

Real output (GDP) and capital growth 1960–1998 (shading illustrates episodes when output exceeds capital growth)

in information, investment ﬂows have been recurrently used to account for the impact of changes in capital on output growth. We expect investment to be a good proxy for capital, considering the procedure that is implemented for constructing the physical capital series (Nehru and Dhareshwar, 1994). The early years of (sub)period I, 1960–1976, witnessed a rise in the level of investment that accompanied the First Five-Year Plan. As a result, output (and capital) growth accelerated during the 1961–1965 period. This trend was quickly reversed in the mid-1960s when the share of imports in GDP increased without being matched by a rise in the export share, mainly because of the cotton crop failure that occurred at the time and the poor export performance of newly nationalized companies. In addition, the foreign exchange drain that accompanied the Yemen war, the withdrawal of US foreign aid and the looming (1967) war with Israel ampliﬁed the downward pressure on domestic investment, which reached very low levels from the mid-1960s until 1973. The drop in investment was not accompanied by a similar decline in output growth. The growth was primarily generated by higher public consumption intended to accommodate increasing involvement of the government and its extensive network of public agencies in economic life and to confront the Israeli aggression.

210

Hanaa Kheir-El-Din and Tarek Abdelfattah Moursi

In the last few years of the ﬁrst subperiod, the level of investment started to rise again and remained comparatively high on average until the mid-1980s. Investment growth stimulated new production opportunities. The reasons for the increase in investment primarily included implementation of a number of proinvestment laws within the context of Infitah as well as an increase in the country’s hard currency resources because of rising world petroleum prices, a regional boom that encouraged inﬂow of workers’ remittances from abroad, reopening the Suez Canal and expansion in tourism. The rise in inﬂow of Arab aid and capital also permitted a large increase in investment. Thus, by 1988, investment reached a peak of 34 percent of GDP. The investment surge, translated into a rise in physical capital growth between 1973 up until 1985 (Figure 6), led to expansion in GDP growth. However, toward the end of the second subperiod (from the mid-1980s to 1990), investment started to fall sharply because of the collapse of international petroleum prices and domestic market distortions that overburdened the government budget. At the same time, these distortions undermined the efﬁcient allocation of investment across sectors and activities. Again, the fall in investment was reﬂected in the decline in the growth rates of capital stock as well as real output. Conversely, the third subperiod marked a fundamental reversal in the relation between capital expansion and output growth in contrast with the previous subperiod. During 1991–1998, the rate of capital growth relatively slowed down because of several reasons. Among these reasons are the sharp reductions in public investment associated with implementation of the reform program and contraction in private investment as a consequence of low domestic savings.10 Nevertheless, despite its low level, GDP growth started to rise by 1992/1993 (see Figure 6), mainly because of enhancements in TFP that were presumably associated with implementation of the stabilization program. The above discussion suggests that large investment allowances may frequently have promoted economic growth during the period 1960–1998. In Egypt, there has always been a domestic resource gap. The discrepancy between domestic savings and investment has usually been bridged by external ﬁnancial resources. Considering these facts, it sounds plausible, at ﬁrst glance, to infer that higher levels of investment, which may be achieved through raising the domestic saving rate and encouraging foreign ﬁnancial capital inﬂow, would result in an immediate increase in economic growth. Indeed, it is rather hard to imagine that such an inference could be questionable. After all, most technical innovations are in a way embodied in some new forms of machinery, equipment and acquisitions that can only be made available through accumulation of investment (Solow, 1957). Our results,

10

The fall in investment is partly seen as an outcome of lower inﬂow of workers’ remittances from abroad and poor performance of the export sector.

Sources of Economic Growth and Technical Progress in Egypt

211

however, give an unambiguous impression that accumulation of physical capital – though necessary – is not always sufﬁcient to sustain efﬁcient growth. Diminishing returns to physical capital (per-worker), illustrated by the curvature of the production function in Figure 2, reveals that increasing investment and thereby capital per-worker is not always conducive to growth. It also raises serious concern that the capital-labor ratio often may have exceeded its maximum efﬁcient threshold. The above ﬁndings suggest that aggregate economic efﬁciency in Egypt would not necessarily be hampered by decreasing capital intensity. This evidence downplays arguments that attribute slow growth to investment deﬁciencies and physical capital shortages.11 Nevertheless, it accentuates the importance of addressing problems underlying unproductive investment and poor economic and institutional management of the existing capital stock as the crucial element constraining growth in Egypt.12 Our results dwell, therefore, on the central role that TFPG and the factors that affect it have to play in the process of economic growth. 3.2. Role of technical progress While TFP is now playing an increasingly vital role as a major source of development in the Egyptian economy,13 from 1960 until 1990, the contribution of TFP to aggregate production growth was negligible. However, with the implementation of ERSAP in 1991, technical progress became the dominant contributor to real GDP growth during the 1990s. We examine that contribution in this section. Table 2 is more or less self-explanatory. It records the large contribution of capital to growth (especially in (sub)period II) and provides further justiﬁcation for diminishing returns to capital (Figure 2). It also reveals that the considerably larger role of TFP in output growth is a direct consequence of the sharp decrease in the contribution of physical capital rather than a result of the deterioration in the share of labor in growth, which actually almost doubled between the two (sub) periods II and III. Figure 7 shows that during the period 1960–1998, TFP was sensitive not only to variation in the level of input growth but also, more importantly, to changes in the aggregate economic and political environment. The diagram shows that temporary dips in TFP coincided with political disturbances resulting from (a) the repercussions of military involvement in Yemen, (b) the two major wars in

11

It is interesting, for example, to observe the similarity of output growth rates for the subperiod between 1967–1976 and the subperiod between 1990–1998 together with the marked dissimilarity of capital per-worker growth rates across the same subperiods, really helping us appreciate the signiﬁcance of this evidence. 12 Such problems include choice of unsuitable and/or outdated machinery, deﬁcient maintenance policies for equipment and incompetent laborers who are unable to deal with modern technology. 13 Recently, Easterly and Levine (2001) document the critical role of TFP and argue that it accounts for most cross-country differences in growth and income.

Hanaa Kheir-El-Din and Tarek Abdelfattah Moursi

212

Table 2.

Input and TFP contributions to economic growth Period I (1960–1976)

Contribution of labor to GDP growth (percent) Share of labor in GDP growth (percent) Contribution of capital to GDP growth (percent) Share of capital in GDP growth (percent) Annual average growth of TFP (percent) Growth of TFP (percent) Contribution of TFP to GDP growth (percent) Share of TFP in GDP growth (percent)

Period II (1977–1989)

Period III (1990–1998)

0.69

0.73

1.01

14.10

11.94

22.12

3.25

6.54

1.54

66.57

107.45

33.88

3.48

1.42

3.98

2.96

0.89

4.58

0.94

1.18

2.00

19.33

19.40

44.00

1967 and 1973 and (c) the cold-shoulder treatment and withdrawal of investment inﬂows from Arab countries to Egypt and assassination of President Sadat by fundamentalist militants in the aftermath of the Camp David Peace Treaty with Israel. The graph suggests, however, that productivity is inﬂuenced by economic forces,which, unlike political events, are liable to have somewhat more lasting impact on TFP. The economic slowdown of the 1980s – instigated by the decline in petroleum and tourism revenues and Arab aid and by investment cuts – had a lingering effect on TFP during 1977–1989. Alternatively, recovery with Infitah and then the reform program accompanied by increase in hard currency availability, because of debt forgiveness, sent TFP rising a little bit above the pre1977 levels. These observations are underscored in Figure 7, which shows that the average annual TFPG remained almost the same during 1960–1976 and 1990–1998, despite serious military turbulence that occurred in (sub) period I. Moreover, looking at Figure 6 and Table 2, we observe that the ﬁrst and third (sub) periods were characterized by relatively higher shares of TFP in GDP growth; and unlike the second (sub) period (II), the rate of output growth exceeded that of the capital in both (sub) periods (I and III). Our observation discloses an inverse relation between the contributions of capital and TFP to output growth, at least within the period under examination. The estimates of TFPG, portrayed in Figure 7, provide further knowledge about the behavioral characteristics of technical progress that are not discernible

Sources of Economic Growth and Technical Progress in Egypt

Figure 7.

213

TFP growth (1960–1998)

from an examination of the TFP levels alone. We start by presenting three stylized facts that are deduced from the observed behavior of technical progress during 1960–1998. First, despite an abrupt decline in the average TFPG during the economic slowdown of the late 1970s and the 1980s (Figure 7), on the whole, the dynamics of technical progress appear to be smooth and normal. Normality is conﬁrmed by the rejection of signiﬁcant skewness and kurtosis in the TFPG series. In addition, TFPG is stationary during 1960–1998, with random ﬂuctuations around a constant mean of approximately 2.83 percent. The stationarity of TFPG is conﬁrmed by means of a Dickey–Fuller (DF) test. The test coefﬁcient and some basic descriptive statistics for technical progress are reported in Table 3. Second, TFPG is uncorrelated with the amount of available capital per-worker. This is revealed by means of an excessively small coefﬁcient of determination (approximately 0.67 percent) obtained from a linear regression of capital per-worker, k, on TFPG and a constant term. Third, because of variations in the relative contribution of inputs, output growth may not always be positively associated with TFPG. These facts are important for understanding the nature of the contribution of technical progress to aggregate growth. The stationarity of TFPG unveils information about its behavior when the economy is temporarily subjected to

214

Hanaa Kheir-El-Din and Tarek Abdelfattah Moursi

Table 3. Statistic Mean (percent) Standard error of mean (percent) Skewnessa a

Descriptive statistics for TFPG Estimate 2.829 2.552 0.113 (0.784)

Statistic Kurtosisa t-test (with constant term)

Estimate 0.939 (0.953) 4.269

Signiﬁcance level reported in parentheses.

unanticipated disturbances. On average, the effects of transient exogenous shocks (e.g. war and assassination of President Sadat) on technical progress are typically short-lived and tend to dissipate quickly, in spite of their considerable impact on TFPG. Thus, while ephemeral impulses associated with introduction of technological innovations or new economic policy (e.g. announcement of Infitah) might have a big impact on technical progress, they would undoubtedly fail to leave an everlasting impression on the average level of TFPG. Indeed, TFPG series, during the period 1960–1998 incorporate only two AOs (in 1976 and 1980) with no encounters of either TC or LS. The small number of AOs, present in TFPG series, rules out the possibility of perpetual introduction of technological or institutional improvements during the 39 years under consideration. Finally, comparing Figures 1 and 7 emphasizes the correspondence between the major turning points in the TFP and TFPG series. The smooth trending pattern characterizing the behavior of TFP, however, suggests that the series could be useful to identify major upturns and downturns in technical progress. So although the nonstationary nature of a masks the speciﬁc details pertaining to the structural behavior of technical progress, the TFP series should be useful for identifying turning points associated with main technological shocks and breaks in production processes. This supports our earlier decision to employ TFP for periodization to avoid ad hoc splitting-up of the sample period into subperiods. The absence of a signiﬁcant positive correlation between GDP changes and TFPG indicates that net shifts in the APF (that averaged approximately 2.83 percent per year over the entire period 1960–1998) are to a great extent associated with changes in the output scale. This result is a bit discouraging. In a sense, it is suggestive of the impotence of Egyptian economic policy (implemented under both planned and market-oriented systems) and of recent reform measures to fully exploit technical progress as a vehicle for promoting economic growth. In the following section, we try to identify different economy-wide variables that, if appropriately controlled, could contribute to raising TFPG and, hence, assert its importance as a key element for motivating the aggregate production and growth processes. 4. Relation between technical progress and selected macroeconomic indicators The growth literature cannot propose compelling theoretical justiﬁcations regarding the selection of a unique group of determinants for growth. Thus, the

Sources of Economic Growth and Technical Progress in Egypt

215

process of choosing speciﬁc explanatory variables from the set of possible determinants of growth has been essentially based on results from an ever-growing body of empirical cross-country inquiries (see for example, Barro, 1991; Barro and Lee, 1993; Barro and Sala-i-Martin, 1995; Sachs and Warner 1997, Makdisi et al., 2000; O’Connell and Ndulu 2000). We follow a conventional approach for selecting the control variables for our technical progress equation. Our choices come from a collection of economywide control variables proposed by other researchers, which are known to have more or less robust effects on growth, and, at the same time, have purportedly played an important role in shaping up the productive performance of the Egyptian economy since the 1960s.14 No causal order is implied by our choice of explanatory variables for the regressions. The dearth of speciﬁc information about socioeconomic forces that simultaneously affect growth and the selected regressors precludes any attempt to impose arbitrary assumptions that may erroneously enforce structural causal relations between the dependent and the independent variables. The remainder of this section presents the estimation results for the TFPG equation regression using the sample period 1960–1998. Our regression model considers six aggregate explanatory variables: general government consumption (GOV), exports (EXP) and imports (IMP) of goods and services, gross domestic ﬁxed investment (GDFI), foreign aid (AID) and credit to the private sector (CRPRIV), each deﬁned as a ratio to GDP.15 Inﬂation (INFLAT) is incorporated in the regression to represent macroeconomic stability. With the exception of inﬂation, the regressors are expressed in logarithmic ﬁrst differences. The (logarithmic) change in population (Dln POP), two dummy variables (DWAR and D91) and a constant term are added to the list of regressors. The dummy variable DWAR that takes on the value 1 in the war years (1967 and 1973) and 0 otherwise, is supposed to capture the destabilizing effects of war exigencies. The other dummy variable, D91 – set to unity

14

Brock and Durlauf (2001) maintain that a fundamental problem in growth analysis – referred to as open-endedness – relates to absence of a formal approach for choosing a unique set of explanatory variables that must be included in regressions. At the present, there does not seem to be a satisfactory way to deal with this problem. Nevertheless, we think that selection of regressors based on countryspeciﬁc growth-related experience should provide some justiﬁcation for our choices. 15 It would have been important to include budget deﬁcit and real interest rate in the list of explanatory variables that are likely to be correlated with technical progress. This might have allowed us to capture the crowding-out effect on private investment by public spending on public investment projects and by current expenditures. We expect such effect to be important given the weight of the public sector and the scope of government intervention in the Egyptian economy. We are unable to introduce these variables in the regression model, however, in view of the unavailability of the required data for a signiﬁcant subset of the estimation horizon.

Hanaa Kheir-El-Din and Tarek Abdelfattah Moursi

216

Table 4. Parameter Dln Dln Dln Dln Dln Dln

GOV/Y EXP/Y IMP/Y GDFI/Y AID/Y CRPRIV/Y

OLS regression estimates for aggregate variables (1961–1998) Value 0.111 0.082 0.131 0.091 0.007 0.077

(2.128) (2.447) (3.164) (3.233) (2.309) (4.449)

Parameter INFLAT Dln POP DWAR D91 Constant

Value 0.003 0.007 0.028 0.027 0.049

(4.699) (0.146) (2.358) (3.618) (7.509)

Note: Robust t-statistics in parentheses.

in 1991 and 0 in other years16 – is designed to account for the inﬂuence on technical progress of introducing reform measures at the beginning of the 1990s. The regression is estimated by means of the ordinary least squares (OLS) method.17 The estimated coefﬁcients and relevant robust t-statistics are exhibited in Table 4. The estimated R2 (percent) for the regression is 59.13. In general, the estimated relationship between the explanatory variables in the regression and technical progress are economically sensible. With the exception of population growth, the reported t-ratios are all statistically signiﬁcantly different from 0 at conventional levels. Tables 5a and 5b present a battery of statistical speciﬁcation and stability tests for the aggregate indicators regression (Johnston and DiNardo, 1997). Overall, the test results support the rejection of misspeciﬁcation and of instability of parameters of the estimated equation. First, the one-step forecasts for the period 1985–1992, reported in Table 5a, ^_ in reveal that the standard error range about the technical progress forecast, a, each year includes the actual value of TFPG. Second, the Ramsey RESET test – substantiates rejection of speciﬁcation error.18 Third, Table 5b records Hansen test statistics for joint and for individual coefﬁcients stability. The tests could not reject the null hypotheses of joint and individual parameter constancy.19 Fourth, panel 8. i, Figure 8, sketches the recursive residuals and relevant 72.0 standard error (SE) bands for the regression. Except for 1993, the recursive residuals lie

16

This dummy has been ﬁrst assumed to take the value 1 in all years starting from the implementation of the reform program in 1991 till the end of the period in 1998; however, its coefﬁcient was insigniﬁcant. The signiﬁcance of the coefﬁcient of the dummy variable, as deﬁned in the text, precisely reﬂects the instantaneous effect of policy changes introduced in 1991 and conﬁrms further that reform is a continuous process which warrants successive and sustained measures that have not been undertaken by the Egyptian government. 17 A DF statistic is computed to test for a unit root in the time series of each explanatory variable. All the test results (not reported) signiﬁcantly reject nonstationarity of the selected control variables. 2 18 The RESET test is implemented using only a^_ . 19 The p-values for all the reported stability statistics could not reject the null hypothesis as they exceed the top signiﬁcance level (0.20) for the test.

Sources of Economic Growth and Technical Progress in Egypt

Table 5a.

217

One-step forecasts analysis and parameter constancy and specification testsa

Year

a_

a^_

a_ a^_

Forecast SE

t-Value

1985 1986 1987 1988 1989 1990 1991 1992

0.014 0.007 0.015 0.014 0.008 0.008 0.024 0.045

0.024 0.007 0.006 0.001 0.017 0.022 0.019 0.007

0.010 0.014 0.009 0.013 0.009 0.014 0.006 0.037

0.020 0.023 0.030 0.024 0.021 0.023 0.025 0.022

0.480 0.610 0.313 0.554 0.414 0.614 0.224 1.665

Tests of Parameter Constancy (signiﬁcance level in square brackets): (i) Over 1985–1992: Chow test statistic F(8,16) ¼ 0.319 [ 0.947]. (ii) Over 1961–1998: RESET test statistic F(1,27) ¼ 1.686 [0.205]. a One-step forecasts are estimated based on the observations from 1960–1998.

Table 5b. Test Joint Variance Constant

Hansen’s instability test

Statistic

Test

Statistic

Test

Statistic

Test

Statistic

1.779 0.052 0.079

GOV/Y EXP/Y IMP/Y

0.046 0.052 0.137

INFLAT AID/Y GDFI/Y

0.055 0.161 0.204

CPRIV/Y WAR D91

0.075 0.079 0.041

between the SE bands, thereby corroborating the previous evidence on parameter stability. The same result is again conﬁrmed by a Chow test (see Table 5a). Finally, panels 8 (a–h) (Figure 8) portray the time varying parameter (TVP) estimates and 72.0 SE bands for the right side variables in the regression.20 A solid line is drawn in each panel to mark the level of the relevant OLS coefﬁcient reported in Table 4. As expected, the diagrams show negligible time variation for almost all the estimates of the explanatory variables in the regression within the sample period. The above statistical tests provide support for using the OLS results. Hence, they are employed to explain the relation between TFPG and the selected aggregate variables. During 1960–1998, public expenditures were generally allocated to pay wages and salaries for a massive number of civil employees and to support public sector industries and large-scale development projects (e.g. the High Dam and heavy industry projects). Furthermore, large ﬁscal expenditures were allocated to meet social welfare targets (e.g. subsidizing basic commodities consumed by people in low-income brackets and securing social services for a considerable

20

The TVPs are estimated by means of the ordinary KF technique.

0.50

8.a GOV/Y

0.6

0.25

0.4

0.00

0.2

-0.25

0.0

-0.50

-0.2

-0.75

-0.4

8.d GDFI/Y

0.02

218

Figure 8. Time-varying parameter (TVP) and recursive residuals for the TFPG regression. 8.g INFLAT

0.01 0.00 -0.01

1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 0.8

8.b EXP/Y

0.6 0.4

-0.04 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997

0.05

8.e AID/Y

0.03

-0.000

-0.050

0.00

-0.4

-0.075

-0.02

-0.100

-0.03

-0.125

-0.04 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 8.C IMP/Y

-0.150 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997

0.75

8.h WAR

-0.025

-0.01

-0.2

0.050 0.025

0.01

0.0

1970 1973 1976 1979 1982 1985 1988 1991 1994 1997

0.04

0.02

0.2

1.0

-0.03

8.f CPRIV/Y

1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 0.050

8.i Recursive Residuals and +/-2.0 SE Bands

0.50

0.5

0.025 0.25

0.0 0.00

0.000

-0.5 -0.25 -0.025

-1.0

-0.50

-1.5

-0.75 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997

-0.050 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997

1971 1974 1977 1980 1983 1986 1989 1992 1995 1998

Hanaa Kheir-El-Din and Tarek Abdelfattah Moursi

-0.02

Sources of Economic Growth and Technical Progress in Egypt

219

fraction of the population). Thus, our measure of government consumption is likely to pick up the adverse effects of distortions associated with nonproductive ﬁscal expenditures and with the size of government (Barro and Sala-i-Martin, 1995). Table 4 depicts a signiﬁcant negative correlation between government consumption (as a share of GDP) and TFPG with a large estimated coefﬁcient that is equal to 0.11. To offset this negative effect, the government ought to consider modifying the current structure of ﬁscal spending. Government expenditures must be geared more toward productive uses (e.g. investments in infrastructure, decentralized institutional development programs and scientiﬁc research) and less toward distortive uses (e.g. excessive government employment and conspicuous ﬁscal consumption). Foreign trade openness – measured as the ratio of imports and exports of goods and services to GDP – is frequently cited as a determinant of growth that is capable of promoting improvements in economic performance through stimulation of price competitiveness and technological diffusion and spillover beneﬁts. Because most countries in the MENA region have high trade ratios, Makdisi et al. (2000) argue that openness may not be an appropriate explanatory variable in growth regressions. Instead, they suggest using Sachs and Warner’s (1997) index that rates a country’s openness on the basis of a set of criteria related to the levels of tariff and nontariff barriers (NTBs) and to the degree of free trade and state control. We do not feel that Sachs and Warner’s index can be accurately computed in the case of Egypt. One reason is related to the paucity as well as the inaccuracy of the available data on NTBs. Instead, we capture the relation between foreign trade and technical progress by introducing export and import ratios, EXP/Y and IMP/Y, separately in the regression. This dichotomy allows us to take account of differences in the contribution of export and import ratios to TFPG arising from variance in the economic behavior of export and import industries. The negative correlation between IMP/Y and TFPG (Table 4) is interpreted as follows. On one side, the share of consumer goods in total imports is quite high, especially after the Infitah. On the other side, imports of physical capital goods, in the form of machinery and equipment, are often outdated. Therefore, while capital imports carry little technological spillover beneﬁts, high-tech imported consumer goods compete viciously with domestic production to the extent of restricting the economy’s production and technical capabilities. In contrast, export promotion compels domestic producers to adopt progressive modes of production to penetrate world markets. In addition, a portion of hard currency export revenues is frequently recycled to promote efﬁciency in exporting industries with further favorable impacts on productivity. These gains encourage technological spillover and know-how beneﬁts for other domestic producers. Together, all these spell signiﬁcant positive association between TFPG and the export ratio, as demonstrated in Table 4. Egypt is one of the largest recipients of foreign ﬁnancial assistance (AID) that has been needed to bridge the gap between the supply and demand for hard currency. Apart from being used to promote large economic development

220

Hanaa Kheir-El-Din and Tarek Abdelfattah Moursi

projects and welfare programs, foreign aid funds are regularly utilized to ﬁnance importation, particularly of basic staples (e.g. wheat), to build up foreign reserves, to stabilize the exchange rate and to ease balance of payments crises. Table 4 illustrates that foreign aid share in GDP is positively related to TFPG with a coefﬁcient estimate of less than 0.01. This weak relation seems somewhat puzzling, bearing in mind Egypt’s prominent position as a recipient of aid. The result, however, conﬁrms that some foreign assistance funds directed to the government or to the public sector are used unproductively. Frequently, bureaucratic and red tape constraints prevent effective utilization of these funds.21 The importance of the private sector in Egypt varied during 1960–1998 because of changes in the country’s political orientation. The changes are reﬂected on the amount of credit available to the private sector as a ratio to GDP (CRPRIV/Y).22 From 1960 to 1980, CRPRIV/Y was fairly low as public sector companies dominated production activities and received most of the available credit. Liberalization measures that were launched in the early 1980s gave a larger and more pivotal role for private sector investment in the economy. Accordingly, CRPRIV/Y increased, reaching a peak in 1998. The increase was not entirely benign. A great deal of that credit was indulged in unproductive highreturns activities including investment in the construction and contracting industry, in speculation on land and real estate and in importation of luxury goods. Moreover, corruption, lax judicial system and absence of a regulatory mechanism to control capital outﬂows permitted some business tycoons to illegally transfer a part of the credit funds abroad. Such transfers had perverse effects on economic performance as well as on the credibility of the private business environment in Egypt. We ﬁnd, therefore, that the private sector credit to GDP ratio and technical progress were signiﬁcantly negatively correlated with an estimated coefﬁcient of 0.08 (Table 4). According to Fischer (1993), inﬂation (INFLAT) reduces growth through its negative effect on investment and productivity expansion. In addition, he argues that the inﬂation rate unequivocally provides a reasonable measure for uncertainty of the macroeconomic environment. During 1960–1998, the average annual inﬂation rate in Egypt (measured by the consumer price index (CPI)) was approximately 10.0 percent with a large variance of 49.1 percent. The economic slowdown of the 1980s was accompanied by a sharp rise in the annual inﬂation rate, reaching a maximum of 23.9 percent in 1986. From the mid-1980s until

21

For example, the European Union (EU) recently warned the Egyptian government of the possibility of withdrawing foreign assistance funds intended to develop the Egyptian industrial sector – in compliance with the EU – Egypt Partnership Agreement – if Egypt failed to utilize these funds productively. 22 According to World Bank (2000), for some countries CPRIV may include credit to public enterprises. We have no clue if this were the case for Egypt. Anyway, we assume that the share of credit to public sector ﬁrms in CPRIV is negligible.

Sources of Economic Growth and Technical Progress in Egypt

221

1998, the annual inﬂation rate gradually fell to 4.3 percent in 1998. On average, the level of inﬂation and its variance were above what would have been desired for providing a stable economic environment. For instance, uncertainty proceeding from large inﬂation-induced distortions in the market for foreign exchange frequently spawned increases in the country’s real exchange rate. The real exchange rate increases led to a reallocation of factors toward production of nontradable goods and against export promotion. Such pattern of resource allocation reduced Egyptian producers’ competitiveness in both international and domestic markets with obvious negative effects on productivity and domestic investment. Table 4 shows that inﬂation is signiﬁcantly – though weakly – inversely correlated with TFPG. The estimated coefﬁcient of INFLAT implies that a rise in the yearly inﬂation rate of 100 percent is associated with a modest decrease in the rate of technical progress of only 0.3 percent per annum. It is possible that the low correlation is a statistical illusion caused by the deliberate doctoring of CPI data to keep ofﬁcial inﬂation ﬁgures below actual levels as a means of assertion of political and social stability.23 Apart from that, during most of the period under consideration, the prices of essential inputs and of basic goods and services were subjected to extensive subsidization and/or mandatory control measures purportedly designed to achieve national welfare targets. The rental values of agricultural land and housing were subjected to similar controls. In addition, the large fraction of labor force holding tenured jobs in government and public sector and labor laws that nearly prohibit dismissal of employees, except in highly unlikely situations, reduced the variability of wages and household labor income. Together with price controls, stability of income and wages may have restrained volatility of the inﬂation rate, which, in turn, artiﬁcially reduced macroeconomic uncertainty and hence the degree of correlation between price ﬂuctuations and TFPG. GDFI (as a percentage of GDP) contains public and private investments in plant and equipment, in industrial buildings, schools, hospitals, construction of roads and railways, etc. and in land improvements. These investments play a catalytic role in facilitating the development process. The productivity of public and private investments differs. It would have been more interesting to capture the selective impact of those differences on TFPG. Unavailability of suitable data rules out this option. According to our regression results, the expansion of domestic ﬁxed investments has the strongest positive correlation with technical

23

Perhaps the following data may help – at least in part – substantiate this allegation. We note that during the last quarter in 2003, annualized CPI inﬂation was about 3.54 percent according to ofﬁcial estimates (IMF, 2004). Meanwhile, Information and Decision Support Center (IDSC) (unpublished) estimates reveal that (annualized) increase in prices of basic food commodities such as rice, ﬂour, oil, sugar and tea reached about 64.09 percent, 15.18 percent, 24.37 percent, 26.42 percent and 27.38 percent, respectively, for the same quarter.

222

Hanaa Kheir-El-Din and Tarek Abdelfattah Moursi

progress. The relatively large parameter estimate conﬁrms the importance of investing in infrastructure and new machinery and equipment for encouraging TFPG (Solow, 1957). It is usually proposed that the increase of population (POP) has an unfavorable effect on technical progress, especially in developing countries. The main intuition behind this proposition is that with a ﬁxed amount of resources (land, capital, savings, etc.), the growth of population would immediately imply fewer resources per person and, thus, lower growth potential. Pritchett (1996) rejects this proposition because resources, primarily, the amount of physical capital and the level of education, are not ﬁxed. Furthermore, his evidence – using cross-section country data – suggests a positive correlation between physical capital growth and rapid population expansion. Figure 5 demonstrates that during the 30-year period from 1961 to 1990, with the exception of the interwar years 1967–1973, the annual growth rate of capital per-worker was high in absolute terms and exceeded the rate of population growth (roughly 2–2.5 percent per year).24 In contrast, the interwar years and the last subperiod 1991–1998 were characterized by low absolute levels of capital expansion and reversal in the relation between physical capital accumulation and population growth. Besides, the education drive (beginning with the 1952 revolution) has deﬁnitely increased educational standards through compulsory primary education and through raising the average number of years of schooling per person. Nevertheless, illiteracy and school dropout rates remain above acceptable levels, partly because of teaching facilities and resource constraints that limit the potential to increase educational opportunities. Together with the change in the relation between capital growth and population expansion, limited educational opportunities forestall the potential for a signiﬁcant positive correlation between TFPG and population growth. The estimated coefﬁcient for population growth is positive (approximately 0.01), but not statistically signiﬁcant (Table 4). The result implies a weak nonnegative association between TFPG and population expansion. Furthermore, it supports the proposition that fast population growth did not hurt the rate of technical progress much, since increases in population were accompanied by – albeit insufﬁcient – rise in the level of education. Finally, the estimated coefﬁcients of the two dummy variables, DWAR and D91, are statistically signiﬁcant and have the expected sign. The estimate for the (1967 and 1973) war dummy variable veriﬁes that Egypt’s persistent conﬂicts with Israel had moderate negative impacts on technical progress. Not only were the two wars politically destabilizing, but they also represented a severe drain on the country’s economic resources. Frequently, investments and foreign

24

Concavity of the APF provides further support for the proposition that the absolute level of accumulation of physical capital per worker was high.

Sources of Economic Growth and Technical Progress in Egypt

223

assistance from Arab and other countries were withdrawn from productive uses to ﬁnance the military forces. The hazardous repercussions of war discouraged domestic and foreign direct investments and encouraged their diversion to other countries in the region, e.g. Tunisia, Qatar and Oman, which provided safe havens for these investments in the region. At the same time, the wars motivated many skilled workers and some of the best Egyptian professionals and experts to migrate to Arab and Western (e.g. USA, Canada and Australia) countries in pursuit of better employment opportunities and higher living standards away from military hostilities. It is easy to understand how such resource and brain drains had adverse effects on the country’s potential for innovation and technical progress. On the whole, the stabilization program and the steps taken for raising economic competitiveness improved macroeconomic management and reduced both internal and external imbalances (World Bank, 2001). The combined effect of improved economic management and increased supply of ﬁnancial resources – spurred by foreign debt forgiveness – raised TFPG. This is demonstrated in Table 4 by a statistically signiﬁcant positive coefﬁcient estimate for the dummy variable D91 (0.027), which almost counterbalanced the negative association between war and technical progress (0.028). 5. Conclusion: reflection on sources of economic growth in Egypt (1960–1998) In this paper, we provide a synopsis of key factors that inﬂuenced the behavior of economic growth in Egypt during the period 1960–1998. This was done in two stages. In the ﬁrst, we studied the contributions of physical inputs (capital and labor) and TFP to aggregate production within a growth accounting framework. Then, using regression analysis, we studied the correlation between technical progress and a selected set of key macroeconomic variables that affect the aggregate performance of the economy. Statistical results from the regression analysis conﬁrmed that all, save one, of the chosen explanatory variables were signiﬁcantly associated with technical progress. Our ﬁndings indicated that from 1960 until the end of the 1980s, capital accumulation was the main driving force behind economic growth. It is puzzling why a country like Egypt with abundant labor and chronically low savings rate would opt for capital-intensive modes of production. We offer two explanations. First, despite tenacious domestic savings shortages, the availability of investment expenditures was not always a signiﬁcantly binding constraint. The geopolitical role of Egypt in the MENA region in general and in the Arab–Israeli conﬂict in particular constantly secured regular foreign investment inﬂows from Arab countries, from the West and/or from the ex-Soviet Union. Signiﬁcant fractions of these inﬂows were allocated to build capital for production and to ﬁnance major national projects. Second, during the peak of the socialist era, in the 1960s, much of the foreign inﬂow that accrued to the state was invested either in newly nationalized or in

224

Hanaa Kheir-El-Din and Tarek Abdelfattah Moursi

public sector companies. The companies were economically inefﬁcient and badly managed. The inefﬁciencies resulted in small marginal returns on investments and losses in proﬁtability. However, output growth was reasonably robust during that period as the government poured more and more investments into losing public companies. On the consumption side, the inward-looking orientation of the country generated perpetual increases in demand for domestic output regardless of quality. Almost the same setting prevailed in the 1970s and the 1980s because of dominance of the public sector and of government intervention in production and in the distribution and allocation of inputs. The substantial quantities of unqualiﬁed labor and prevailing employment laws fostered the adoption of capital-intensive production techniques, which kept the growth rate during 1960–1989 on average above 6 percent per annum. High levels of output growth, initiated by accelerated capital utilization, were helpful in hiding some of the negative effects of productive inefﬁciency and mismanagement on economic proﬁtability. The increase in capital intensity, however, was incapable of generating a sufﬁcient level of technological dynamism that could continuously induce positive technical progress (Steindel and Stiroh, 2001). The lack of dynamism, together with limited labor productivity, which is probably the main source for diminishing returns to capital exhibited by our APF estimates, caused a gradual reduction in the level of investment. The decrease in investment led to an unprecedented fall in the contribution of physical capital to growth after 1991, and, a consequent increase in the share of TFPG in economic growth from 1991 to 1998. The increasing importance of technical progress is better seen as an outcome of the decline in the rate of capital accumulation rather than as a result of basic improvements in the management and organization of the production sphere in the economy. That is why the rise in TFPG in that period has not been quite sufﬁcient to stimulate sustainable economic growth. Low saving and investment growth rates in Egypt curtail the prospects for a substantial increase in the rate of capital accumulation in the near future. Enhancing labor productivity requires radical repair of the educational system and implementation of human capital development programs, which by nature are not feasible in the short run. Thus, by elimination, it seems natural to focus on analyzing the behavior of technical progress and on the variables that inﬂuence it. We concentrate on macroeconomic variables. Our results show that TFPG is signiﬁcantly correlated with the main aggregate policy variables in Egypt. In particular, during the entire period from 1960 to 1998, reductions in inﬂation, government spending, imports and increases in exports, foreign aid and domestic ﬁxed investment are signiﬁcantly correlated with positive variation in technical progress. In sum, the relative abundance of inexpensive physical capital historically encouraged the adoption of capital-intensive methods of production. The accessibility of low-cost capital allowed a national political target of supporting unproﬁtable producers to be maintained, especially in the public sector and to

Sources of Economic Growth and Technical Progress in Egypt

225

encourage industrial production. Excessive use of capital inputs gave rise to diminishing marginal returns to capital. Hence, human capital and skilled labor shortages imposed serious constraints on aggregate production capacity. These constraints could be relaxed by introducing tangible improvements in the quality of human labor capital via suitable long-run education and training programs. In the short run, however, stimulating technical progress, through appropriate macroeconomic policy intervention, seems to be the most viable alternative that is available – at present – for raising economic growth. Acknowledgments The authors would like to thank Hashem Pesaran, Jeffrey Nugent, Nadeem Ul-Haq, Sherine El-Shawarby, Alaa El-Shazly and two anonymous referees for their extremely helpful comments. We are also grateful to Mai Ashraf, Mai El-Mossallamy, Sally Farid, Iman El-Shair and Rasha Reda for their valuable research assistance. Naturally, all errors and omissions remain exclusively our responsibility. References Abdellatif, L. (2003), ‘‘Egypt’s manufacturing sector: factor inputs and TFP over half a century’’, GDN Library: Research papers from GDN activities, revised version, http://www.gdnet.org/activities/global_research_projects/explaining_growth/country_studies/#G. Barro, R. (1991), ‘‘Economic growth in a cross-section of countries’’, Quarterly Journal of Economics, Vol. 104, pp. 407–433. Barro, R. and J. Lee (1993), ‘‘International comparisons of educational attainment’’, Journal of Monetary Economics, Vol. 32, pp. 363–394. Barro, R. and X. Sala-i-Martin (1995), Economic Growth, New York: McGrawHill. Bisat, A., M. El Erian and T. Helbing (1997), ‘‘Growth, investment and saving in the Arab economies’’, IMF Working Paper, WP/97/85. Brock, W. and S. Durlauf (2001), ‘‘Growth empirics and reality’’, The World Bank Economic Review, Vol. 15, pp. 221–224. CAPMAS (various issues), Labor Force Sample Survey, Cairo: CAPMAS. Chen, E. (1997), ‘‘The total factor productivity debate: determinants of economic growth in East Asia’’, Asian-Pacific Economic Literature, Vol. 11, pp. 18–38. Collins, S. and B. Bosworth (1996), ‘‘Economic growth in East Asia: accumulation versus assimilation’’, Brookings Paper on Economic Activity, Vol. 2, pp. 135–203. Dowling, M. and P. Summers (1998), ‘‘Total factor productivity and economic growth-issues for Asia’’, The Economic Record, Vol. 74, pp. 170–185.

226

Hanaa Kheir-El-Din and Tarek Abdelfattah Moursi

Easterly, W. and R. Levine (2001), ‘‘It’s not factor accumulation: stylized facts and growth models’’, The World Bank Economic Review, Vol. 15, pp. 177–219. EFG-Hermes (1997), Country and Capital Market Update March 1997, Cairo: Egyptian Financial Group. Fischer, S. (1993), ‘‘The role of macroeconomic factors in growth’’, Journal of Monetary Economics, Vol. 32, pp. 485–512. Go´mez, V. and A. Maravall (1997), ‘‘TRAMO and SEATS: instructions for the user,’’ Banco de Espan˜a, www.bde.es. Hamilton, J. (1994), Time Series Analysis, Princeton, New Jersey: Princeton University Press. Harrison, A. (1996), ‘‘Openness and growth: a time-series, cross-section analysis from developing countries’’, Journal of Development Economics, Vol. 48, pp. 419–447. Information and Decision Support Center (IDSC) (unpublished), Daily Prices for Basic Commodities Database, Information Sector, Cairo, Egypt: Information and Decision Support Center. IMF (2004), International Finance Statistics on CD-ROM, May edition, IMF. Johnston, J. and J. DiNardo (1997), Econometric Methods, Fourth Edition, New York: McGraw-Hill, International Editions. Maddison, A. (1987), ‘‘Growth and slowdown in advanced capitalist economies: techniques of quantitative assessment’’, Journal of Economic Literature, Vol. 25, pp. 649–698. Makdisi, S., Z. Fattah and I. Limam (2000), ‘‘Determinants of growth in The MENA countries’’, Paper for World Bank Global Development Network. Moursi, T. and A. Kamel (1989), ‘‘The relationship between unemployment and its dispersion and the rate of change of money wage rates and of prices in Egypt’’, Proceedings of the First Conference of the Economics Department, February 20–22, Faculty of Economics and Political Science, Cairo University. Nehru, V. and A. Dhareshwar (1994), ‘‘New estimates of total factor productivity growth for developing and industrial countries’’, Policy Research Working Paper No.1313, The World Bank. O’Connell, S. and B. Ndulu (2000), ‘‘Africa’s growth experience: a focus on sources of growth’’, Paper for World Bank Global Development Network. Pissarides, C. (2000), ‘‘Labor markets and economic growth in the MENA region’’, Paper presented at the Global Research Project: Explaining Growth sessions held in Cairo, Egypt, March. Press, W., B. Flannery, S. Teukolsky and W. Vettering (1988), Numerical Recipes in C, New York: Cambridge University Press. Pritchett, L. (1996), ‘‘Population growth, factor accumulation, and productivity’’, Policy Research Working Paper, R/1567, The World Bank. Sachs, J and A. Warner (1997), ‘‘Natural resource abundance and economic growth’’, Center for International Development and Harvard Institute for International Development.

Sources of Economic Growth and Technical Progress in Egypt

227

Solow, R. (1957), ‘‘Technical change and the aggregate production function’’, Review of Economics and Statistics, Vol. 39, pp. 312–320. Steindel, C. and K. Stiroh (2001), ‘‘Productivity: what is it, and why do we care about it?’’, Federal Reserve Bank of New York, http://www.ny.frb.org/ rmaghome. World Bank (1994), ‘‘Private sector development in Egypt, the status and the challenges’’, A World Bank Report prepared for the Conference Private Sector Development in Egypt: Investing in the Future, Cairo, October 9–10, 1994, The World Bank. World Bank (2000), World Development Indicators 2000, on CD ROM, The World Bank. World Bank (2001), Egypt social and structural review, Report No. 22397-EGT, The World Bank. Appendix A: Review of the methods used to estimate the APF Consider an (n 1) random vector of variables, zt, containing the observed values of n variables at date t with t ¼ 1, y, T. A state-space representation of the dynamics of z in terms of observed (xt) and unobserved (nt) vectors can be given by the following system of equations zt ¼ Bxt þ C nt þ wt

ðA:1aÞ

nt ¼ Dnt1 þ vt

ðA:1bÞ

where xt is a (k 1) vector of predetermined random variables, nt is an unobserved (or state) vector of dimension (r 1) and B 0 and C 0 are (n k) and (n r) matrices of parameters, respectively. The observation equation (A.1a) relates the observable variables zt to the unobservable states, such that wt is a vector of white noise measurement errors. The state equation (A.1b) describes the dynamics of the evolution of the vector of states over time. The matrix of parameters D is of dimension (r r) and vt is a vector of random shocks. Assume that the disturbances vectors vt and wt are uncorrelated at all lags and that G for t ¼ t 0 ðA:2aÞ Eðwt wt Þ ¼ 0 otherwise Eðvt v0t Þ ¼

H for t ¼ t 0 otherwise

ðA:2bÞ

where G and H are matrices of dimension (n n) and (r r), respectively. The unknown value of TFP in period t can be computed by casting it within a state-space representation. Drawing on Solow’s (1957) highly touted study, a Cobb–Douglas APF, in intensive form, can be written as yt ¼ b þ akt þ at

ðA:3Þ

Hanaa Kheir-El-Din and Tarek Abdelfattah Moursi

228

where lower case letters represent the logs of output and capital per-worker (y and k, respectively), a denotes the log of TFP and b is a constant parameter.25 We stipulate that the behavioral dynamics of TFP exhibits a considerable degree of persistence over time. Consequently, the evolution of the unobserved variable of states, at, can be captured comprehensively by means of a model, which belongs to the class of ARMA processes of orders p and q, ARMA(p,q). This implies that the dynamics of TFP are stable and fairly simple during the sample period under consideration.26 Accordingly, the ARMA(p,q) model for at (t ¼ 1, y, T) in our state-space framework for TFP can be presented as follows:

at ¼

p X

di ati þ t þ

i¼1

q X

gj tj

ðA:4Þ

j¼1

where et N(0,s2) are identically independently distributed (iid) white noise shocks and di (i ¼ 1, .., p) and gj (j ¼ 1, .., q) are any constant coefﬁcients. Our model, composed of equations (A.3) and (A.4), describes a state-space system for TFP with equation (A.3) representing the observation equation and equation (A.4) being the state equation. With the exception of dropping wt in equation (A.1a), our model corresponds to the prototypical state-space representation described by equations (A.1a and A.1b) and (A.2a and A.2b). By direct substitution of equations (A.3) and (A.4) into the state-space system (A.1a and A.1b) and (A.2b), and noting that n ¼ 1, our growth model can be written in general form with the following deﬁnitions: B0 ¼ ½b a xt ¼ ½1 kt 0 C 0 ¼ ð1; 0; :::; 0Þ

25

Assuming CRS, the production function speciﬁcation represented by equation (5) corresponds to the form y ¼ at.f(k) where f is a twice continuously differentiable concave function satisfying the standard neoclassical production function regularity conditions. With this speciﬁcation, any shift in the APF is exclusively attributable to the shift term at. (Solow, 1957). 26 Hence, the assumption rules out sudden economic improvements, which might result, say, from transfer of technology from abroad. We doubt that Egypt experienced such improvements during the period under study. To be sure, we explored the possibility of AO, TC and LS in the capital, output and capital per worker series with TRAMO/SEATS (Go´mez and Maravall, 1997). With the exception of one AO in series k in 1976, we are not able to detect any other outliers/shocks in the three series.

Sources of Economic Growth and Technical Progress in Egypt

2

d1 61 6 6 60 6 6: 6 6 6: 6 60 6 D¼6 60 6 60 6 6 6: 6 6: 6 6 4: 0

d2 0

... ...

dp 0

0 g1 0 0

g2 ...

... ...

gq 0

1

...

0

:

:

:

:

:

: :

: :

: :

: :

: :

: :

: :

: :

... ...

... ...

1 0

0 0 0 0

... ...

... ...

0 0

...

...

0

0 1

0

...

0

: :

: :

: :

: :

1 :

... :

0 :

: ...

: ...

: 0

: : 0 0

: ...

: ...

: 1

0 :

229

3

0 07 7 7 : 7 7 : 7 7 7 : 7 7 07 7 7 07 7 07 7 7 : 7 7 : 7 7 7 : 5 0

nt ¼ ðat ; at1 ; . . . ; atp ; t ; t1 ; . . . ; tq Þ0

where the matrices and vectors B 0 , xt, C 0 , D, nt and H are of dimension (1 2), ((p+q+2) 1), (2 1), ((p+q+2) (p+q+2)), ((p+q+2) 1) and ((p+q+2) (p+q+2)), respectively. In addition, we note that each of the upper and lower square submatrices, to the right of the vertical dotted line in matrix H, are of dimension ((p+2) (p+2)). Given suitable initial values for a, s2 and the parameter vectors d and c, the matrices D, nt and H would be known and the Kalman-ﬁltering technique could be implemented to compute optimal estimates of at, conditional on current and past values of yt and kt. The parameters and the state variable that appear in equations (A.3) and (A.4) and the matrices and vectors portrayed above are estimated for the Egyptian economy by means of the Kalman ﬁltering technique using a sample of annual data on y and k.

230

Hanaa Kheir-El-Din and Tarek Abdelfattah Moursi

The initial guesses for the capital share in income, a, and the coefﬁcient b are estimated from a double-log OLS regression of yt on real capital per-worker and a constant. In addition, a dummy variable (D91t), that takes on the value unity in 1991 and 0 otherwise, is added as an explanatory variable in the regression to account for variance in the level of utilization of capital per-worker owing to the introduction of economic reform measures. The highly signiﬁcant statistical estimate of the dummy variable coefﬁcient in that regression (Table A.1) strongly suggests modifying the observation equation. Accordingly, equation (A.3) and the vector B0 are rewritten, respectively, as follows: yt ¼ b þ akt þ fD91t þ at

ðA:30 Þ

and B0 ¼ [b a f] where f is any parameter. The initial guess for f that is employed in the Kalman-ﬁltering algorithm is taken from the same OLS production function regression from which a is estimated. The estimates of a and f from the production function regression are also used to compute a crude estimate of TFP (aˆt) during the sample period such that ^ t f^ D91t a^ t ¼ yt b^ ak ðA:5Þ ^ ^ where a^ , b and f are the parameter estimates of a, b and f, respectively. The Akaike Information Criterion is employed to guide the selection of an appropriate order (p,q) of an ARMA process for a^ t . Our results (not reported) suggest that a^ t may be adequately described by an ARMA(1,5) model. The same order (1,5) is selected for the state equation (A.4) in our empirical state-space representation of TFP. Furthermore, the estimated coefﬁcients from the ARMA equation for a^ t are used in the Kalman-ﬁltering algorithm as initial guesses for d and gi (i ¼ 1, y, 5) in equation (A.4). There is no commonly accepted procedure to determine an initial guess for the variance–covariance matrix H. We stipulate that the estimated square standard error (SSE) from the ARMA(1,5) regression equation (A.4) for the aˆt process is a reasonable initial value for s2. The SSE estimate is used to construct the initial matrix H. Employing the initial guesses, the estimation of the APF parameters and unobserved state variable (at) is carried out using the standard KF by means of the maximum likelihood (ML) method.27 Convergence is achieved smoothly after 13 iterations. The estimates of the parameters, their t-statistics and the loglikelihood function (Ln F) are displayed in Table A.1. The parameter estimates, exhibited in Table A.1, are used to provide estimates for TFP during the period under consideration. On observing the pattern of signiﬁcant coefﬁcients, the table delineates that – with the exception of g4 – all the parameter estimates of the production function and

27

The Broyden, Fletcher, Goldfarb and Shanno (BFGS) algorithm (Press et al., 1988) is used in ML estimation.

Parameter b a

KF system, OLS, ML and ARMA in the errors estimated parameter values of the per-worker aggregate production function OLS

MLa

Systema

0.008 (1.411) 0.606 (6.815)

0.008 (1.900) 0.606 (13.624)

2.058 (4.121) 0.649 (4.223) 0.991 (74.270) 0.605 (3.261) 0.344 (1.589)

d g1 g2 R2 (%)

56.33

ARMAeb,c

Parameter

MLa

g3 0.549 (3.758) 0.508 3.402 – – 0.202 (1.263) 43.26

0.327 (1.692) 0.166 (1.120) 0.289 (1.655) 0.042 (1.883)

g4 g5 f s2 Ln F

Systema

0.001 (4.116) 122.363

ARMAeb,c 0.153 1.011 0.437 3.716 0.705 (5.480) 0.039 (1.793)

137.924

Note: (t-Statistics in parentheses.) BFGS algorithm employed in estimation. b Gauss–Newton algorithm employed in estimation. c For succinctness, the state-space parameter notation, a, d, g1yg5 and f is borrowed to denote the ARMAe parameters aARMA, dARMA, gARMA1ygARMA5 and fARMA, respectively. a

Sources of Economic Growth and Technical Progress in Egypt

Table A.1.

231

232

Hanaa Kheir-El-Din and Tarek Abdelfattah Moursi

the state equation are signiﬁcant at conventional levels. In particular, the coefﬁcient estimates of the APF and the autoregression (AR) coefﬁcient (d) of the state equation are of high quality according to the reported t-statistics. The estimated value of d is notably near unity, suggesting that the state variable TFP process is characterized by a substantial degree of persistence. This ﬁnding provides some statistical support for the assumption maintained in the construction of the state equation that TFP exhibits a high degree of persistence over time. The estimate of the share of capital in income reported in Table A.1, roughly 0.65, is above the commonly assumed value, 0.4, for growth accounting exercises and other studies on the Egyptian economy (see for example, World Bank, 1994 and Pissarides, 2000). Our result, however, is consistent with recent international evidence as well as with other ﬁndings for the Middle East and North Africa (MENA) region, which suggest that the capital share in income is considerably higher than 0.5 (Bisat et al., 1997). Abdellatif’s (2003) study of the industrial sector in Egypt provides further support for our ﬁnding. Using micro-data for the industrial sector, she shows that the share of capital in industrial output (value added) is on average – for different plant sizes – greater than 0.65. Table A.1 also contrasts our KF system estimates of the parameters of the APF with standard (single-equation) estimates of a and b, obtained directly from equation A.328 The latter estimates are derived by means of the OLS and the ML methods. As expected, while both OLS and ML yield the same estimates for a and b, the ML estimates tend to be relatively more efﬁcient as indicated by the tstatistics reported in Table A.1. Moreover, we ﬁnd that the value of a estimated from the single-equation regression (0.606) is quite close to the analogous estimate resulting from the KF system (0.649). However, a comparison between the computed log-likelihood functions, exhibited in Table A.1, provides evidence conﬁrming that the multi-equation system outperforms the single-equation regression. The only qualm associated with this result is that – contrary to the estimate of b – the single-equation ML (as well as the OLS) estimate of a is somewhat more efﬁcient than the multi-equation system estimate. This should not be too alarming for two reasons. First, relative efﬁciency of the single-equation estimate of a is most likely the spurious result of omitting some of our a priori knowledge about the dynamic behavior of a. Second, the short sample period (1960–1998) involved in estimation is not sufﬁcient to allow us to attach large signiﬁcance to differences between the estimates of the t-statistics obtained from the single- and multiple-equation models. Figures A.1 and A.2 sketch the TFP and TFPG series, respectively, calculated on basis of the different single- and multi-equation system parameter estimates. In addition, the two series are recalculated for an ad hoc value for a equal to 0.4,

28

The single-equation APF is estimated without the dummy variable D91. Introducing the dummy variable does not have any qualitative effect on any of the results.

Sources of Economic Growth and Technical Progress in Egypt

Figure A.1.

233

Sensitivity of TFP to the value of a

1 0.9 0.8 System

0.7

a

0.6 0.5 α=0.4

0.4 0.3

α=0.606

0.2 0.1 0 60

65

70

75

80

85

90

95

Year

which has been adopted frequently in the earlier studies (World Bank, 1994). Visual inspection reveals that the TFP level is sensitive to the selected value of a.29 In general, the TFP level series computed with the system and the ML/OLS estimates of a are more or less comparable. Figure A.2, however, shows that the single-equation model could not distinctly resolve variances in TFPG associated with values of a ranging between 0.4 and 0.6. In contrast, the TFPG series, based on the richer dynamic behavior of the underlying system estimates, appears to capture variations in technical progress associated with different values of a as well as some of the distinctive realities of Egyptian economic performance such as the rise in TFPG during the 1990s. Alternatively, the system results reported in Table A.1 suggest the presence of a unit root in a, which is indicated by the parameter estimate of d (0.991). This may affect our inference based on the KF approach. We notice that the KF approach that we use in this paper is equivalent to ﬁtting equation A.3 with ARMA speciﬁcation in the errors.30 The latter approach has the

29

The nonmonotonic sensitivity of (log) TFP estimates to the value of a disclosed by Figure A.1 follows from the difference in the functional speciﬁcation of the single-equation (OLS/ML) and of the multiple-equation (system) models (see Table A.1) used to derive those estimates. 30 We thank Hashem Pesaran and Jeffrey Nugent for pointing out that equivalence and for suggesting the procedure for comparing our KF estimates with the ARMA in the errors estimates of the parameters of the APF.

Hanaa Kheir-El-Din and Tarek Abdelfattah Moursi

234

Figure A.2.

Sensitivity of TFPG to the value of a α=0.4

0.1 0.08

System 0.06 0.04 0.02 0 -0.02

α=0.606 60

65

70

75

80 Year

85

90

95

advantage that it can be easily adapted to deal with the possibility of a unit root in TFP. To compare the results from the two approaches, we employ the ARMA speciﬁcation in the errors (ARMAe) to estimate the coefﬁcients of the APF. We begin by taking the ﬁrst difference of both sides of equation A.3. Then we estimate the regression of dy on dk without an intercept, but inclusive of the 1991 dummy variable in ﬁrst difference form (dD91) dyt ¼ a_ ARMAt þ aARMA dkt þ fARMA dD91

ðA:6Þ

_ which measure We assume, also, that the residuals from the regression A.6 (a), the TFP growth, are allowed to have an ARMA(p,q) speciﬁcation a_ ARMAt ¼

p X i¼1

dARMAi a_ ARMAt i þ ARMAt þ

q X

gARMAj ARMAt j

ðA:7Þ

j¼1

Similar to the empirical KF system, we choose the order (p,q) ¼ (1,5) to describe the ARMAe process and we invoke the Gauss–Newton procedure to estimate the parameters of equations A.6 and A.7. Despite many tedious trials with different initial guesses, the selected ARMAe (1,5) model failed to converge after a reasonable number of iterations. Repeated attempts, however, have suggested that the convergence properties of the model are signiﬁcantly enhanced when the ﬁrst coefﬁcient of the MA process (gARMA1) is restricted to 0. We impose that restriction directly by excluding the MA(1) process from the estimation of the empirical analog of equation A.7. Following that strategy, the estimated ARMAe (1,2-5) model converges smoothly after 21

Sources of Economic Growth and Technical Progress in Egypt

Figure A.3.

235

TFP estimates from the KF system and the ARMA specification in the errors model

2

Period III

Period II 1.5 aARMA

a

Period I 1

0.5 a

0 65

70

75

80

85

90

95

Year

iterations. The parameter estimates obtained from this model are exhibited in Table A.1 along with the corresponding t-statistics and the R2 (percent) for the regression. Table A.1 shows that the system and the ARMAe estimates are qualitatively equivalent. (In particular, the estimate of the share of capital in income (0.55) is still above the commonly assumed value, 0.4, and is relatively close to the value estimated from the KF system, 0.65.) To be sure, we compare the TFPG and the TFP estimates derived from the state-space system (a_ and a) with the corresponding estimates (a_ ARMA and aARMA) resulting from the ARMAe model. The aARMA series are obtained by accumulating the residuals from equation A-6 (a_ ARMA ) from some given value (say 0) to compute estimates of TFP (aARMA) up to a scalar. The two sets of TFP and TFPG series are depicted in Figures A.3 and A.4, respectively. Visual inspection of the two diagrams substantiates the similar behavior of the TFP and of the TFPG series (particularly their respective turning points) estimated by means of the two approaches. This result is further conﬁrmed by a set of Ljung-Box Q-statistic tests (not reported).

Appendix B: Source of data (A) World Development Indicators (2000). GDP at market prices [GDPMP]; real GDP at constant 1987 prices [GDP (Y)]; total labor force; inﬂation [INFLAT]; general government consumption [GOV]; gross domestic ﬁxed

Hanaa Kheir-El-Din and Tarek Abdelfattah Moursi

236

Figure A.4. TFP growth estimates from the KF system and the ARMA specification in the errors model

0.2

Period III

Period II

Period I

0.15

a⋅ ARMA

a⋅

0.1 0.05 0 a⋅

-0.05 -0.1 65

70

75

80

85

90

95

Year

investment [GDFI]; credit to private sector [CRPRIV]; exports [EXP] and imports [IMP] of goods and services; foreign aid [AID] and population [POP]. (B) Other sources. Data on unemployment rates are obtained from Moursi and Kamel (1989) and CAPMAS (various issues); real capital stock at constant 1987 prices (LE) [k87] data for 1960–1990 from Nehru and Dhareshwar (1994) and data after 1990 are updated using real ﬁxed investment at constant 1987 prices [FI87] (World Bank, 2000) and assuming a depreciation of 4 percent per annum as follows: kðt þ 1Þ ¼ FI 87 ðtÞ þ ð1 0:04Þk87 ðt 1Þ; t ¼ 1991; . . . ; 1997.

CHAPTER 8

Sources of Growth and the Output Gap for the Turkish Economy Kıvılcım Metin O¨zcan, U¨mit O¨zlale and C - ag˘rı Sarıkaya Abstract In this study, we analyze the growth dynamics of the Turkish economy. After investigating the sources of growth and comparing both the output and the inflation performance with the other MENA countries, we find that the Turkish economy has not achieved a sustainable growth path and has a much worse inflation performance. Then, we take a closer look at the volatile output performance and discuss the role of the wrong macroeconomic policies on the unsustainable output performance. Utilizing an output gap measure, we find that the short-term capital flows seem to be the main driving force in generating excessive fluctuations in the output. Based on these findings, we can say that the high economic growth rates, which the Turkish economy has achieved in the recent years owe much to massive short-term capital inflows. At this stage, it is too early to evaluate the effects of the recent structural reforms on the long-term output dynamics.

Keywords: Turkish economy, emerging markets, output gap, economic growth JEL classifications: C1, E3, O4

Corresponding author. CONTRIBUTIONS TO ECONOMIC ANALYSIS VOLUME 278 ISSN: 0573-8555 DOI:10.1016/S0573-8555(06)78008-7

r 2007 ELSEVIER B.V. ALL RIGHTS RESERVED

238

Kıvılcım Metin O¨zcan, U¨mit O¨zlale and C - ag˘rı Sarıkaya

1. Introduction Despite several stabilization and structural adjustment programs, all of which were backed by the World Bank and the International Monetary Fund (IMF), the Turkish economy can be characterized as having an unsustainable output growth and an inﬂationary environment over the last four decades. Similar to other emerging market economies, a boom–bust cycle on the output dynamics can be observed. After the collapse of the exchange rate-based stabilization program in 2001, which was followed by the deepest ﬁnancial crisis that Turkey has ever experienced, the economy has started to recover with very high economic growth rates. Moreover, unlike the previous recovery periods, the price stability objective of the monetary authorities seems to have been achieved, with inﬂation rates being under 10% as of mid-2005. However, it is an open question whether these recent positive developments after the crisis stem from a major structural change in the macroeconomic dynamics or simply reﬂect the expansionary phase of the boom–bust cycle. On the one hand, with the IMF’s supervision, the government and the central bank took important steps to achieve ﬁscal discipline and low levels of inﬂation, which are the two necessary preconditions for a sustained long-run growth performance. While the government followed a tight ﬁscal policy to put the debt stock on a sustainable path, the central bank started to implement a form of inﬂation targeting, where it is clearly stated that the main objective of the monetary policy is price stability. On the other hand, it can be argued that the still-high real interest rates and the increasing appetite of the investors to invest in the emerging market economies have induced massive short-term capital inﬂows, thus helping the Turkish economy to perform at high economic growth rates without any pressure on price dynamics. As the recent experiences in the emerging markets suggest, high growth rates that are mainly driven by short-term capital ﬂows are not sustainable and generally lead to excessive ﬂuctuations in the macroeconomic dynamics in the long run. Moreover, with short-term capital inﬂows, pressure on exchange rates is relieved, which clearly affects inﬂation performance positively both through the lower cost of imported intermediate inputs and the prices of the imported ﬁnal goods. However, as soon as the concerns regarding the ﬁnancing of the current account deﬁcit increase, a possible adjustment in the exchange rates has the potential to disrupt the price dynamics and generate lower economic growth. As a result, it is not clear whether the recent positive developments in the Turkish economy imply a sustainable long-run growth performance or simply signal a temporary improvement. To address the above-mentioned issue, this study analyzes the output dynamics of the Turkish economy. First, beginning with the 1960s, we present an overview of the economy from a growth perspective. Then, we provide a brief comparison of the Turkish economy with the other MENA countries. The main ﬁnding in these two sections is that the Turkish economy has been far from

Sources of Growth and the Output Gap for the Turkish Economy

239

having a sustainable growth path especially during the last two decades. Therefore, following the comparison of the Turkish economy with the other MENA countries, we discuss the negative outcomes of the macroeconomic policies implemented over the last two decades, with special reference to the ﬁnancial crises in 1994 and 2001. One main result of this discussion is that the short-term capital ﬂows are becoming increasingly important in shaping macroeconomic dynamics. Thus, we continue our analysis with an empirical investigation of the inﬂuence of short-term capital ﬂows on the excessive ﬂuctuations in the output. In this respect, we utilize an output gap measure and show its comovement with the capital inﬂows. Using this gap measure, we also display the business cycle characteristics of the Turkish economy. Such an exercise will not only enable us to develop an idea about the recent macroeconomic framework, but it will also reveal which macroeconomic variables are procyclical or countercyclical as well as whether these variables lead or lag behind the ﬂuctuations in the output. Section 6 concludes the discussion. 2. An overview of the Turkish economy from a growth perspective This section provides a brief historical overview of the Turkish economy from a growth perspective by identifying periods by stabilization policies and the growth path followed.1 In this section, we will mainly focus on capital formation and the sources of growth. The evaluation of the Turkish economy is analyzed by dividing the entire period into three phases: 1963–1980, 1980–1999 and from the millennium onward. 2.1. Inward-oriented strategy: 1963– 1980 period While there was no conscious planning or coordination of economic policies in the 1950s, a new constitution was formulated under the revolutionary government in 1960–1961. One of its key changes was social and economic development planning. The role of the state gained importance by following the import substitution policies together with economy-wide planning. During the First and Second Five-Year Plans, the traditional strategy of modernization and industrialization in a mixed economy was attempted systematically. In both plans, the main aim was to achieve a higher rate of growth and investment in a balanced economy. Some social issues addressed in the plans were more equitable income distribution, better regional balance, land reform, economic organization and

1

See, for example, Aricanli and Rodrik (1990), Celasun and Rodrik (1989), Ekinci (1990, 2000), Metin-Ozcan et al., (2001) and O¨zatay (1999) among others for more detailed analyses and discussions of these and related issues of the Turkish economy.

240

Kıvılcım Metin O¨zcan, U¨mit O¨zlale and C - ag˘rı Sarıkaya

the efﬁciency of the State Economic Enterprises (SEEs) (for more detail, see Celasun and Rodrik (1989), S- enses (1991) and Metin (1995, pp. 17–18)). The pace of economic growth accelerated during the 1960s and most of the 1970s was also good. In fact, the GDP increased at an average annual rate of 6.4% during the First Plan period (1963–1967), 6.7% during the Second Plan period (1968–1972) and 7.2% during the Third Plan period (1973–1977). In the 1960s, development occurred with moderate reliance on foreign aid. In the 1970s, it was supported by large increases in workers’ remittances, but after 1974 by excessive reliance on foreign borrowing. The long-term growth of industry has been particularly rapid, averaging 9.2% between 1963 and 1977. This led to a substantial structural change in the economy. Agricultural output growth averaged 3.3% per annum during this period, somewhat above the annual population growth rate of 2.5%. Population growth remained fairly high (2.5% per annum). Since GDP growth was fairly high, this allowed a relatively fast growth in per capita income. However, the rising income level does not appear to have been accompanied by a better income distribution. In fact, the 1973 household income survey showed that the income was unequally distributed. The top 10% of the population received at least 40% of the income while the bottom 40% received about 11.5%. Considerable inequality within the agricultural sector (owing largely to extremely unequal land distribution) was the main cause of the overall income inequality (World Bank Country Study (1980, p. 11)) During the 1963–1977 period, both real private investment and real public investment increased at almost the same pace, averaging 10.8% and 11.4% per annum, respectively. This revealed the crowding-in effect in both sectors, which is in line with inward-oriented strategy (see I˙smihan (2003, p. 144). From 1973 to 1979, public sector investment increased tremendously with a 22.1% annual rise, while private investment grew by 10.2%. During the 1960s, the macroeconomic environment was stable compared to that in the 1970s. In fact, the inﬂation rate, which is regarded as an important indicator of macroeconomic instability, was quite low during the 1960s (5.2% per annum) compared to that in the 1970s (27% per annum). Starting in the mid-1970s, the economy become even more unsteady since Turkey had delayed its internal adjustment to the external shocks of this period via reserve decumulation initially and excessive short-term borrowing later on (Celasun and Rodrik, 1989). In addition to macroeconomic instability, from the mid-1977 onward, the political environment became more unstable owing to successive weak coalition governments. In 1978–1979, the crisis in the economy deepened on account of inﬂation and balance of payment difﬁculties, so the government needed to implement more effective policies. On January 24, 1980, the government announced a major stabilization program, which included a series of new economic measures that were intended to solve the high inﬂation, economic stagnation and unmanagable balance of payment problems.

Sources of Growth and the Output Gap for the Turkish Economy

241

Table 1. Selected indicators of the Turkish economy (1980–1999) Output and capital formation Annual average growth rate Real GNP Real private ﬁxed investment Real public ﬁxed investment Real public ﬁxed core infrastructural investment Real public ﬁxed non-core infrastructural investment Composition of public investmenta Core infrastructural investment (as % of total) Non-core infrastructural investment (as % of total) Macroeconomic instabilitya Inﬂation rateb(%)

4.2 6.1 1.6 2.7 0.8 50.5 49.5 61.8

Source: I˙smihan et al. (2005), Table 1, p. 241. a Simple period average. b Percentage change in the GNP deﬂator.

2.2. The 1980s and the 1990s Turkey began taking serious steps to liberalize and strengthen its economy a full 25 years ago. In 1980, Turkey made a fundamental decision to switch its economic strategy from an inward-oriented growth to an outward-oriented one. Prior to the reform program, tariff barriers were high, state ownership prevailed in key sectors and competition was strangled by regulations. The 1980 program had stabilization as well as structural aspects with efforts to liberalize trade and ﬁnance. During the outward-oriented period of 1980–1999, the real GNP grew at an average annual rate of 4.2% (see Table 1 which provides summary information on the Turkish economy for the 1980-1999 period).2 The economic growth rate was higher during the 1980s (5.2% per year) compared to that in the 1990s (3.2% per year). Capital formation performance in the private sector was better compared to that of the public sector during this period. Real private (public) investment grew at an average annual rate of 6.1% (1.6%) from 1980 to 1999. The most important development in this period was the changing role of the state in the investment process. The share of core public infrastructural investment in total public investment rose from 37.3% in the inward-oriented period to 50.5% in the outward-oriented one. While the private investment to GNP ratio (in current prices) rose from 12.8% in the 1980s to 18.1% in the 1990s, the public investment/GNP ratio dropped from 8.8% in the 1980s to 6.2% in the 1990s owing to budgetary pressures.3

2

For further discussion, see I˙smihan et al. (2005, p. 241). See Conway (1990), Uygur (1995) and I˙smihan et al. (2005, p. 242) for a detailed discussion of private and public investment behavior in Turkey.

3

242

Kıvılcım Metin O¨zcan, U¨mit O¨zlale and C - ag˘rı Sarıkaya

Turkey experienced a very severe ﬁnancial crisis in early 1994 mainly owing to unsustainable ﬁscal balance and monetization of the domestic debt. GNP contracted by 6.1% during 1994, which is the peak rate of contraction of the Turkish economy during the 1963–1999 period. Similarly, real public investment fell dramatically by about 40% from 1993 to 1994. This is evidence of the negative effect of macroeconomic instability on the ﬁscal ‘‘ability’’ of governments to make investments. Real private investment, in contrast, contracted only moderately (about 5%). Inﬂation peaked in 1994 with 107.3% per annum and the Turkish lira depreciated by>70% against the US$ in 1994 (I˙smihan et al., 2005, p. 242). Following this major crisis in mid-1994, Turkey adopted an IMF-based standby agreement to eliminate the severe ﬁnancial crisis. The inﬂation rate decreased from 107.3% in 1994 to 87.2% in 1995. However, macroeconomic instability continued until the late 1990s, and public sector balances were unsustainable owing to heavy reliance on domestic borrowing. In addition, the East Asian and Russian crises of 1997–1998 and the two devastating earthquakes of 1999 had a negative impact on the Turkish economy. In December 1999, Turkey signed a 3year IMF-based stand-by agreement.4 The objectives of the program were tight ﬁscal and monetary policies, ambitious structural reforms and the use of a predetermined exchange rate path as a nominal anchor. The disinﬂation program had a major impact on banks’ balance sheets. At the end of 2000, the Turkish banking system was seriously affected by increasing interest, exchange rate and credit risks because of the short-term maturity structure of ﬁnancial sources used in the banking sector and the open positions of the Turkish banks. Owing to a major banking crisis, unfortunately, this program failed in the early 2001 and the real GNP contracted by 9.4% in 2001. Then, backed by the IMF and the World Bank, Turkey signed another program, called ‘‘Transition to a Strong Economy,’’ to eliminate the conﬁdence crisis and ﬁnancial instability. The strategy was strongly based on market orientation and openness to the world economy. An important pillar of the program consisted of a renewed effort to eliminate structural weaknesses that had not been fully tackled by the 2000 program, particularly by strengthening governance and improving economic management.

2.3. The millennium onward The trend in economic growth during the period from 1998, the year prior to the economic crisis that affected the early 2000s, to the end of 2004 may be analyzed in two subperiods: 1999–2000 and 2001–2004. In the ﬁrst subperiod, the average annual increase in the GDP was around 1.1% and in the second subperiod 3.7%. If the 1999–2004 period is taken as a whole, the rates of growth of the GNP and GDP are 2.9% and 2.3%, respectively. In spite of the improvement throughout

4

See Ekinci (2000) for the details of program.

Sources of Growth and the Output Gap for the Turkish Economy

243

the 2002–2004 period, the negative effects of the two banking crises experienced in 1999 and 2001 were still being felt (see Figure 1 to observe the GDP growth rates). A high GDP growth rate of 7% was achieved in 2002, followed by 4.8% in 2003 and 9% in 2004. The Turkish economy displayed an especially high growth performance in 2004. After the 2001 banking crisis, inﬂation had been on a steady downward trend and came close to the single-digit level as of 2004. Starting from the ﬁrst quarter of 2002, high growth ﬁgures were achieved without interruption over a period of 10 quarters. Therefore, there are indications that the Turkish economy has been free of the structure, which is characterized by boom–bust cycles since the 1990s and has embarked on a high-growth process (Pre-Accession Economic Programme, State Planning Organization (SPO) (2004, p. 27)). At this point, it is necessary to discuss whether these positive developments caused expectations to change permanently. The course of Turkish economy has clearly seen improvement in expectations since 2001 (see Figure 2 below). The upper-left panel of the ﬁgure indicates that business conﬁdence has been stable and dominated by optimists since the beginning of 2002. Moreover, the consecutive achievements in reaching the year-end inﬂation targets since 2002 have increased the credibility of macroeconomic policies and have helped inﬂation expectations maintain a downward trend since then. In line with the robust economic growth and disinﬂation process, improved economic fundamentals have led to a more conducive environment for employment and investment decisions (right panel of the ﬁgure). However, although business tendency survey results point to a signiﬁcant improvement in the expectations, it is too early at this stage to claim that the changes in these expectations are permanent. Private ﬁxed capital investment emerged as the other element that drove growth in 2004 (see Independent Social Scientists’ Alliance (ISSA) Report, 2005 and Pre-Accession Economic Programme, State Planning Organization (SPO) (2004) for further discussion). Figure 1.

GDP growth rates GDP Growth Rate

0.1 0.08 0.06 0.04 0.02

-0.06 -0.08

20 02

20 00

19 98

19 96

19 94

19 92

19 90

19 88

19 86

-0.04

19 84

19 80

-0.02

19 82

0

Kıvılcım Metin O¨zcan, U¨mit O¨zlale and C - ag˘rı Sarıkaya

244

Figure 2.

Selected indicators from the CBRT business tendency survey

50

20

30

10 0

10 % -10

%

-10 -20

-30

Business Confidence (Optimists-Pessimists)

Jul-05

Jul-04

Jan-05

Jul-03

Jan-04

Jul-02

Jan-03

Jul-01

Jan-02

Jul-00

Total Employment Tendency (More-Less)

70

PPI Inflation Expectations for the Next Twelve Months

Jul-05

Jul-04

Jan-05

Jul-03

Jan-04

Jul-02

Jan-03

Jul-01

Jul-05

Jan-05

Jul-04

Jan-04

Jul-03

Jan-03

Jul-02

Jan-02

Jul-01

Jan-01

Jul-00

Jan-00

Jul-99

Jan-99

0

Jan-02

10

Jul-00

20

Jan-01

30

Jan-00

40

Jul-99

50

Jan-99

10 0 -10 -20 -30 % -40 -50 -60 -70 -80

60

%

Jan-01

Jul-99

Jan-99

Jul-05

Jul-04

Jan-05

Jul-03

Jan-04

Jul-02

Jan-03

Jul-01

Jan-02

Jul-00

Jan-01

Jul-99

Jan-00

-50 Jan-99

-40

-70

Jan-00

-30

-50

Investment Tendency for the Next Twelve Months (More-Less)

When the total savings–investment balance, which is given in Table 2 is considered, the ﬁrst point to be noted is that some revival in private sector ﬁxed capital investment has been observed (13.7%) in 2004. However, this ﬁgure is still far behind the averages of the private ﬁxed investment to GDP ratios of 1990–1994 (17.9%) and 1995–1999 (17.4%) periods. It is also observed that this revival is largely ﬁnanced by increases in external resources (4.6% in 2004) and that the domestic savings/GDP ratio, which was showing an increasing trend since the 2000s remained at the same level as during 1990–1994. The second point to be raised is that the ratio of private investment to GDP was realized at 11.3% in 2002 and maintained at this level in 2003 along with a signiﬁcant increase recorded in private investment in 2004 (13.7%). The third point is that the ratio of public investments to GDP, which was around 5%–6% during the 1990s (see Table 2), decreased to as low as to 4.0% in 2004 as a result of the tight ﬁscal policy implemented. In addition, stocks are rising and investments, especially public investments are steadily declining. In fact, there is no possible economic explanation of the observation that stock increases have become a considerable determining factor in economic growth. However, Independent Social Scientists’ Alliance (ISSA) Report (2005, p. 11) put forth two reasons, which shed light on the phenomenon. First, it seems that the fact that

Sources of Growth and the Output Gap for the Turkish Economy

Table 2.

245

Total savings–investment balance (as percentage of GDP)

Year Total investment Fixed capital investment Public investment Private investment Stock changes Total savings Domestic savings Foreign savings

1990–1994

1995–1999

2001

2002

2003

2004

23.0 24.2 6.3 17.9 –1.2 23.1 21.3 1.8

23.2 22.9 5.5 17.4 0.3 23.3 19.2 4.1

15.9 17.2 5.3 11.9 –1.3 15.9 18.2 –2.3

21.3 16.6 5.3 11.3 4.7 21.3 19.8 1.5

22.8 15.5 4.2 11.3 7.3 22.8 19.5 3.3

25.9 17.7 4.0 13.7 8.2 25.9 21.3 4.6

Note: The savings–investment balance is based on State Institute of Statistics (SIS) ﬁgures of the GDP by expenditures. Public investments in the table include the investments of the SEEs in addition to those of the general government.Source: Pre-Accession Economic Programme (2004, p. 31, Table 2.1).

the economy has recently become more dependent on imported inputs and consequent decreases in value added/gross production value ratios in all sectors including industry in the ﬁrst place is not duly considered by the SIS, which still uses the technical coefﬁcients of the 1990s. Hence, when high estimations are made for GDP growth (based on gross production values), ﬁnal demands, which cannot be explained by changes in investment, consumption and net exports turn into virtual demand elements and reﬂect themselves in stock changes as a residual (see also Somc- ag˘ (2004)). A less likely second possibility is the underestimation of private consumption expenditures, which are not as closely followed as investments by the State Planning Organization (SPO), and carrying the impact of this faulty estimation to stock changes (see Independent Social Scientists’ Alliance (ISSA) Report (2005, p. 11)). Finally, the total domestic savings had an upward trend after 2000. The large increases of domestic demand that were observed in 2003 and 2004 have increased the economy’s need for foreign savings. The most unpleasant result of the high growth rate was its failure to generate employment. Throughout 2004, unemployment remained a problem of the economy awaiting solution (see State Institute of Statistics (SIS), 2004). An important observation relating to labor market data for the period following 2002 is the presence of sharp increases in the national income together with overly slow expansion in employment, which is known as jobless growth in the literature (see UNCTAD, 2002, 2003).5 Therefore, while the contribution of employment to growth was negative owing to the decline in employment, the contribution of

5

Turkey’s jobless growth starting from the late 1990s and the early 2000s is believed partly to have been due to tightening labor market regulations so as to be more like the EU. The labor market will be one of the key factors in ensuring a sustainable growth environment and enhancing the competitiveness of the economy during the process of EU accession (see further discussion, Pre-Accession Economic Programme, State Planning Organization (SPO) (2004, p. 80).

Kıvılcım Metin O¨zcan, U¨mit O¨zlale and C - ag˘rı Sarıkaya

246

capital stock and total factor productivity (TFP) to growth increased considerably, reaching 52.2% and 58.5%, respectively, in 2003 and 2004 (Pre-Accession Economic Programme, State Planning Organization (SPO) (2004, p. 31). After these characteristics of the Turkish economy are presented, it will be insightful to compare the overall performance of the Turkish economy with the MENA countries. The next section provides such a comparison. 3. Comparison with the MENA countries In this section, ﬁrst, the economic growth and inﬂation performance of the Turkish economy will be compared to those of some MENA countries. Second, using the results of the previous studies that employed a growth accounting method decomposing aggregate GDP growth into labor, capital and technical progress for the MENA countries, these decomposed variables are compared with the ones obtained from the Turkish economy. Interesting results are obtained when the inﬂation and output performance of the Turkish economy are compared with those of the non-oil MENA countries (see Table 3). Excluding Turkey, the countries in the non-oil MENA region have performed remarkably well in terms of inﬂation. As Table 3 shows, the average annual inﬂation rate for the region is 3.4% when the Turkish economy is not taken into consideration. The fact that the Turkish economy had a 56% average inﬂation rate for the sample period clearly shows why it should be treated as an outlier in its inﬂation performance. When the standard deviation of the inﬂation, which can be regarded as a source of volatility, is calculated, it is seen that the inﬂation rate is also volatile. Turkish output growth was also very volatile compared to the other non-oil MENA countries during this period of>25 years. However, despite the good inﬂation performance over the last two decades, the economic growth of the MENA region has been very slow, as Table 4 indicates. Although the region needed to grow to at least 5%–6% to reduce high-level Table 3.

Inflation and output performances for the period 1980–2003: Turkey and non-oil MENA countries

Country Turkey Non-oil MENA Egypt Jordan Morocco Tunisia Syria

Inﬂation rate

Standard Deviation

Output growth rate

Standard Deviation

56.9 3.40 4.40 3.40 1.97 2.50 4.6

22.7 0.029 0.028 0.023 0.020 0.016 0.054

4.08 4.56 6.2

4.78 0.025 0.024

a

3.3 4.1 5.8

0.024 0.015 0.039

Note: Non-oil MENA countries comprises Egypt, Jordan, Morocco, Syria and Tunisia. Turkey is not included in the calculations.Source: Our calculations are based on the IFS CDROM data. a We could not ﬁnd the data in the IFS CDROM.

Sources of Growth and the Output Gap for the Turkish Economy

247

Table 4. GDP and population growth in Turkey and the MENA countries Country

Periods

Turkey

1980–1990a 1990–1998a 1980–2004a 1980–1990a 1990–1998a 1980–2003c

MENAb

GDP Growth

Population Growth

4.5a 5.0a 4.0a 2.0c 3.0c 4.5a

2.6a 1.7a 2.4a 3.2c 2.3c 1.1a

Source: World Bank Development Indicators, CDROM, 1999, World Development Report, 1983, 1989, 1999/2000. a Own calculations. b It includes non-oil MENA countries, which are Egypt, Jordan, Morocco, Tunisia and Syria. c Sekkat (2003), Table 1.

unemployment, the average growth rate was only 2% for 1980–1990 and 3% for 1990–1998. Moreover, during the 1981–90, 1991–98 and 1980–2000 periods, the real GDP per capita growth in the region was 0.4%, 0.2% and -0.07%, respectively (see Sekkat (2003) Table 1, p. 5). The Turkish economy, in contrast, grew to an average of 5.4% during the 1980–1990 period. However, mostly owing to the ﬁnancial crisis in 1994 and the Russian crisis in 1998, the economy grew at a slower pace between 1990 and 1998. Finally, the major economic crisis in 2001 led to even less economic growth rates between 2001 and 2004. These ﬁndings clearly show that the Turkish economy suffered high and volatile inﬂation. The high economic growth rates were mostly achieved by stimulating the aggregate demand, which did not put the economy on a sustainable growth track and resulted in an inﬂationary environment. In Table 5, we investigate the growth rates in the factors of production across subperiods for the non-oil MENA countries and Turkey. It is evident from Table 5 that the Turkish economy suffered a sharper decrease in growth rate during 2001–2003 and exhibited more volatility compared to the non-oil MENA countries. Another important ﬁnding is that the growth rate of labor turned out to be negative after the major economic crisis in February 2001. Unemployment rose and the contribution of labor to growth was negative. In Table 6, by utilizing a growth accounting framework6 we further elaborate the differences in growth dynamics across the MENA7 countries and Turkey.

6

The growth accounting method measures the contributions of the growth in capital per-worker and TFP to the growth of the real output per-worker. The measurement of TFP is the simplest econometric method and it is commonly used to evaluate past and future performance of growth in an economy. During the calculation of TFP growth, a two-factor growth accounting framework is usually used with labor and quantity of capital. 7 For the growth accounting exercise for the MENA countries, see Makdisi et al. (2000), Senhadji (2000), Sekkat (2003), Bosworth and Collins (2003) and Hakura (2004).

248

Table 5.

Kıvılcım Metin O¨zcan, U¨mit O¨zlale and C - ag˘rı Sarıkaya

Growth rates of factors of production: Turkey and the non-Oil MENA countries Growth Rate

Country

Period

Turkey

1960–1980a 1981–1990a 1991–1997a 1990–2000b 2001–2003b 2003b 1960–1980 1981–1990 1991–1997 1960–1980 1981–1990 1991–1997 1960–1980 1981–1990 1991–1997 1960–1980 1981–1990 1991–1997

Egypta

Jordana

Moroccoa

Tunisiaa

a

Output

Capital

Labor

5.18 5.35 3.96 4.10 1.80 5.80 6.77 4.93 4.68 5.76 3.19 4.18 5.87 3.57 2.36 6.61 3.72 4.30

7.60 4.63 5.48 7.60 4.30 4.00 7.95 8.75 2.28 10.95 6.99 8.20 8.25 4.95 3.59 6.48 4.60 3.92

1.42 2.59 3.14 1.50 –0.7 –1.0 2.20 2.52 2.84 3.22 5.04 6.00 2.71 2.46 2.61 2.84 2.78 2.96

Sekkat (2003) and FEMISE Report on the Euro Mediterranean Partnership (2003 p. 20, Box 3). Pre-Accession Economic Programme State Planning Organization (SPO) (2004), p. 31, Table 2.2.

b

Despite the considerable literature that has employed such a framework, there are only a limited number of studies, which concentrate on the Turkish case.8 Although Turkey has been characterized by rapid growth in human capital, to the best of our knowledge there is no TFP calculation, which considers human capital accumulation.9 Recently, the SPO, which is essentially based on the work of Saygili et al. (2005), calculated growth accounts for Turkey (1990–2003) in the Pre-Accession Economic Programme, State Planning Organization (SPO) (2004). However, the role of human capital accumulation is neglected in this analysis. Recently, in their growth accounting exercise, Ismihan and Metin-O¨zcan, (2005, p. 16) calculated a human capital-augmented labor series by using the average years of schooling of adult population (aged 15 years and over) and the rate of return to schooling, which is approximately 10%.

See Maras-liog˘lu and Tiktik (1991), Temel and Saygili (1995), Saygili (1999), Temel et al. (2000), Bosworth and Collins (2003), Saygili et al. (2005), Ismihan and Metin-O¨zcan, (2005, p. 12–15) on growth, capital accumulation and the employment level of the Turkish economy. 9 TFP growth is calculated as a ‘‘residual’’ from growth accounting formula and TFP growth can be thought of as a measure of gains (or losses) in economic efﬁciency. 8

Sources of Growth and the Output Gap for the Turkish Economy

Table 6.

249

Contributions to economic growth by factors of production Contribution to Growth

Country

Period

Capital (%)

Labor (%)

TFP (%)

Turkey

1960–1980a 1981–1990a 1991–1997a Early 2000sb 1960–1980 1981–1990 1991–1997 1960–1980 1981–1990 1991–1997 1960–1980 1981–1990 1991–1997 1960–1980 1981–1990 1991–1997 1980–2000f 1980–2000f 1980–2000f

4.11c 2.50c 2.96c 0.8b 4.29 4.73 1.23 5.91 3.78 4.43 4.45 2.67 1.94 3.50 2.48 2.12 0.99 1.12 0.85

0.65c 1.19c 1.44c 0.3b 1.01 1.16 1.31 1.48 2.32 2.76 1.25 1.13 1.20 1.31 1.28 1.36 –0.06 0.98 –1.09

0.42c 1.65c –0.45c 2.3b 1.46 –0.95 2.32 –1.63 –2.90 –3.00 0.17 –0.23 –0.78 1.81 –0.04 0.82 –1.00 –0.15 –1.86

Egypta,c

Jordana,c

Moroccoa,c

Tunisiaa,c

MENA countries Non-oil MENA countriesd Oil MENA countriese a

Sekkat (2003) and FEMISE Report on the EURO Mediterranean Partnership (2003, p. 20, Box 3). Ismihan and Metin-O¨zcan (2005, Table 4, p. 21 alpha ¼ 0.50). c p ¼ 0.54 d Non-oil MENA countries comprise Egypt, Jordan, Morocco, Syria and Tunisia. e Oil- MENA countries comprise Algeria, Islamic Republic of Iran, and Kuwait. f Hakura (2004, p. 24, Table 4). b

Table 6 presents the contributions of the various factors to economic growth for Turkey and different groups of MENA countries.10 In Turkey, both TFP and capital accumulation were important sources of growth since the 1960s. However, starting from the 1970s macroeconomic instability has increased and become an important problem of the Turkish economy during the late 1980s and the 1990s. During this period political environment was unstable owing to several coalition governments. As a result, TFP growth exhibited signiﬁcant variability over time. In this decade, a rise in TFP growth is remarkable. We believe that the following factors have contributed to the speedup in TFP: (i) successful lowering of inﬂation, (ii) structural reforms and stable political environment, (iii) accession to EU membership (see Ismihan and Metin-O¨zcan, 2005, p. 25).

10

To our knowledge, a comparison between Turkey and the MENA countries has not been made before. Sekkat (2003) compares the growth performance of Morocco and the MENA countries.

250

Kıvılcım Metin O¨zcan, U¨mit O¨zlale and C - ag˘rı Sarıkaya

The basic conclusion to be drawn regarding the Turkish economy is that, from 1990 onward, while the contributions of capital and labor to the economic growth rate are below the average of the MENA as well as the non-oil MENA countries, the Turkish economy performs remarkably well in terms of TFP. The important contributions of both capital and labor to the economic growth rates between 1960 and 1990 were replaced by TFP in the following years. Consequently, our message from this section is clear: despite recent improvements, when the MENA countries are taken as a benchmark, the Turkish economy suffered a high and volatile inﬂation for the whole sample. In addition, in terms of growth performance, it should be noted that the contribution of TFP to the output growth rate is much higher for the Turkish economy, despite the fact that the contributions of labor and capital are below the average. Finally, in terms of output, it should also be added that the growth performance of Turkey changes signiﬁcantly over time, which leads to excessive ﬂuctuations in output around its potential level. Such a ﬁnding necessitates further focusing on these ﬂuctuations. For this purpose, in the following sections, we ﬁrst elaborate on the effects of the macroeconomic policies implemented in generating these negative outcomes. Then, we try to build a link between short-term capital ﬂows and the output gap. 4. Macroeconomic policies and the financial crises Several studies point out that macroeconomic problems intensiﬁed during the 1990s. Thus, it is not surprising to witness two ﬁnancial crises for this particular period, which resulted mainly from incorrect macroeconomic policy decisions.11 In this section, we investigate these policy mistakes. In addition, we focus on two important negative characteristics of the economy for this particular period: speciﬁcally, the lack of ﬁscal discipline and the untimely liberalization of the ﬁnancial markets. As will be clariﬁed below, these two interrelated factors are signiﬁcant, but not solely responsible for the two ﬁnancial crises, which could have been avoided with the right macroeconomic policy decisions. We start our discussion with the currency crisis of 1994. 4.1. The 1994 currency crisis As mentioned in the second section, in April 1994, the Turkish economy suffered a major currency crisis, during which the domestic currency depreciated by almost 70% against the US dollar, interest jumped overnight to 700%, and as a result, economic growth decreased by 6%. As discussed in O¨zatay (2000), despite the worsening macroeconomic fundamentals in the economy such as the

For a detailed analysis of the Turkish economy for the sample period, see Ertug˘rul and Selc- uk (2001), O¨zatay (1999) and S- enses (1991).

11

Sources of Growth and the Output Gap for the Turkish Economy

251

ﬁscal discipline, there was no major sign of a macroeconomic-wide crisis. The risk premium in the economy was quite steady and there was no increase in inﬂation volatility. However, as the Treasury attempted to change the ﬁnancing mechanism of the deﬁcit, canceled the debt auctions and relied heavily upon Central Bank resources, a currency crisis became inevitable.12 It is also interesting to note that the macroeconomic fundamentals in the precrisis and postcrisis periods remained almost the same, implying that such a major crisis could actually have been avoided if these important policy mistakes in the debt ﬁnancing mechanism had not been made. As a result, it can be claimed that the currency crisis in 1994 should not be attributed to the weak macroeconomic performance of the economy. It stems rather from the policy mistakes of the Treasury in ﬁnancing the debt. As discussed further in O¨zatay (2000), there is a policy lesson that can be deduced from the 1994 crisis. In the absence of an effective stabilization program, the government should have done its best to continue debt ﬁnance. Given that the Turkish economy ran very high public deﬁcits during the 1990s, any failure of the Treasury to ﬁnance the debt would naturally result in a currency crisis. 4.2. The 2001 financial crisis The ﬁnancial crisis in February 2001 was the deepest crisis that has ever been experienced in the Turkish economy. To better understand the causes of this crisis, one has to analyze the exchange rate-based stabilization program that was launched in December 1999. The main goal of the program was to bring inﬂation down and satisfy ﬁscal discipline. Based on these two goals, the government used the exchange rate as a credible anchor for inﬂationary expectations. However, similar to the other exchange rate-based stabilization programs that were implemented in other emerging market economies, there was a real appreciation of the domestic currency, and an increasing reliance on capital inﬂows to ﬁnance the current account deﬁcit. At this point, for the program to be successful and to reduce the sensitivity of the economy to capital ﬂows, certain conditions should have been satisﬁed. First, the government should have taken effective measures to ensure ﬁscal discipline. More importantly, the banking sector, which was relied on heavily to ﬁnance the domestic debt, should not have had structural problems. As mentioned in Akyu¨z and Boratav (2001), since the banking sector was largely dependent on high-yielding T-bills, a rapid disinﬂation process, which lowered the real rates in the economy, made the banking sector vulnerable to the implemented stabilization program. Economic agents soon recognized that the program was nowhere near satisfying the goals, which

12

There was a strict belief by the government that the debt dynamics could be changed easily and the high real interest rates could be reduced with this attempt. However, the ﬁnancial markets reacted negatively to this policy change, which resulted in a ﬁnancial crisis.

252

Kıvılcım Metin O¨zcan, U¨mit O¨zlale and C - ag˘rı Sarıkaya

resulted in massive capital outﬂows. As a result, the program had to be abandoned and the domestic currency depreciated signiﬁcantly. The impact of the ﬁnancial crisis on the real side of the economy was also detrimental: the economy shrank at the unprecedented rate of 9.4% in 2001 and the unemployment rate increased from 6.50% in the last quarter of 2000 to almost 12% at the beginning of 2002. When the two economic crises are compared, it can be seen that the 1994 crisis stemmed mainly from the Treasury’s mismanagement of the public debt. For the 2001 crisis, in contrast, the vulnerability of the banking sector and the characteristics of the stabilization program caused the crisis. However, one common feature of these two crises is that the macroeconomic fundamentals were weak and there was no solid macroeconomic structure. Among these macroeconomic fundamentals, ﬁscal dynamics and the liberalization of the ﬁnancial markets emerge as two important characteristics of the Turkish economy. Next, we analyze these two features. 4.3. Fiscal dynamics Contrary to other emerging market economies, Turkey experienced relatively modest accumulated ﬁscal debts up to the mid-1990s. Although the government ran budget deﬁcits, these were largely monetized and interest rates remained at steady levels, excluding the crisis period in 1994. As a result, it can easily be claimed that the main source of the high and persistent inﬂation in the Turkish economy was the lack of ﬁscal discipline. Since the budget deﬁcits were mainly monetized, the size of the accumulated ﬁscal debt could remain at moderate levels. However, as mentioned in Pamukc- u and Yeldan (2005), beginning in the mid-1990s, there were two additional characteristics, which made the implementation of the ﬁscal policy much more difﬁcult. First, the monetization of the debt was no longer an option for the Treasury, which gave way to ﬁnancing the domestic debt through very high real interest rates. Such a strategy not only increased the accumulated ﬁscal debt but also the risk premium required by investors to demand government bonds.13 Second, as a result of these developments, the average maturity of the domestic debt became very short, which made debt ﬁnancing more difﬁcult for the Treasury. In addition, the banking sector bailouts right after the February 2001 ﬁnancial crisis further intensiﬁed the debt management problem. As a result, the government started a new IMF-led program, which relied heavily on ﬁscal restraint. To achieve these goals, the ﬁscal authority aimed to generate a primary budget surplus of 6.5% for the public sector in proportion to the GNP. Although a budgetary discipline seems to be satisﬁed in recent years, real

13

One reason for the high-risk premium was the probability that the government would renege on the domestic debt.

Sources of Growth and the Output Gap for the Turkish Economy

253

interest rates are still very high and the debt to GDP ratio has not decreased signiﬁcantly. Pamukc- u and Yeldan (2005) consider these results totally expected and argue that the debt burden in the economy cannot be handled with the present IMF-led program. They add that, to achieve long-term ﬁscal stability, a detailed restructuring of the terms of the Turkish debt obligations is necessary. 4.4. Liberalization of the financial markets In the last two decades, an increasing number of developing countries have decided to liberalize their ﬁnancial markets. The main motive behind these liberalization attempts was to restore growth and stability by raising savings. The Turkish economy was no exception in this context, and there were intense liberalization attempts, starting at the beginning of the 1980s. However, the most important date for these attempts was the capital account liberalization in August 1989. When the period following the capital account liberalization is observed, it can easily be seen that the foreign direct investments, which promote long-run growth and do not cause any speculative activity, occurred at an almost negligible level. Short-term capital ﬂows were very important and often led to macroeconomic uncertainty through increased ﬁnancial instability. The economy also observed a divergence of domestic savings away from ﬁxed capital investments toward speculative ﬁnancial markets. As a result, with a weak ﬁnancial structure and a shallow ﬁnancial market, massive short-term capital ﬂows caused volatility in output growth and uncertainty in exchange rates. It could be argued that ﬁnancial liberalization could change the structure of the overall economy in a positive manner, by promoting investment in the real side of the economy via extending credits from the ﬁnancial markets. At the same time, the promising sectors could develop more easily owing to increasing capital ﬂows to these sectors. However, as Boratav and Yeldan (2002) show, the liberalization of the ﬁnancial markets was also unable to change the structural sectoral composition, the nature of market concentration and the behavior of proﬁt margins in the post-1980 period. In this context, the liberalization attempts were effective only on the ﬁnancial markets and could not restore longrun economic stability. The liberalization of the ﬁnancial markets could be considered as untimely. In an economy with a weak ﬁnancial structure, shallow exchange rate markets and very high real interest rates owing to the lack of ﬁscal discipline, it was almost inevitable that short-term capital ﬂows would have negative effects on macroeconomic dynamics. The changing risk appetite of foreign investors could easily lead to instability in the economic performance. To make this evident, the next section gives a clearer perspective concerning the relationship between excessive ﬂuctuations in output and short-term capital ﬂows.

254

Kıvılcım Metin O¨zcan, U¨mit O¨zlale and C - ag˘rı Sarıkaya

5. Output gap and business cycle characteristics The ﬁndings in the previous sections show that there have been excessive ﬂuctuations in the output dynamics for the Turkish economy during the last two decades. In this context, it can be stated that there existed a boom–bust cycle, where a sustainable growth path could not be achieved. This characteristic of the Turkish economy differentiates it from its industrialized counterparts, necessitating a thorough analysis of these excessive ﬂuctuations. Therefore, in addition to our discussion of the sources of growth, we investigate this issue in brief. For this purpose, we utilize an output gap measure, where the gap is taken as the difference between the actual output and the potential output. While a positive gap implies that the actual output was above its potential level, thus exerting pressure on the price dynamics, a negative gap implies the opposite. However, measuring the output gap is controversial, especially for emerging market economies. There are two main reasons for this. First, conventional approaches, such as the HP ﬁlter and the quadratic time ﬁlter, are purely statistical methods, which do not convey any economic information. Second, as stated by Aguiar and Gopinath (2004), emerging markets are prone to frequent shocks that affect the potential output. Therefore, these economies typically do not have very smooth trends.14 Taking these factors into account, we adopted the output gap measure that was derived exclusively for the Turkish economy in Ece et al. (2005). As can be seen in the appendix, this gap measure is obtained from a semi-structural timeseries model, which represents the general characteristics of the Turkish economy. Thus, it is not subject to the criticisms that apply to conventional methods and it can be conveniently used in our analysis. Figure 3 displays the actual output, the estimated potential output and the output gap. At ﬁrst sight, it can be seen that the crisis and recovery periods are captured well. The output gap measure displays the fact that, several times in the sample period, the Turkish economy was subject to severe shocks that created an unstable environment, ending with rapid contractions in the economic activity. As a consequence, output gap estimates exhibit sharp movements. According to the output gap ﬁgure, expansion periods were generally interrupted by economic crises so that the last decade seems to have witnessed three separate periods of recession. Such an observation validates both the excessive boom–bust cycle peculiar to emerging markets and the ‘‘unsustainable growth path’’ that the Turkish economy has followed in the last decade.

14

For a discussion of the HP ﬁlter, see Butler (1996) and Boone et al. (2001).

Sources of Growth and the Output Gap for the Turkish Economy

Figure 3.

255

Actual output, potential output and output gap Actual and Potential Output

10.5 10.4 10.3 10.2 10.1 Actual Output Potential Output

10.0

04Q1

01Q1 01Q1

03Q1

00Q1 00Q1

02Q1

99Q1 99Q1

98Q1

97Q1

96Q1

95Q1

94Q1

93Q1

92Q1

91Q1

90Q1

9.9

Output Gap 8 6

% deviation

4 2 0 -2 -4 -6 -8 04Q1

03Q1

02Q1

98Q1

97Q1

96Q1

95Q1

94Q1

93Q1

92Q1

91Q1

90Q1

-10

5.1. Role of short-term capital flows Emphasis can be placed on the signiﬁcance of short-term capital ﬂows as a major determinant of the above-mentioned output gap, especially after the capital account liberalization of 1989. Capital ﬂows that are accompanied by weak ﬁnancial structure have led to a number of economic crises with severe social and economic costs in many emerging market economies. A similar argument can be valid for Turkey considering the growth performance in the aftermath of the ﬁnancial liberalization policy. The experiences of the last decade especially point to the dependence of growth performance on capital ﬂows (Figure 4). In fact, economic conditions providing speculative arbitrage opportunities were a result of unsuccessful disinﬂationary efforts, which dragged the economy

Kıvılcım Metin O¨zcan, U¨mit O¨zlale and C - ag˘rı Sarıkaya

256

Figure 4.

Capital flows and GDP growth

Net Capital In flow, billions dollar, left axis

2004

2003

-12 2002

-15 2001

-8

2000

-10

1999

-4

1998

-5

1997

0

1996

0

1995

4

1994

5

1993

8

1992

10

1991

12

1990

15

GDP Growth, %, right axis

into ‘‘slowed down depreciation-correction’’ cycles (Ertug˘rul and Selc- uk, 2001). In this respect, the ﬂuctuations during the 1990s can be characterized as a ﬁnancial cycle determined by capital movements. Therefore, periods of capital inﬂows coincided with high economic growth periods followed by subsequent real depreciation and crisis periods. In an effort to control inﬂation as well as ease domestic borrowing, the policies focused on slowing down the depreciation of the Turkish lira. However, policies oriented toward controlling the exchange rate generally lead to an initial decline in interest rates and a real appreciation of the domestic currency owing to the slow catch-up of the inﬂation rate. As a stylized fact, similar to the general characteristic of exchange rate-based stabilization programs, an initial upsurge in domestic demand and real activity along with real appreciation give rise to an increase in imports. The subsequent deterioration of the current account balance and concerns regarding its sustainability reverse the process and trigger capital outﬂows. Hence, the economy comes to its initial point again with high interest rates, depreciated domestic currency, contracted domestic demand and an improved current account balance. Such a macroeconomic instability results in the rising volatility of the growth rate and thus creates mini business cycles, the duration of which shortens as a result of the huge and rapid changes in the size and direction of short-term capital ﬂows. Figure 5 demonstrates the tight relationship between net capital inﬂows and the real exchange rate, except for the years 1998 and 1999. During the 1995–1998 period, capital ﬂows had been quite stable at positive levels and the real exchange was appreciating. It was possible to sustain the appreciation of the

Sources of Growth and the Output Gap for the Turkish Economy

Figure 5.

257

Capital flows, real exchange rate and business cycle

8

160

6

150

4

140

2

130

Net Capital Inflow, bln.$, left axis

Jan-04

Jan-03

Jan-02

Jan-01

Jan-00

Jan-99

80 Jan-98

-8 Jan-97

90 Jan-96

100

-6 Jan-95

-4

Jan-94

110

Jan-93

120

Jan-92

0 -2

Real Exchange Rate, right axis

8 6 4 2 0 -2 -4 -6 -8

Net Capital Inflow, bln.$

Jan-04

Jan-03

Jan-02

Jan-01

Jan-00

Jan-99

Jan-98

Jan-97

Jan-96

Jan-95

Jan-94

Jan-93

Jan-92

-10

Output Gap, %

Turkish lira in spite of two consecutive external shocks, the Asian and Russian economic crises; the latter resulted in a huge capital outﬂow in the third quarter of 1998. Obviously, such an appreciation leads to a temporary increase in output through two distinct channels. First, a surge in consumption is observed, which increases the aggregate demand. Second, an increase in the imported intermediate and capital goods is observed, which further increases output in the economy. In terms of monetary policy, it should be noted that during the 1995–1999 period, the main goal for the central bank was to maintain ﬁnancial stability through a real exchange rate targeting strategy. Thus, other than pursuing a clear price stability objective, the central bank mainly tried to stabilize the negative effects of the volatile capital ﬂows on the real exchange rate for the sample period.

Kıvılcım Metin O¨zcan, U¨mit O¨zlale and C - ag˘rı Sarıkaya

258

Table 7. Cross correlations of GAPt with STCFt7i (t–5)

(t–4)

(t–3)

(t–2)

(t–1)

(t)

(t+1)

(t+2)

(t+3)

(t+4)

(t+5)

0.05

0.05

0.19

0.36

0.58

0.54

0.23

–0.11

–0.30

–0.37

–0.18

As a result, it can be claimed that short-term capital ﬂows stand out as a very important factor in generating the excessive ﬂuctuations in output performance. The comovement of the output gap and the capital inﬂows in Figure 5 provide further support for this claim. In addition, the close relationship between these two series does not seem to have broken down in recent years. Such an observation makes us think that the high economic growth rates and the performance of the Turkish economy above its potential level in recent years owe much to the increasing short-term capital inﬂows. At this stage, it is still early to state whether the Turkish economy has entered a sustainable growth path. The recent developments more closely resemble an expansionary phase of a boom–bust cycle. As a ﬁnal exercise, it can also be claimed that short-term capital ﬂows are attracted by the state of the economy, i.e. the output gap. When the economy performs at high economic growth rates, investors might have incentives to invest in Turkish ﬁnancial securities, which, in turn, generate short-term capital inﬂows. To investigate this issue, we ﬁrst checked the cross correlations of the output gap and the short-term capital ﬂows. Then, we performed a Grangercausality test between these two variables. The results can be seen below. For the 1992Q1–2004Q4 period, the cross correlations between the output gap (GAP) and short-term capital ﬂows (STCF) show that the highest correlation (0.57) is observed between GAPt and STCF t-1 . Therefore, STCF may be a leading indicator of the output gap (Table 7). However, since correlation itself cannot be interpreted as a causality between the two variables, we also perform a Granger causality test by estimating a Vector Autoregression (VAR) system. According to the test results, past values of STCF can be used to forecast GAP at the 5% signiﬁcance level, while the opposite is not true. Therefore, we ﬁnd a one-way causality between the two variables and shortterm capital ﬂows are the Granger-cause of the output gap (Table 8). This ﬁnding is an empirical one, which also has an underlying economic reason. Short-term capital inﬂows lead to a real appreciation of the domestic currency, which may affect the output gap from different channels.15 First of all, a real appreciation of the Turkish lira may trigger the demand for imported goods by bringing their relative prices down, thereby increasing the output gap. It may also lead to a narrowing of the output gap through an increase in the potential output by inducing ﬁrms to import capital goods from abroad. In contrast, in a

15

For a more detailed discussion of the effect of the real exchange rate on the output gap, see Ece et al. (2005).

Sources of Growth and the Output Gap for the Turkish Economy

Table 8.

259

Granger causality test between GAP and STCF

Dependent variable: GAP Excluded STCF Dependent variable: STCF Excluded GAP

w2 5.7456

P-value 0.0165

w2 0.1361

P-value 0.7122

country like Turkey, which is heavily dependent on intermediate goods’ imports to expand production, a real appreciation may also increase the supply in the economy. Regardless of which effect dominates, short-term capital ﬂows surely have a signiﬁcant impact on consumption, investment and production decisions, and thus on the output gap, mainly through the exchange rate channel. 5.2. Business cycle facts for Turkey After stressing the role of the short-term capital ﬂows in generating the output gap, this subsection illustrates the business cycle facts for Turkey by examining how the cyclical behavior of the selected macroeconomic variables are related to the output gap. Such an exercise reveals which factors lead or lag behind the ﬂuctuations in the output as well as showing whether these variables are positively (procyclical) or negatively (countercyclical) associated with the output gap. Following Kydland and Prescott (1990), we ﬁrst characterize the business cycle facts for Turkey by using the HP-ﬁlter to detrend selected real factors and then we compare their cyclical behavior with our output gap estimate. We employed the HP-ﬁlter in decomposing the real factors for two reasons. First, we wanted the business cycle analysis to be consistent with the exercise that Kydland and Prescott (1990) provided. Second, it would require a signiﬁcant amount of additional work to come up with a different structural model for each factor and then employ the Kalman ﬁlter to decompose these factors into their trend and cyclical components. Therefore, we decided to adopt the main variable of interest (the output gap) from a structural framework while we used conventional techniques for the other variables. The main implications of the analysis did not change even when we derived the output gap from the HP-ﬁlter. The cyclical behavior of the series is evaluated according to the degree of comovement with the output gap and depending on whether there is a phase shift in the movement of a variable relative to the cycle. A positive (negative) contemporaneous correlation coefﬁcient is interpreted as the variable being procyclical (countercyclical). If the largest correlation coefﬁcient appears in column (t-i), where i>0, then the series tends to move i quarters before the real GDP and leads the cycle. Similarly, the largest correlation coefﬁcient appearing in column (t+j), where j>0, indicates that the series lags behind the gap by j quarters. The cyclical facts are summarized in Table 9. At ﬁrst glance, one obvious fact is that most of the selected variables are procyclical and coincide with the

Kıvılcım Metin O¨zcan, U¨mit O¨zlale and C - ag˘rı Sarıkaya

260

business cycle in Turkey. Moreover, major output components, private consumption and private investment tend to move together with the cycle. Our output gap estimate is closely tied to private consumption making it procyclical, as their contemporaneous correlation is 0.82. Durable consumption also comoves with the cycle with a high correlation. Private investment, one of the main determinants of potential output, exhibits a strong contemporaneous relationship with the output gap, so private investment is said to be procyclical. Moreover, among the subitems of private investment expenditures, machineryequipment investment is closely related to the cycle whereas building investment is found to be acyclical. Selected indicators of resource utilization, such as capacity utilization, number of workers and working hours, move with the output gap simultaneously. Finally, the relationship between labor costs and output gap is found to be a lagged one rather than a contemporaneous one. Both real wages and real unit wages move with a one-quarter lag relative to the cycle. Therefore, the cyclical behaviors of real wages and real unit wages turn out to be consistent with the standard adjustment process in the sense that a positive output gap puts upward pressure on real marginal costs through increased demand for labor. Finally, it can be claimed that the output gap measure that is employed throughout the analysis can be criticized for being purely statistical and lacks the theoretical background. For this purpose, to check the robustness of the results, we used another output gap measure, which is derived from Dees et al. (2005). This measure also takes into account the evolution of trend output of Turkey’s trading partners. As Tables B.1, B.2 and B.3 in Appendix B shows, while we ﬁnd slightly smaller correlations between output gap and the other relevant variables, the main implications of the business cycle analysis do not change. In addition, we ﬁnd that the short-term capital ﬂows again Granger-cause the output gap, which supports the previous empirical results.

Table 9.

Cross correlation of output gap (Gapt) with selected variables (1988Q2–2004Q4)

Private consumption Durable consumption Private investment Machinery equipment Construction Capacity utilization Number of workers Working hours Labor productivity Real wagea Real unit wagea a

(t–5)

(t–4)

–0.13 –0.18 –0.12 –0.14 –0.01 –0.02 –0.01 0.01 –0.07 –0.02 0.15

–0.09 0.09 0.34 0.64 0.82 –0.08 0.12 0.37 0.62 0.74 –0.08 0.09 0.28 0.53 0.66 –0.09 0.09 0.31 0.60 0.72 –0.02 –0.01 0.00 0.06 0.11 0.03 0.17 0.36 0.53 0.59 0.00 0.08 0.23 0.37 0.49 –0.01 0.09 0.26 0.44 0.54 –0.06 0.12 0.29 0.48 0.53 0.09 0.22 0.37 0.52 0.61 0.22 0.25 0.27 0.38 0.51

Correlation in 1999Q1–2004Q4.

(t–3) (t–2) (t–1)

(t)

(t+1) (t+2) (t+3) (t+4) (t+5) 0.60 0.52 0.56 0.60 0.16 0.25 0.44 0.41 0.20 0.62 0.62

0.27 0.24 0.32 0.33 0.16 –0.09 0.22 0.17 –0.02 0.42 0.52

0.02 0.05 0.07 0.05 0.12 –0.29 –0.05 –0.12 –0.12 0.20 0.32

–0.11 –0.03 –0.09 –0.13 0.17 –0.37 –0.23 –0.30 –0.12 –0.04 0.03

–0.17 –0.07 –0.18 –0.22 0.17 –0.27 –0.32 –0.33 0.02 –0.16 –0.17

Sources of Growth and the Output Gap for the Turkish Economy

261

6. Conclusion Unsustainable output growth and high inﬂation are the two prevailing macroeconomic characteristics of the Turkish economy. These two factors also indicate a boom–bust cycle, which the economy has experienced more severely in the last decade. In this respect, as of mid-2005, after the negative effects of the 2001 crisis were partly eliminated, there has been considerable discussion about whether the recent positive developments in the output and the inﬂation dynamics are the result of a new macroeconomic policy framework or whether they simply reﬂect an expansionary phase of such a business cycle. Taking this discussion as our reference point, we tried to analyze the growth dynamics in the Turkish economy. Throughout the study, we handled the issue in three ways. First, we presented an overview of the Turkish economy from a growth perspective and compared its performance in terms of output and inﬂation with the other MENA countries. The main ﬁnding from this approach is that Turkey has not followed the right macroeconomic policies to have a sustainable growth path. Moreover, the Turkish economy has suffered from high and volatile inﬂation. In addition, the growth accounting exercise shows that the TFP is the main cause behind the recent increase in output growth rates for the Turkish economy. Then, we discuss the characteristics of these incorrect macroeconomic policies, with a special reference to the two ﬁnancial crises in 1994 and 2001. This discussion brings to the fore an important link between short-term capital inﬂows and excessive output ﬂuctuations. Finally, we performed an empirical investigation of the role of the short-term capital ﬂows in generating these ﬂuctuations. It can be seen clearly that there is a very close relationship between the output dynamics and the capital inﬂows. The periods of high growth rates also coincide with the periods of massive short-term capital inﬂows. The fact that these capital inﬂows reached a maximum by mid-2005 makes us think that the recent positive developments owe much to the increases in the capital inﬂows. Thus, we can say that it is too early to judge the success of the new macroeconomic policy framework that was put into effect after the 2001 crisis. The effects of this new framework will be better understood when the short-term capital inﬂows decrease to moderate levels. Lastly, based on the output gap measure, we analyzed the business cycle characteristics of the Turkish economy. As a result, we found that most of the major output components tend to move with the cycle. Moreover, the indicators of resource utilization move with the output gap simultaneously. Finally, a positive output gap exerts an upward pressure on the marginal costs through the labor demand channel. References Aguiar, M. and G. Gopinath (2004), Emerging market business cycles: the cycle is the trend, Mimeo, University of Chicago.

262

Kıvılcım Metin O¨zcan, U¨mit O¨zlale and C - ag˘rı Sarıkaya

Akyu¨z, Y. and K. Boratav (2001), ‘‘The making of the Turkish ﬁnancial crisis’’, prepared for a conference on Financialization of the Global Economy, PERI, December 7–9, University of Massachusetts, Amherst, MA. Aricanli, T. and D. Rodrik (1990), The Political Economy of Turkey: Debt, Adjustment, and Sustainability, London: Macmillan. Boone, L., M. Juillard, D. Laxton and P. N’Diaye (2001), ‘‘How well do alternative time-varying parameter models of the NAIRU help policy makers forecast unemployment and inﬂation in the OECD countries?’’, IMF Working Paper, Draft version. Boratav, K. and E. Yeldan (2002), ‘‘Turkey, 1980–2000: ﬁnancial liberalization, macroeconomic (in)-stability and patterns of distribution’’, Mimeo, January. Bosworth, B. and S.M. Collins (2003), The empirics of growth: an update, Brookings Institution. Butler, L. (1996), ‘‘A semi-structural method to estimate potential output: combining economic theory with a time-series ﬁlter’’, The Bank of Canada’s New Quarterly Projection Model (QPM), Part 4. Celasun, M. and D. Rodrik (1989), ‘‘Debt, adjustment and growth: Turkey’’, in: J. Sachs and S.M. Collins, editors, Developing Country Debt And Economic Performance, Country Studies, Chicago: University of Chicago Press. Conway, P. (1990), ‘‘The record on private investment in Turkey’’, in: T. Aricanli and D. Rodrik, editors, The Political Economy of Turkey: Debt, Adjustment, and Sustainability, London: Macmillan. Dees, S., F. Di Mauro, M.H. Pesaran and L.V. Smith (2005), ‘‘Exploring the international linkages of the Euro area: a global VAR analysis’’, CESifo Working Paper No. 1425. Ece, D., H. Kara, F. O¨g˘u¨nc- , U¨. O¨zlale and C - . Sarikaya (2005), ‘‘Estimating output gap for the Turkish economy’’, Unpublished Manuscript, The Central Bank of the Republic of Turkey, Research Department, Ankara, Turkey. Ekinci, N. (1990), ‘‘Macroeconomic developments in Turkey: 1980–1988’’, METU Studies in Development, Vol. 17(1–2), pp. 73–114. Ekinci, N. (2000), ‘‘The IMF and reforming the public sector in Turkey’’, METU E.R.C. Working Papers in Economics, No. 00/04, Ankara, METU. Ertug˘rul, A. and F. Selc- uk (2001), ‘‘A brief account of the Turkish economy 1980–2000’’, Russian and East European Finance and Trade, Vol. 37(6), pp. 6–28. Femise Report on the EURO Mediterranean Partnership (2003), ERF, Cairo, Egypt and Institut de la Mediterranee, France. Hakura D. S. (2004), ‘‘Growth in the Middle East and North Africa’’, IMF Working Paper WP/04/56. Independent Social Scientists’ Alliance (ISSA) Report (2005), ‘‘On economic and social life in Turkey in early 2005’’, http://www.bagimsizsosyalbilimciler.org. I˙smihan, M. (2003), ‘‘The role of politics and stability on public spending dynamics and macroeconomic performance: theory and evidence from Turkey’’, Unpublished PhD Thesis, The Graduate School of Social Sciences of Middle East technical University, Ankara.

Sources of Growth and the Output Gap for the Turkish Economy

263

Ismihan, M. and K. Metin-O¨zcan (2005), ‘‘Sources of growth in Turkish economy, 1960–2004’’, ERF 12th Annual Conference Paper, Cairo, Egypt. I˙smihan, M., K. Metin-O¨zcan and A. Tansel (2005), ‘‘The role of macroeconomic instability in public and private capital accumulation and growth: the case of Turkey 1963–1999’’, Applied Economics, Vol. 37, pp. 239–251. Kydland, F. and E. Prescott (1990), ‘‘Business cycles: real facts and a monetary myth’’, Federal Reserve Bank of Minneapolis Quarterly Review, Spring, pp. 3–18. Ljung, L. and T. Soderstrom (1983), Theory and Practice of Recursive Identiﬁcation. The MIT Press. Makdisi, S., Fattah Z. and I. Limam (2000), ‘‘Determinants of growth in the MENA countries’’, The World Bank sponsored project on the Global Development Network. Maras-liog˘lu, H. and Tiktik, A. (1991), Tu¨rkiye Ekonomisinde Sekto¨rel Gelis- meler : U¨retim, Sermaye Birikimi veI˙stihdam 1968– 1988, Devlet Planlama Tes- kilati, I˙ktisadi Planlama Bas- kanlig˘i, DPT: 2271- I˙PB: 428. Metin, K. (1995), The analysis of inflation: the case of Turkey (1948– 1988), Ankara: Nurol Publishing Co., Capital Markets Board Publication No. 20. Metin-Ozcan, K., E. Voyvoda and E. Yeldan (2001), ‘‘Dynamics of macroeconomic adjustment in a globalized developing economy: growth, accumulation and distribution, Turkey 1969–1998’’, Canadian Journal of Development Studies, Vol. 22(1), pp. 219–253. O¨zatay, F. (1999), ‘‘Populist policies and the role of economic institutions in the performance of the Turkish economy’’, Yapi Kredi Economic Review, Vol. 10(1), pp. 13–26. O¨zatay, F. (2000), ‘‘The 1994 currency crisis in Turkey’’, Journal of Policy Reform, Vol. 3, pp. 327–352. Pamukc- u T. and Yeldan, E. (2005), ‘‘Country proﬁle: Turkey, public sector and ﬁscal policy issues’’, Mimeo, March 2005. Saygili, S- . (1999), Technical change efficiency, growth and exports: the case of Turkish economy, Unpublished PhD thesis, University of Kent at Canterbury, England. Saygili, S- ., C. Cihan ve H. Yurtog˘g˘lu (2005), ‘‘Tu¨rkiye Ekonomisinde Sermaye Birikimi Verimlilik ve Bu¨yu¨me: 1972–2003, DPT yayin No. 2686, Ankara. Sekkat, K. (2003), ‘‘The sources of growth in Morocco: an empirical analysis in a regional perspective’’ ERF Working Paper, Cairo. Senhadji, A. (2000), ‘‘Sources of economic growth: an extensive growth accounting exercise’’, IMF Staff Papers, Vol. 47(1), pp. 129–157. S- enses, F. (1991), ‘‘Turkey’s stabilization and structural adjustment program retrospect and prospect’’, The Developing Economies, Vol. 29, pp. 210–234. Somc- ag˘, S. (2004), ‘‘Confession of imaginary stock by the SIS’’, http://www.selimsomcag.org. State Institute of Statistics (SIS) (2004), Household Labor Surveys, Ankara, Turkey.

264

Kıvılcım Metin O¨zcan, U¨mit O¨zlale and C - ag˘rı Sarıkaya

State Planning Organization (SPO) (2004), Pre-Accession Economic Programme, Ankara: Republic of Turkey: SPO, November 2004. Temel, A., ve S- . Saygili (1995), ‘‘An estimation of gross ﬁxed capital formation’’, in: T. Bulutay, editor, Investment and the Labour Market in Turkey: Proceedings of a Seminar Held in Ankara 7 December 1995, Ankara; SIS. Temel, A., E. Boyar and S- . Saygili (2000), Tu¨rkiye Ekonomisinde Bu¨yu¨me ve Ekonomik Yapi Deg˘is- meleri: 1946-1999, Unpublished Manuscript, State Planning Organization, Ankara, Turkey. UNCTAD (2002, 2003), Trade and Development Report, 2002 and 2003, New York: UNCTAD. Uygur, E. (1995), ‘‘Recent estimates of investment and private investment behaviour in Turkey’’, in T. Bulutay, editor, Investment and The Labour Market in Turkey: Proceedings of a Seminar Held in Ankara 7 December 1995, Ankara; SIS. World Bank Country Study (1980), Turkey: policies and prospects for growth, Washington, DC, USA: The World Bank. Appendix A: The output gap estimation for the Turkish economy This appendix brieﬂy introduces the output gap model and the methodology that is used in Ece et al. (2005). The system of equations, in which the output gap is derived as the difference between the actual output and the potential output, is as follows: (1) Inﬂation-output gap dynamics: pt ¼ a1;t pt1 þ a2;t pt2 þ a3;t gapt1 þ a4;t reert þ nt (2) Actual output decomposition: yt ¼ yt þ gapt (3) Potential output equation: yt ¼ yt1 þ mt1 þ Zt (4) Potential output growth rate equation: mt ¼ ð1 rt Þm0 þ rt mt1 þ t (5) Output gap dynamics: gapt ¼ g1;t gapt1 þ g2;t rt þ g3;t DI t þ g4;t reert þ Bt where pt is the inﬂation rate deﬁned as the logarithmic difference of quarterly seasonally adjusted consumer price index (CPI), gapt the unobserved output gap, reert the logarithmic difference of the real effective exchange rate, yt the logarithmic seasonally adjusted real gross domestic product, y*t the unobserved potential output, mt the potential output growth rate, rt the ex post

Sources of Growth and the Output Gap for the Turkish Economy

265

real interest rate based on 3-month Treasury auction rates and DIt the demand index, which is constructed from the Business Tendency Survey of the Central Bank of the Republic of Turkey. In contrast, vt, Zt, et and xt represent shocks to the system, which are assumed to be i.i.d. with zero mean and constant variances. It is important to remind that the parameters of the system are time varying. Therefore, one has to make a time-series speciﬁcation for the evolution of these parameters. It is assumed that each time-varying parameter follows a random walk. Such a speciﬁcation can be defended on theoretical grounds: Since any structural change on the dynamics of the model–thus the system parameters–cannot be known a priori, it is intuitive to specify a random-walk process for each parameter. As a result, the system includes nine more equations, where each time-varying parameter follows a random-walk process. Finally, it should be noted that the output gap variable, which needs to be estimated within the system, appears in multiplicative form with the time-varying parameters, which, again, are to be estimated. Thus, the model has a non-linear characteristic, where the standard Kalman ﬁlter is inappropriate and should be replaced by the extended Kalman ﬁlter algorithm, which is designed for estimating non-linear state space models. For details about the extended Kalman ﬁlter algorithm, one can either read the appendix in Ece et al. or ‘‘Theory and practice of recursive identiﬁcation’’ by Ljung and Soderstrom (1983), from which the appendix in Ece et al. (2005) is derived. Appendix B: Robustness tests with the output gap measure from Dees et al. (2005) Table B.1.

Cross correlations of GAPt with STCFt7i (1992Q1–2003Q4)

(t–5) (t–4) (t–3) (t–2) (t–1) (t) (t+1) (t+2) (t+3) (t+4) (t+5) –0.17 –0.09 0.09 0.23 0.49 0.47 0.24 0.04 –0.17 –0.21 –0.07 Table B.2.

Granger causality test between GAP and STCF (1992Q1–2003Q4)

Dependent variable: GAP Excluded STCF

w2 7.7550

P-value 0.0207

Dependent variable: STCF Excluded GAP

w2 2.0692

P-value 0.3554

266

Private consumption Durable consumption Private investment Machinery equipment Construction Capacity utilization Number of workers Working hours Labor productivity Real wagea Real unit wagea a

Cross correlation of output gap (Gapt) with selected variables (1988Q2–2003Q4) (t–5)

(t–4)

(t–3)

(t–2)

(t–1)

(t)

(t+1)

(t+2)

(t+3)

(t+4)

(t+5)

–0.17 –0.21

–0.16 –0.19

–0.02 –0.02

0.13 0.12

0.35 0.30

0.52 0.47

0.44 0.38

0.32 0.28

0.17 0.17

0.02 0.05

–0.04 0.00

–0.11 –0.13

–0.13 –0.12

0.00 0.00

0.12 0.15

0.32 0.36

0.45 0.48

0.43 0.46

0.35 0.34

0.20 0.18

0.04 0.00

–0.03 –0.07

–0.01 0.18 –0.07 –0.04 –0.04 –0.54 –0.45

–0.07 0.22 -0.04 –0.02 –0.05 0.09 –0.47

–0.08 0.33 0.05 0.07 0.08 –0.38 –0.37

–0.09 0.46 0.18 0.21 0.10 –0.21 –0.22

–0.07 0.60 0.29 0.34 0.31 0.03 –0.01

–0.05 0.68 0.40 0.45 0.36 0.29 0.22

0.03 0.43 0.40 0.39 0.19 0.48 0.47

0.11 0.21 0.30 0.28 0.11 0.43 0.48

0.16 0.00 0.15 0.13 –0.03 0.37 0.42

0.24 –0.15 0.00 –0.04 –0.08 0.22 0.29

0.29 –0.17 –0.08 –0.09 –0.06 0.18 0.15

Correlation in 1999Q1–2004Q4.

Kıvılcım Metin O¨zcan, U¨mit O¨zlale and C - ag˘rı Sarıkaya

Table B.3.

CHAPTER 9

Economic Growth and Economic Policy in Iran: 1950–2003 Ahmad R. Jalali-Naini Abstract The Iranian experience reveals, in a number of ways, the potentials, problems, and challenges that have been confronted by other populated oil-exporting countries – a tale of high expectations turning into mediocre performance. Three distinct growth periods, with different political and policy environments, are identified over the 1950–2003 period. From the 1960s up to the first oil boom (1973–1974) the economy exhibited high GDP growth and low inflation rates. The oil boom brought in large financial resources, but also induced soft-budget constraints, and expanded the public sector. The huge increase in oil-financed government spending raised growth temporarily above its trend; however, by creating inflation and opportunities for rent seeking, it resulted in economic and social instability. The positive wealth and saving effects of the oil boom did not have a long-lasting effect on either economic growth rates or on per-capita income levels. Since 1974 economic growth has been volatile and factor-intensive and total factor productivity did not make a significant contribution to economic growth. The evidence also does not support that human capital accumulation has strongly contributed to economic growth. While fixed investment/GDP ratios in Iran compare favorably with high growth economies, the growth performance does not; too many resources have been spent to produce only an average growth performance. It is argued that Iran’s relatively rich resources, high rates of accumulation of physical and human capital, but a mundane growth performance and continued dependence on oil reflect inadequate economic institutions, price/incentive distortions, policy coordination and rentseeking problems, and the effect of the Iran–Iraq war. In the episodes with superior

Corresponding author. CONTRIBUTIONS TO ECONOMIC ANALYSIS VOLUME 278 ISSN: 0573-8555 DOI:10.1016/S0573-8555(06)78009-9

r 2007 ELSEVIER B.V. ALL RIGHTS RESERVED

268

Ahmad R. Jalali-Naini

growth performance, economic and institutional uncertainty and transaction costs were lower, the price structure was more rational, and government policies allowed wider participation by the private sector. Keywords: Economic growth, Iran, government policy JEL classifications: O40, O53, Q38 1. Introduction Iran’s growth experience over the 1950–2003 period is, in many ways, similar to the opportunities, problems, and challenges that have been confronted by other populated oil-exporting countries. It is a case of high expectations turning into mediocre performance. The positive wealth and saving effects of the oil boom did not have a long-lasting effect on economic growth rates. Since 1974 economic growth has been volatile and factor-intensive and total factor productivity (TFP) did not make a signiﬁcant contribution to economic growth. While ﬁxed investment/GDP ratios in Iran compare favorably with high growth economies, the growth performance does not. In this paper we will argue that Iran’s relatively rich resources, high rates of accumulation of physical and human capital (HC), but mediocre growth rates and continued dependence on oil reﬂect inadequate economic institutions, price/incentive distortions, policy coordination, and rentseeking problems – and the effect of the Iran–Iraq war in the 1980s. This paper identiﬁes three distinct phases of economic growth in Iran during the 1950–2003 period, compares the growth performance of the Iranian economy with other developing countries (DCs), identiﬁes the sources of growth, and examines the inﬂuence of oil and government economic policies on economic growth. The paper is organized as follows. Section 2 reviews three distinct growth periods in Iran and sectoral growth patterns. Section 3 reviews the modern ‘‘stylized facts’’ of growth and their empirical relevance in Iran. In Section 4 we use the conventional growth accounting methods to measure the sources of economic growth. In Section 5 the inﬂuence of important growth correlates such as oil, ﬁscal variables, macroeconomic uncertainty, price distortions, and ﬁnancial deepening for the Iranian economy will be tested through standard growth regressions. Section 6 discusses oil, the state, and rent seeking and growth performance. Section 7 presents a summary and some policy options. 2. Aggregate growth performance and sectoral growth patterns: 1950–2003 Modern economic growth is predicated on a number of social, economic, and institutional transformations; ﬁrst and foremost being the growth of the domestic market. Market production increasingly supplanted ‘‘natural economy’’ during

Economic Growth and Economic Policy in Iran: 1950– 2003

269

the ﬁrst half of the 20th century in Iran. Demographic changes, migration, urbanization, oil discovery, and transportation breakthroughs accelerated this process. Population growth, particularly in the urban areas picked up: the share of urban population rose from around 20 percent at the turn of the 20th century to 40 percent in 1955 – while the size of the population nearly tripled. Trade between rural areas and the towns expanded as expansion of road networks and highways reduced transportation costs quite signiﬁcantly (Issawi, 1971). Safer and more secure roads facilitated greater local and regional trade. Growing oil revenues stimulated urbanization and monetization of the economy. The rural areas supplied growing quantities of food and agricultural cash crops to the urban areas. During the early 1920s to the mid-1950s a justice system was created and a set of legal codes for trade and commerce was enacted. Moreover, a public administration system, a ﬁscal system, and a national bank were created. The number of small and large manufacturing ﬁrms increased signiﬁcantly during the 1930s; oil revenues ﬁnanced imports of capital and intermediate goods for this sector.1 This period also witnessed growth in wage-laborer employment on a signiﬁcant scale. Back-cast projections put the GDP growth rate estimates during the 1937–1950 period at 4.42 percent per annum and indicates that the growth rate had picked up in the 1950s.2 Based on the available data, we can distinguish three different growth phases or growth cycles in Iran: 1959–1977 (Persian calendar years 1338–1356), 1978–1988, and 1989–2003. The time trend for real GDP, non-oil GDP, and per-capita GDP are shown in Figures 1 and 2. The average growth rates and growth ﬂuctuations, measured by their standard deviations, are shown in Table 1. The average growth rate of non-oil GDP in constant prices during the ﬁrst period was 10.2 percent per annum. A clear fall in the growth trend can be observed in the 1978–1988 interval covering the revolution and the Iran–Iraq war. In the third period, there is another break in the trend as economic growth resumes following the end of the war. The average annual growth rate of non-oil GDP in the third period was 5.4 percent. On a per-capita basis, the distinctions between the three different periods are starker. The average non-oil GDP per-capita growth rate in the second period is about -4 percent. 3. Stylized facts of growth The main features of the economic growth of nations are described by the stylized facts. The original stylized facts of growth (Kuznets and Kaldor) focused on a number of ‘‘grand’’ ratios and their evolution through time, speciﬁcally: potential

1

In 1926 only 30 manufacturing establishments employing >50 workers including the oil industry were in existence. By 1956, the total number of large manufacturing units had grown to 114. Total employment in this sector had grown to around 520,000 persons by 1956; the great majority of the employees worked in small artisan-type workshops (Bharier, 1971). 2 The above period is equivalent to the 1316–1329 period in the Persian calendar, see, Khavari (2001).

Ahmad R. Jalali-Naini

270

Figure 1.

Real GDP Trend Iran

500000

400000

300000

200000

100000

0 60

65

70

75

80

85

90

95

100

GDP, constant 1997 prices, in Billion Rials Non-Oil GDP, constant 1997 prices, in Billion Rials Source: The Central Bank of the Islamic Republic of Iran, henceforth BMI

Figure 2.

Per capita Real GDP and Non-Oil GDP in Iran

28000 24000 20000 16000 12000 8000 4000 60

65

70

75

80

85

90

95

100

GDP per capita, 000 Rials, 1997 prices Non-Oil GDP per capita, 000 Rials, 1997 prices (source BMI)

Source: BMI

Economic growth in Iran, selected periods (in percentages)

1959–2003 a

1978–1988

1989–2003

Non-oil GDP

GDP

Non-Oil GDP

GDP

Non-Oil GDP

GDP

Non-Oil GDP

GDP

5.93 6.90

5.27 7.26

10.2 6.6

9.91 4.84

0.424 6.006

2.1 8.21

5.47 3.41

5.13 3.84

Non-oil GDP per worker 3.4 6.4

Non-oil GDP per population 3.17 6.8

Non-oil GDP per worker 8.0 5.9

Non-oil GDP per population 7.06 6.65

Non-oil GDP per worker 2.67 5.4

Non-oil GDP per population 3.93 5.59

Non-oil GDP per worker 2.31 2.54

Non-oil GDP per population 3.72 3.11

44

44

18

18

11

11

15

15

Mean Standard deviation

Mean Standard deviation Observations

1959–1977 a

Source: BMI. a Constant 1997 prices.

Economic Growth and Economic Policy in Iran: 1950– 2003

Table 1.

271

272

Ahmad R. Jalali-Naini

output and per-capita income grow steadily over the long run; capital stock growth is about the same as output; capital labor ratio grows at a fairly steady rate; the shares of proﬁt and wages in output are fairly stable in the long run. Chenery and Syrquin (1975) detected a general pattern of structural transformation and sectoral shift in the DCs in the process of growth. More recent empirical works (Mankiw et al., 2000; Mankiw, 1995; Klenow and Rodrı´ guez–Clare, 1997; Barro, 2001; Easterly and Levine, 2001) have given rise to a number of other empirical ﬁndings on whether growth is driven by factor accumulation or productivity increases over time, on the effects of HC, national economic policies, and institutions on growth. Among these ﬁndings are: (a) Non-convergence in large cross-country samples: per-capita incomes diverge over the long run among developed countries and DCs because growth rate differentials persist. (b) Growth is highly volatile and not persistent, but factor accumulation is persistent. (c) TFP (or more appropriately the ‘‘residual’’) rather than factor accumulation accounts for much of the growth differentials among countries. (d) National policies and good institutions have a signiﬁcant effect on economic growth performance. For the speciﬁc case of economic growth of Iran, we elaborate the changes in the ‘‘grand’’ ratios and other stylized facts in the following paragraphs. 3.1. Grand ratios and structural change In the course of economic growth capital per worker and output per worker grow steadily over time. In a standard production function setting, higher capital per worker and (Hicks-neutral) technical progress are the main determinants of output per worker. The data indicates that the main determinant of output per worker in Iran (as measured by real non-oil GDP per non-oil worker) is capital per worker and the contribution of technical progress to higher output per worker is positive but very little. Figure 3 shows that the indexes of capital per worker and non-oil output per worker during the 1959–2003 period have similar trends. The long-run comovement between these two non-stationary variables is empirically veriﬁed by the existence of a cointegrating vector between the log of output per worker and the log of capital per worker.3 Output per worker increased at a quicker pace than capital per worker during 1964–1974. This trend was reversed during the 1975–1987 period.

3

The trace statistics test indicates the existence of a cointegrating vector between the two variables at the 1 percent level for the 1959–2000 period. The trend variable capturing the effect of TFP or technical progress is positive, but has a very low value of 0.0016.

Economic Growth and Economic Policy in Iran: 1950– 2003

273

Figure 3. Capital per Worker 2.0 1.8 1.6 1.4 1.2 1.0 0.8 60

65

70

75

80

85

90

95

100

Index of the log of Real Non-Oil GDP per Worker (1959=1) Index of the log of Real Capital per Worker Source: Calculated from data management and plan organization data

Another set of empirical observations in DCs relates to the structural transformations that an economy undergoes in the process of growth. In the early and intermediate stages of economic growth, such countries undergo signiﬁcant and rapid structural transformations. A number of studies (Kuznets, 1971; Chenery and Syrquin, 1975; Kongsamut et al., 1997) suggest that structural changes accompany growth and that the pattern of structural change is dependent on the type of resource endowment, the size of the economy, population, and the type of development strategy. The multicountry empirical study by Chenery and Syrquin (1975) on the morphology of growth indicates that the pattern of sectoral change is fairly similar and on this basis it attempted to delineate universal patterns of development. Some of the prevalent economic changes that occur with higher income per-capita are: a decline in the share of agriculture; an increase in the share of output and employment in industry and services; a sustained increase of the ratios of saving, investment, and government (tax) revenue to GDP; and a large increase in the share of industrial exports in total exports. For primary commodity-oriented economies (also oil exporters) changes in the sectoral composition of output, tax revenue/GDP, and the share of industrial exports in total exports deviates from the ‘‘normal’’ path. For the oil-exporting countries the service sector is larger and the manufacturing sector is smaller than the average for non-primary DCs within the same income class. The presumption that predicted changes in the sectoral composition of output was indicative of a more balanced growth path or that countries that followed the

274

Ahmad R. Jalali-Naini

‘‘normal’’ pattern could create and/or sustain growth momentum was later dismissed. Cross-section growth studies could only support the association between growth and saving/investment rates and not between the sectoral composition of output and growth. Consistent with the above-mentioned empirical ﬁndings, agriculture’s share in GDP has shrunk since the early 1960s and the share of the industry and services sector has increased owing to differences in growth rates. The real volume of value-added in the industry and mines sector outgrew other sectors during the pre and post revolution periods. The growth rate of value-added in the industry and mines sector was quite high during the 1960–1977 period. None the less, the share of the manufacturing sector in GDP in Iran relative to that for middle-income countries was lower during these periods. Moreover, the share of industrial exports in total exports in Iran has always been signiﬁcantly less than the average for the middle income countries. During 1960–1977 period, the value-added in the oil sector outgrew GDP, consequently its share in GDP rose, particularly during the oil-price jump of 1973–1974. In contrast, after 1977 owing to a permanent decline in crude oil production and a reversal of the oil terms-of-trade during 1985–1999, the oil sector’s share in GDP declined signiﬁcantly. In the 1990s the share of nonoil GDP rose and reached the mid-1960 levels. Table 2 shows the sectoral growth rates for four distinct periods. Agriculture experienced a growth rate of 4.35 percent per annum and had the lowest growth volatility throughout the 1960–2003 period. Growth rates in the oil sector have been the most volatile. 3.2. Persistence of factor accumulation and divergence in growth rates4 Growth rates differ signiﬁcantly among countries and regions for an extended period of time owing to a host of reasons. The growth rate of GDP in Iran (and also in the MENA countries) varied substantially during the last four decades: it accelerated during the 1960s and the 1970s, but decelerated during the 1980s and the 1990s.5 This was most notable for the oil exporters in spite of huge terms of trade improvements and large-scale wealth transfers to the oil-exporting countries between 1973 and 1983. Figure 4 and Table 3 show real GDP per-capita growth rates for Iran, East Asia and Paciﬁc, South Asia, middle-income countries, and the world for the 1960–2004 period. Owing to the lack of data for the

4

Simple versions of ﬁnite-growth models (Solow and Swan) predict an inverse relationship between per-capita income and growth rates. Endogenous growth models and multiple equilibrium growth models are capable of providing theoretical explanations for the persistence of growth rate differentials or persistence of per-capita income gap between countries. 5 Klenow and Rodriguez-Clare (2004) show that during 1975–2000 the world economy and a number of its regions experienced a growth slowdown. For instance, growth rates of Asian countries were 3.2 and 2.8 percent during 1960–1975 and 1975–2000, respectively, with the ﬁxed investment rates of 14.5 and 19.9 percent in the corresponding periods.

Economic Growth and Economic Policy in Iran: 1950– 2003

Table 2.

275

Sectoral growth rates, selected periods (in percentages) Agriculture

Industry and Mines

Oil

Services

GDP

1960–1977 Mean Standard deviation Coefﬁcient of variation Observations

4.29 4.85 1.13 18

14.21 11.02 0.77 18

9.7 8.35 0.86 18

11.52 8.7 0.75 18

9.91 4.84 0.48 18

1978–1988 Mean Standard deviation Coefﬁcient of Variation Observations

4.7 2.67 0.56 11

0.44 8.94 20.3 11

0.83 48.2 58.07 11

2.05 7.26 3.54 11

2.1 8.21 3.90 11

1989–2002 Mean Standard deviation Coefﬁcient of variation Observations

4.15 5.39 1.29 14

8.08 9.09 1.12 14

2.47 8.20 3.31 14

5.01 2.67 0.53 14

5.0 3.96 0.79 14

1960–2002 Mean Standard deviation Coefﬁcient of variation Observations

4.35 4.50 1.03 43

8.46 11.34 1.34 43

5.08 24.86 4.89 43

5.93 8.70 1.46 43

5.24 7.34 1.40 43

Note: sectoral growth rates are based on constant 1997 price series. Source: calculated from National Income data published by the Central Bank of Iran.

earlier years, the coverage for MENA is limited to the 1974–2004 period.6 During the 1960–1977 period Iran overmatched growth rates in East Asia, South Asia, the middle-income countries, and the world. However, in the post-1974 period Iran’s growth rates were below those for East Asia and Paciﬁc, South Asia, the middle-income countries, and MENA. The threshold investment rate for high GDP growth per-capita economies is 22 percent (Ghosh and Wolf, 1998) and the average GDP growth rate among this group was above 4 percent. The average ﬁxed capital investment per unit of GDP (both in current prices) in Iran is higher than this threshold: for the 1959–2003 period it was 27.5 percent (Figure 5), and those for the ﬁrst, second, and third subperiods were 28.04, 27.79, and 26.59 percent per annum, respectively. The growth rates for these corresponding periods were 9.91, 2.1, and 5.1 percent, respectively, and were not strongly correlated with period-speciﬁc gross investment rates. On a percapita basis, the growth rates for the corresponding periods were: 7.06, 3.93, and 3.72 percent per annum, respectively. The above numbers are consistent with the proposition that ‘‘growth is not persistent, but factor accumulation is’’ (Easterly and Levine, 2001; Klenow and Rodriguez–Clare, 2004). Moreover, in

6

During 1963–2003 East Asia and Paciﬁc outperformed per capita GDP growth of Iran by close to 3 (and of the oil-exporting countries by >3) percent per annum.

Ahmad R. Jalali-Naini

276

Figure 4. GDP growth rates 20 16 12 8 4 0 -4 -8 -12 -16 65

70

75

East Asia MENA

80

85

90

95

100

World Iran, constant local currency Source: World Bank and BMI

Table 3.

Mean Standard deviation Period

Average GDP growth rates in selected regions and Iran (percent) Asia and Paciﬁc

South Asia

Middle Income

World

MENA

Iran PPP ($)

Iran*

Iran NonOil GDP*

6.798 4.16

4.68 2.47

4.22 1.80

3.63 1.55

3.79 3.18

2.77 7.14

5.27 7.18

5.92 6.83

1960–2004

1960–2004

1960–2004

1960–2004

1974–2004

1974–2004

1959–2004

1959–2003

Note: all series are in constant (2000) dollar PPP, except for the last two columns marked with * which are in 1997 prices, local currency. Source: World Bank Development Indicators and BMI.

contrast to AK class of models, the period growth rates were not increasing in the investment rate. Fixed investment/GDP rates were close to their long-run average in the second period, but growth rates collapsed owing to the effect of the revolution, the Iran–Iraq war, and inappropriate economic policies (see Sections 5 and 6). In the third (post-war period 1989–2003), the ﬁxed investment/GDP rate decreased by 1.2 percent compared to the war period, but economic growth picked up signiﬁcantly, by nearly 7.3 percent. Higher rates of investments ﬁnanced by oil revenues and distributed by the government have raised capital/output ratios during oil booms, but the effect of this on growth has been counterbalanced by lower investment efﬁciency (higher incremental capital/output ratio) – owing to, among other things, the lack of a

Economic Growth and Economic Policy in Iran: 1950– 2003

277

Figure 5. Gross Fixed Investment Rate in Iran 44 40 36 32 28 24 20 16 60

65

70

75

80

85

90

95

100

Gross Fixed Investment per unit of GDP (%) Flexible Trend (%) Source: BMI

developed ﬁnancial market and budgetary rent seeking. As a consequence, the huge wealth transfer and positive government-saving boost from higher oil prices during 1974–1985 in Iran and in most other oil-exporting countries did not result in a signiﬁcant and permanently higher per-capita income level or a higher growth path – as implied from either conventional (Solow–Swan) or endogenous growth models. There has not been a growth miracle story for the oil exporters, and it does not seem that there will be one in the ofﬁng.

3.3. Growth volatility and shock persistence Economic growth in Iran like most MENA countries and other DCs is unstable and shows signiﬁcant ﬂuctuations. Growth rate volatility in Iran is nearly twice as much as in the industrial countries.7 Unlike South and East Asia, growth in MENA (and Latin-America) has been volatile. Output series and estimates for TFP in the MENA region are twice as volatile as those in developed countries (Senhadji, 1999). These observations are consistent with the ﬁndings of Easterly et al. (2000) that the volatility of growth for DCs is much higher than that in the OECD countries.8

7 8

For details see Jalali-Naini (1999). They also found that employment and real wages are also more volatile in the DCs.

Ahmad R. Jalali-Naini

278

Table 4. Country

Measures of persistence for GDP series in selected MENA countries Egypt

Iran

Morocco

ARIMA 1.89 (1,1,0) 2.41 (1,1,0) 0.641 (0,1,1) Spectral density 1.75 2.68 0.264

Saudi Arabia 1.44 (1,1,0) 1.98

Tunisia

Turkey

0.258 (0,1,2) 0.905 (1,1,0) 0.256 0.94

Note: the sample period for Iran is 1959–2000, and for other countries is 1960–1999. Source: calculated from real GDP series, World Bank.

Aside from volatility, economic growth in Iran shows a signiﬁcant degree of shock persistence. The estimates of persistence, based on spectral density and auto regressive integrated moving average (ARIMA) are given in Table 4. The order of ARIMA and the associated measures of persistence for Iran and some selected countries are presented.9 The estimates of persistence based on spectral density and ARIMA for Iran during the 1959–2000 period are both substantially higher than unity – which is the measure of persistence for a random-walk process. These imply persistence of growth shocks; a pick up in the growth rate is followed by movements in the same direction.10 What could be behind shock persistence? In the mainstream macro models (excluding RBC) nominal shocks are presumed to have only temporary effects on output, though nominal or real rigidities can prolong their effects. Easterly et al. (2000) argue that, given that real wages are more ﬂexible in DCs, the orthodox Keynesian and monetarist interpretation that monetary shocks have a long-lasting effect on output growth cannot be maintained. In their opinion, the shallowness of ﬁnancial institutions rather than incomplete wage ﬂexibility in the standard macro models is the cause of excess output volatility.11 Jalali-Naini and Naziﬁ (2002) provide indirect support for this view by showing that the effects of positive as well as negative nominal shocks on output die out quickly and there are

9

The results of the ARIMA speciﬁcation with the highest value of Akaike Information Criterion and Schwarz Bayesian Criterion are reported here (Pesaran and Pesaran, 1997). In most cases, a ﬁrstorder autoregressive process in the ﬁrst difference of the log of GDP (ARIMA 1,1,0) gave the best ﬁt. In Morocco and Tunisia a moving average process was selected. 10 Some words of caution are worth mentioning here. Given that our sample of observations is limited, the statistical problem of determining long-run characteristics (e.g. permanent component to output ﬂuctuations) from ﬁnite samples exists. The evidence cited here cannot be taken as the evidence of the existence of a permanent component – even though the evidence is consistent with such an interpretation. Given the length of our sample, the evidence of shock persistence in Iran can also be consistent with a mean reversion story, that is, a slow reversion of output to a deterministic trend. What can be established with the evidence produced here is that, even if trend reversion cannot be rejected owing to ﬁnite data, trend reversion does not occur quickly. 11 Countries can smooth shocks with greater capital account openness and this might result in less growth volatility. Greater openness, however, might increase volatility through sudden capital outﬂows. In any event, a better access to international capital markets can partially be achieved through a more developed domestic ﬁnancial sector.

Economic Growth and Economic Policy in Iran: 1950– 2003

279

Table 5. Estimates of production function parameters, selected regressions

K L HL RSINTM

I

II

III

IV

V

Phillips–Hansen 0.416

Phillips–Hansen 0.452

OLS 0.519 0.430

OLS 0.479

OLS 0.509 0.420

0.577

0.490 0.159

0.513 0.0005

Note: sample period, 1959–2000.

no signiﬁcant dynamic multiplier effects. If not attributable to monetary variables, persistence of shocks could have resulted from the real variables (e.g. major economic policy shifts, institutional instability, and terms of trade shocks).12 4. Sources of growth: factor accumulation or productivity Growth studies have been interested in ﬁnding out whether per-capita income differentials across countries are due to different saving and investment rates in physical capital or differences in productivity growth and HC accumulation.13 Some cross-country studies show that capital formation is the main driver of growth (Mankiw, 1995; Young, 1995), but others (Klenow and Rodrı´ guez–Clare, 1997; Easterly and Levine, 2001) argue that it is TFP. In this section, we attempt to provide an answer to this question by examining the sources of growth in Iran and use standard methods to examine the relative importance of capital accumulation and TFP in the growth process. Several forms of aggregate production functions with slightly different factor speciﬁcations were estimated for Iran. The Cobb–Douglass function yielded superior results and the estimated coefﬁcients from ﬁve different regressions covering 1959–2000 are reported in Table 5. Each column in the table speciﬁes the estimation methods used: Phillips–Hansen, 1990 (PH) and ordinary least squares.14 The inputs in the production function are physical capital stock (K), employed workers (L), and in regression I, III, and IV the number of workers

12

Makdisi et al. (2005) point to the signiﬁcance of the terms of trade shocks in the Arab countries. These factors inﬂuence growth via a higher pace of innovation, quicker adaptation, creation of stronger incentives for investment, and improved resource allocation. 14 The estimated parameters of regression I in Table 5 are a PH estimate of a Cobb–Douglass production function for non-oil GDP (value-added). The PH method is appropriate for estimation when a single cointegrating vector exists between a set of I(1) variables. Augmented Dicky Fuller test determined that the elements of the production function are I(1). A cointegration test indicated the existence of a single vector between them. 13

280

Ahmad R. Jalali-Naini

weighted by their HC.15 The capital stock was estimated using perpetual inventory method.16 We follow the general practice in the literature by taking a stock measure of education, that is, the average educational attainment of the labor force, as the proxy for HC.17 A period dummy variable is included to capture the break in the production function owing to the war and revolution in the post-1977 period. In regressions II and III we added the deviation of real imported intermediate goods from trend as a capacity adjustment factor (a shift variable) to improve the ﬁt.18 When lower oil revenues and foreign currency shortages result in import rationing, imported intermediate goods/raw materials are below trend, thereby reducing the capacity utilization rate of the production units that have insufﬁcient inputs and materials for further processing. In these periods the shift factor assumes a negative value. Different speciﬁcations and estimations resulted in slightly different factor elasticities and TFP estimates.19 The elasticity of GDP with respect to the capital stock for the post-revolution period (1978–2003) was also estimated; its value is close to the average of the estimates in Table 5.20 The impact of HC in the growth accounting exercise was tested. First, HC was included as a separate factor of production. However, the coefﬁcient of HC was not statistically signiﬁcant. Inclusion of HC as a weighting factor for labor (educationaugmented labor) yielded better results. In regression I, the elasticity of output with

15

Lucas (1988) developed an intertemporal optimizing model that incorporates externalities (spillovers) originating from HC. A simpliﬁed Lucas-type production function is:

yt ¼ AK at ½uht 1a where y is output per worker, k the capital per worker, u the fraction of time devoted to production of goods, and (1-u) is the time allocated for production of h (HC per worker). The increment to capital stock is equal to savings. An estimate of this function for 1966–2000 yielded a ¼ 0.518, slightly larger than the capital elasticity obtained in Table 5 regressions, but the estimated elasticity of HC per worker was not statistically signiﬁcant. 16 The gross ﬁxed capital formation series used for estimating K is based on the Central Bank data. 17 Empirical studies have used a variety of proxies for education: enrollment ratios, the ratio of educated adults to the adult population, and the average years of schooling, that is, past educational attainment as a measure of current stock of HC. The estimated time series for HC stock in Iran was extracted from the Iranian household budget surveys for the 1972–1998 period and was extended forward for 1999 and 2000. 18 The production function speciﬁcation here is of the following form: Y t ¼ At ½K a ðH t Lt Þ1a f ð:Þ where f(.) measures the factors that cause a shift in the production function aside from productivity shocks. It has a general functional form; the speciﬁc form is ascertained through empirical estimation. 19 The homogeneity condition was not imposed, but the estimated elasticities are very close to 1. 20 The estimated value is 0.483 and it was obtained from a simple intensive form of a Cobb–Douglass function. Note that, the reliability of the estimates is dependent on the quality of the data, which is not very high.

Economic Growth and Economic Policy in Iran: 1950– 2003

281

respect to HC-weighted labor is 0.57 and with respect to capital it is 0.41. Based on the results reported in Table 5 we calculated the contribution of capital, labor, and TFP to growth. The estimates indicate that TFP for the 1959–2000 period did not contribute to the economic growth in Iran. When the growth rate of HC is also taken into consideration in the calculations, the contribution of factor inputs to growth exceeds 100 percent, i.e. TFP contribution was negative and the growth process was factor-intensive. If we disregard HC and consider only simple labor (estimates III and V), TFP contribution to growth becomes positive, but is still limited. In this case, the only periods when TFP contributed to growth were the 1961–1972 and 1989–2000 periods. The oil boom of the early to mid-1970s resulted in a government-led investment boom, overaccumulation of capital, and a short period of capital-intensive growth. TFP growth during the 1973–1977 period was negative and declined further during the 1978–1988 period. Solow residuals from the growth accounting equations Granger-cause output, but not the reverse, indicating that the growth process (higher levels of economic activity) does not affect productivity. This could imply the absence of learning by the economic units in the growth process. The TFP estimate from regression I (Table 5) is regressed on a selected list of variables presumed to inﬂuence productivity and the result is reported in Table 6. These include measures of macroeconomic instability and unbalanced macroeconomic policies such as the rate of inﬂation (Fischer, 1993), the size of the government, a measure of openness (deﬁned as exports plus imports as a ratio of GDP),21 and imported intermediate goods – to separate the effect of import rationing from TFP estimates. The rate of inﬂation and the government size negatively inﬂuence TFP – as might be expected. The coefﬁcient of openness is also negative. Since the most important factor that determines the conventional measures of openness in Iran is the magnitude of oil revenues – it inﬂuences the magnitude of exports as well as the capacity to import – the increase in this ratio could be due to oil price variations and does not reﬂect efﬁciency-induced gains in exports. Output was included as an additional regressor to see if growth, hence learning, inﬂuences TFP. However, the estimated coefﬁcient was insigniﬁcant indicating no inﬂuence, and hence was dropped. 4.1. HC and TFP Since a large fraction of government resources in Iran is spent on publicly funded education, evaluating the effect of HC on growth is important. The average annual growth rate of HC in Iran over the 1959–2000 period was a fairly strong 3.75 percent and 4.1 and 3.93 percent for 1959–1977 and 1978–1988, respectively. Endogenous growth models (Romer, 1986; Lucas, 1988) rely on

21

Several other measures such as the ratio of non-oil exports to GDP were attempted and none of them produced the expected results.

Ahmad R. Jalali-Naini

282

Table 6. Determinants of TFP Variable

Coefﬁcient

t-Statistic

Log of private investment/non-oil GDP 0.119 4.273 Log of openness 0.050 2.66 Log of intermediate-goods imports 0.0625 4.167 Log of government expenditures to non-oil GDP 0.079 2.474 Inﬂation rate 0.014 2.496 Constant 0.187 1.726 DUM 1976 0.083 3.489 DUM 1981 0.066 2.804 AR(1) 0.492 3.994 R2 0.787 Akaike information criterion 4.3 Adjusted R2 0.732 Schwarz criterion 3.92 Durbin–Watson statistic 2.314 F-statistic 14.36 Note: sample period, 1963–2000.

HC or on knowledge accumulation to deliver unbounded growth.22 Investment in HC can also deliver social externalities to offset diminishing returns to private capital. Macroeconomic treatment of HC either focuses on the growth rate of HC (Lucas, 1988) or its level (Nelson and Phelps, 1966). In Lucas (1988) the accumulation of HC (education) has a positive effect on the long-run growth rate. Thus, differences in growth rates across countries can be attributable to differences in the rates of capital accumulation.23 From a microeconomic perspective we expect individuals with more education to earn more, a proposition well corroborated by many empirical studies in different countries including Iran (Saedi, 2002). The increase in the average years of schooling in Iran since the 1960s has been quite substantial.24 Higher education attainment rates should increase the average incomes, especially if positive externalities are associated with it. However, many empirical studies, particularly in the MENA region, show that HC does not have a signiﬁcant impact on economic growth or on the growth rate of output per worker.25 As a number of writers using growth accounting approach (Benhabib and Speigel, 1994; Islam, 1995) suggest, HC as a separate factor of production, has either zero or negative direct effect on growth, but the main channel through which education inﬂuences growth is through its effect on TFP. Benhabib and Speigel (1994) argue that HC affects productivity through two channels. First,

22

In the scheme of Arrow (1962) learning by doing occurs without R&D expenditure. Learning and knowledge accumulation can also occur as a result of R&D expenditure (Romer, 1990a,b). 23 Mankiw et al. (1992) showed that by adding HC, their Solow (1956)-augmented model provided a better explanation for per-capita income differentials among countries. 24 For more details see. Salehi-Isfahani (2005). 25 Pritchett (1996) raises a challenging question, ‘‘where all the education has gone in MENA?’’

Economic Growth and Economic Policy in Iran: 1950– 2003

Table 7.

283

Determinants of TFP (from regression III, Table 5)

Variable List Inﬂation rate Log(HC) Log(Private investment/non-Oil GDP) Log(Government size/non-oil GDP) Trend R2 Adjusted R2 Durbin–Watson statistic

Coefﬁcient

t-Statistic

0.226 0.517 0.095 0.074 0.018 0.331 0.229 2.15

2.86 2.97 2.75 1.83 2.84 3.85 3.6 3.267

Akaike information criterion Schwarz criterion F-statistic

Note: sample period, 1963–2000.

a` la Romer (1990), HC raises the capacity of countries to innovate new technologies suited to their domestic economy. Second, a` la Nelson and Phelps (1966), HC increases the ability of nations to adopt imported technologies. The higher the level of HC, the higher is the speed with which DCs can catch up with technology leaders. To account for the indirect effect of HC on growth, our index of HC was regressed on the estimated TFP from regression III plus three other variables, including inﬂation and the size of the government.26 The coefﬁcient of HC was found to be positive and statistically signiﬁcant (Table 7), conﬁrming the positive relationship between HC and TFP as indicated by Benhabib and Speigel (1994) and Islam (1995).27 We can conclude by observing that capital accumulation and higher employment were the main drivers of growth in Iran. TFP contribution to growth in the Iranian economy during the 1959–2000 period has been limited – negative if HC is taken into consideration in the calculations.28 There is evidence that HC positively inﬂuences TFP; however, the magnitude of TFP growth has been limited. 5. Economic policy, strategy, and growth performance during 1950–2003 In the early development models lifting of saving and investment rates was the key for higher growth rates. In the dual gap models the binding constraint was

26

Note that in this regression (shown in Table 5) L and not the HC-weighted L is used for estimation. We could not test for the impact of R&D on TFP because we do not have a reliable estimate. It should be mentioned that examination of the three-digit and four-digit ISIC manufacturing data shows that R&D, including on-the-job training expenditures, for the year 1996 was on average about 0.2 percent of total revenue, quite small by any yardstick. 28 The estimates of TFP should be taken with caution. Aside from the usual caveat of low-quality data and variation in capacity utilization rates, we observe that under imperfect competition the Solow residual is a weighted average of TFP growth and the output/capital ratio, and the weights are determined by the market power of the ﬁrms – as measured by the inverse of price-markup. 27

284

Ahmad R. Jalali-Naini

considered to be foreign savings. The pax Britanica pattern of international division of labor-limited less developed countries’ (LDCs) export earnings and imposed a resource constraint on growth. LDCs also suffered from incomplete, missing markets, and the weakness of private institutions. Given the above backdrop, import substitution strategies and activist governments were thought to be the appropriate policy remedies. These ideas ﬂourished in Latin America and for a relatively short period, these policies were embraced by numerous countries in Asia and MENA, including Iran.29 5.1. High growth years: 1959– 1977 Iran exhibited a stylized feature of an oil-based developing country in the 1950s: exporting raw materials (oil) with limited low forward and backward linkages between the oil sector and the rest of the economy. This environment rendered the ﬁscal (revenue) impact of the growing oil exports much more inﬂuential than its intersectoral demand/supply articulation with other parts of the economy. As in nearly all LDCs in their earlier stages of development, pervasive market imperfections and an economically and politically weak private sector gave way to the view that the state should step in directly to replace missing markets, correct market imperfections, and allocate oil revenues. The basic development thinking in Iran since the mid-1950s has been planning a framework in which the oil industry serves as the engine of growth for the economy supplies surpluses (savings) for investment in other sectors. The allocation of oil resources to promote growth has traditionally been the cornerstone of development plans in Iran. The ﬁve-year plans (ﬁve plans prior to the revolution and three after it) reﬂected the sectoral priorities and the general development strategy envisioned by government bureaucrats and policy makers. The bureaucrats who oversee the distribution of (oil) resources found it in their interest to maintain this practice. Import substitution industrialization (ISI) strategy became an additional feature of the policy framework in Iran during the 1960s and the early 1970s – and also in the post-revolution period. Disbursement of oil revenues through the ﬁscal budget contributed signiﬁcantly to the growth of household incomes and aggregate demand. In time, this process created a mismatch between the proﬁle of domestic demand and the structure of domestic production. The rising domestic demand for modern non-durable and durable consumer goods could not be met within the limited domestic production capacity that existed in the early 1960s. This formed the market basis for the ISI strategy during the Third and Fourth Development Plans (1962–1973). Low inﬂation, stable equilibrium exchange rates, and rising oil revenues created a

29

Iran and a large number of DCs had a higher growth rate during the 1960s until the ﬁrst oil-price jump in 1974, than in the later decades. As noted by Easterly and Levine (2001) the median per-capita income in LDCs was 2.5 percent during 1960–1979s but was 0 during the 1980–1999 period.

Economic Growth and Economic Policy in Iran: 1950– 2003

285

macroeconomic environment conducive to high growth rates in the industrial sector and non-oil GDP during the 1959–1973 period. The ﬁrst and second phases of import substitution (Hirschman, 1968) moved along at a good speed in this period. The above non-inﬂationary macro environment was undermined by a huge increase in oil revenues in the 1973–1977 period. The rise in oil revenue was quickly translated into a large upward revision of planned government consumption and investment expenditures in the Fifth Plan (1973–1977) without any justiﬁcation on efﬁciency grounds. The consequent ﬁscal and monetary expansion motivated rent-seeking activity, and further fueled rapid rates of growth of aggregate demand in an already overheated economy leading to the emergence of severe domestic bottlenecks and a rapid pickup of inﬂationary pressures. The oil windfall boosted domestic income and expenditures, but also allowed for an opendoor policy during the Fifth Plan. Rapid increases of aggregate demand and appreciation of the real exchange rate, which made foreign products relatively cheaper to buy, encouraged a huge surge in imports of goods and services. In spite of an active and dominant role by the government, economic policies were on the whole market-friendly, uncertainty over property rights were limited, the private sector was allowed to grow, and the economy became more open. Under this environment the economy during 1959–1976 experienced its highest recorded growth performance. While this policy regime did well on growth it does not get high marks on social justice or on managing the oil boom. Despite a huge increase in oil revenues, particularly after 1973, a comprehensive and workable social distribution program was not implemented. Unfulﬁlled expectations of signiﬁcantly higher living standards for the low and lower-middle income groups during the oil boom fueled political discontent, culminating in the 1979 Islamic revolution. The growth process in this period came to an end partly because of income distribution-motivated political upheavals. 5.2. The revolution and war: 1978– 1987 The economic policy perspectives changed radically in the early post-revolution days. The oil boom of the previous years had fueled distributional and social justice concerns. Nationalization of large/medium private companies and an inward-looking strategy – along the lines advocated by the Structuralists/ Dependency (Frank, 1967; Dos Santos, 1971) – was an approach to grow independently of the ‘‘world capitalist system.’’ This and the goal of improving income redistribution became the dominant view.30 Consequently, a domestic version of

30

In the ‘‘Dependency’’ view obstacles to development are imposed from outside, i.e. are exogenous, and underdevelopment is a consequence of articulation of the domestic economy or the ‘‘periphery’’ with the ‘‘center.’’

286

Ahmad R. Jalali-Naini

the Structuralists’ ‘‘non-capitalist’’ strategy31 for development became the prevalent policy agenda. Greater state control on prices, extensive nationalization of private enterprises and foreign trade, inward-orientation, and expansion of social subsidy programs were the basic features of the new policy regime. Measures to improve income distribution were not conceptualized in the context of ‘‘redistribution with growth’’ and at times were seen as alternatives to growth. The exigencies of the war with Iraq helped to further reinforce these ideas. This broad consensus emerged at a time when ISI and the activist-state model had failed to deliver sustained growth and external balance in a number of DCs. Two main inﬂuences combined to produce an unsatisfactory growth record during the 1978–1988 period. The more obvious was the debilitating effect of the war on the domestic economy and destruction of valuable productive assets and a very large decline in oil revenues during the mid-1980s. Less obvious, but nonetheless signiﬁcant, were the effect of heightened insecurity regarding the property rights, the gradual replacement of the market mechanism and the private sector by ineffective administrative allocation schemes, and poorly managed public enterprises. Overbureaucratization of the economy, important instances of which were administrative pricing and central allocation of goods, credit, foreign exchange, and imports took hold. Suppressing the prices of goods, credit, and foreign exchange below their market values, coupled with government-directed allocation, encouraged a vast increase in rent-seeking opportunities. This price structure made it more proﬁtable to allocate efforts to obtain inputs, foreign currency, and credit at subsidized rates than to allocate money and effort for innovation and efﬁciency. The internal and external real exchange rates signaled resources out of tradables while, owing to the war, the quantity of oil exports declined. The budget deﬁcits grew, partly because of government ﬁnance of war-related expenditures. Higher risks for the private sector resulted in a reduction of private investment and chronic capital ﬂight. Under these adverse conditions, total private sector investments per unit of GDP declined and its composition changed; private sector investment in machinery and equipment decreased relative to the share of private investment in construction. As in bureaucratized economies, the economy tended to end up in a low-effort, lowgrowth equilibrium (Brixiova and Bulii, 2001) in this period. With the collapse of oil revenues in 1986, the continuation of the above policy package was no longer possible; low oil revenues created an economic environment favoring wider private sector participation and greater efﬁciency. As a reﬂection of the coming changes, a partial nominal devaluation was implemented in 1986 to form a multiple exchange rate system, raising the rates for exporters.

31

A strategy in which the state assumes a dominant and leading role in resource allocation and through state-owned enterprises in the production process. However, unlike the Soviet and Eastern European countries’ system, the state allows a limited role for the local private sector.

Economic Growth and Economic Policy in Iran: 1950– 2003

287

5.3. Post-war economic growth and policy shift: 1988– 2003 In the years following the Iran–Iraq war, reconstruction and economics became a more pressing social and political issue. Per-capita incomes and the investment rates had declined precipitously with the decline in oil revenues during 1985–1987 (Figure 5). The heavily state-controlled war economy had become an obstacle to growth. Implementation of measures to control growth of government expenditures and to stimulate private sector investment and exports had become inevitable. Two contending views on how to manage the post-war economy had emerged: those who wanted a more gradual, inward-oriented, government-controlled approach to this transition, and the ‘‘reformists.’’ The reformers wanted less government control, more reliance on the market and price mechanism, wider private sector participation, and trade/exchange rate liberalization for transition to a more open economy. These policies were similar to the structural adjustment policies which were in vogue at the time and which a large number of DCs had implemented. The reformers’ views prevailed and were immersed in the First (1989–1993), Second (1995–1999), and Third Development (2000–2004) Plans. The issues faced by the policy makers and planners were clear. (1) A rapidly growing young population with several years of repressed demand needed more consumption, educational opportunities, and jobs. However, the domestic production capacity to accommodate that higher consumption demand was highly inadequate. This called for investments in infrastructure and production capacity. (2) Domestic saving was insufﬁcient to ﬁnance the required investment in infrastructure and equipment. A big chunk of the government budget and state-owned banks’ credit was allocated to the poorly performing public enterprises. Public enterprises did not generate proﬁts and did not contribute to government revenue. (3) Oil and non-oil exports had declined signiﬁcantly (Figure 6), resulting in chronic foreign exchange shortages and creating a major constraint in ﬁnancing capital goods imports. The capital goods import/GDP ratio had fallen signiﬁcantly (Figure 7). (4) There were substantial price distortions in the economy, especially a highly overvalued exchange rate, negative real interest rates, and low energy prices.32 (5) Prior to the First Development Plan, the price of energy had fallen in real terms. While the nominal price of energy carriers were allowed to rise in the Second and Third Plans, the real prices rose very moderately. In 1997 total energy subsidies amounted to 18.1 percent of GDP and in 2002 it was $16.9

32 See Karshenas and Pesaran (1995) and Pesaran (2000) for the reform issues and the extent of subsidies for other major products.

Ahmad R. Jalali-Naini

288

Figure 6.

Exports

28000 24000 20000 16000 12000 8000 4000 0 60

65

70

75

Oil Exports (in 000 US $)

80

85

90

95

100

Non-Oil Exports (in 000 US $) Source: BMI

Figure 7.

Imported Capital Goods per Unit of GDP

4

3

2

1

0 40

45

50

55

60

65

70

Imported Capital Good/GDP Source: calculated from data on GDP and imports series (BMI)

75

Economic Growth and Economic Policy in Iran: 1950– 2003

Figure 8.

289

Short and Long-term Deposit Rates

10

0

-10

-20

-30

-40 60

65

70

75

Five-Year Deposits (%)

80

85

90

95

00

One-Year Deposits (%)

Note: Prior to 1988 only intermediate-term deposit rates, calculated from BMI data on deposit rates and prices

billion or >14.5 percent of GDP.33 The ofﬁcial exchange rate had remained unchanged for a long time despite a relatively higher domestic inﬂation rate. Consequently, the black-market premium in 1989 had increased to 2000 percent (Pesaran, 1992), creating and encouraging vast rent-seeking opportunities and activities. A highly overvalued domestic currency had drawn resources out of tradables, and domestic production at the prevailing exchange rates was no longer proﬁtable in many sectors. (6) Domestic real interest rates, particularly on short-to-intermediate certiﬁcates of deposits, were negative (Figure 8). Moreover, like other ‘‘repressed’’ ﬁnancial markets, banking sector resources were widely used for quasi-ﬁscal operations. Moreover, state-directed (and subsidized) credit programs had done little to spur private investment (Jalali-Naini and Khalatbari, 2003). (7) Foreign savings (current account deﬁcits) of the mid-1980s were largely used to ﬁnance government current expenditures or the war efforts. Rationing the dwindling oil revenues was sufﬁcient neither to signiﬁcantly increase the domestic investment rate nor to recapitalize public enterprises and to ﬁnance reconstruction of the economy, including the oil sector.

33

For more details see Iran Energy Report (2005).

290

Ahmad R. Jalali-Naini

The options for the economic planners were to: (a) Mobilize and generate savings in excess of what the existing heavily statecontrolled economic structure could deliver. This required the creation of incentives through price signals for additional supplies of capital and conservation of valuable resources such as energy. Subsidized energy prices had (and since then have) induced a high rate of growth of fuel consumption and its cost was ﬁnanced out of government savings. (b) Reduce the degree of real exchange rate misalignment to manageable proportions to economize its use in domestic production and consumption of imported goods and shifting relative prices in favor of tradables. Moreover, some restrictions on exported commodities were lifted. These measures were meant to encourage movement of capital and labor into non-oil exports and domestic production of import competing goods, hence increased supply of much needed foreign exchange by the private sector. However, no comprehensive measure to lift protection of domestic industry was attempted. Domestic protection emanated largely from a quantitative import control system, which cut imports during 1993–2000 by more than one-third of the levels in 1990–1992. Import tariffs were of secondary importance since the average level of tariffs in Iran was not high compared to that of the other DCs.34 (c) Reform and implement measures to develop the domestic ﬁnancial sector. (d) Resort to foreign borrowing as another avenue through which additional resources could be mobilized. External borrowing was exercised though not managed efﬁciently. To remove tradable price distortions, several rounds of partial devaluation were implemented to unify various currency rates around the free-market rate during 1990–1993. These attempts were largely unsuccessful because they were inconsistent with the expansionary ﬁscal stance and low foreign currency reserves of the Central Bank. Put differently, the expenditure-switching policies (devaluation) were not backed with expenditure-reducing (monetary and ﬁscal) policies to achieve a real depreciation. Falling oil prices, foreign debt repayment, and limited access to international ﬁnancial markets, in the absence of the appropriate ﬁscal measures rendered the inﬂation tax as the balancing variable in the system.35 The inﬂation rate picked up signiﬁcant momentum in 1993 and reached its height in 1995 (Figure 9). As a consequence, a real depreciation was not achieved by the exchange rate uniﬁcation policy in 1992 and the Central Bank had to implement several rounds of additional nominal devaluations. The subsequent pickup in the inﬂation rate turned the social and political

34

A major opening up of the economy for imports materialized only as a result of a huge increase in oil revenues during the 2002–2005 period. 35 Fiscal measures to boost public savings are the most effective method to raise national savings (Easterly et al., 1994).

Economic Growth and Economic Policy in Iran: 1950– 2003

291

Figure 9. Inflation Rate Trends 50 40 30 20 10 0 1960

1965

1970

1975 1980 1985 1990 Inflation Rate (cpi, annual %)

1995

2000

Source: calculated from CPI index (BMI)

environment against the reform package, and the process was stalled. With the election of a new president, the critics of the reform strategy gradually took charge of the majority of economic policy-making organizations. However, falling oil prices and economic realities forced the new group to opt for the same reform agenda. The Third Plan was based on a low oil price estimate of around $12 per barrel. Consequently, it called for privatization, exchange rate uniﬁcation, elimination of quantitative trade restrictions, reform of the ﬁnancial markets, privatization of banks, and more private sector (domestic as well as foreign) investment to achieve external balance and higher growth. The general economic policies in the post-war period became more marketfriendly compared to those in the 1978–1988 period; the relative price and exchange rate distortions were partially corrected, incentives for wider private sector participation became stronger, and the extent of property rights’ uncertainties diminished. In this environment growth performance improved. Persistence of relatively low oil prices until the early 2000 was one reason policy makers pushed for greater economic efﬁciency, higher non-oil exports, and greater private sector investment. While government investment outpaced private investment, mainly owing to higher costs of acquisition of capital goods and crowding out of private investment by an expansionary ﬁscal policy in the mid1990s, private investment rose signiﬁcantly since 1996 as a share of aggregate ﬁxed capital investment (Figure 10).36 Note that, in general, the government’s

36

One should bear in mind that some of the private companies are ﬁrst- or second-generation companies created by some government entity. There is no data to separate the purely private ones from those whose lineage goes back to some government organization or ministry.

Ahmad R. Jalali-Naini

292

Figure 10.

Government Share of Fixed Capital Investment

70

60

50

40

30

20 60

65

70

75

80

85

90

95

100

Share of Government in Aggregate Fixed Investments %

Source: calculated from GNE data (BMI)

share of aggregate investment in the Iranian economy has been large in the preas well as post-revolution years. The shares were 41.9, 45.5, and 37.2 percent in the ﬁrst, second, and third periods, respectively. Although growth performance improved in the post-war period, no major progress was achieved to improve income distribution, in spite of huge subsidies paid out by the government. The reform policies mentioned above have been relegated and distributional issues and extension of subsidies have assumed a higher priority since 2003 owing to a signiﬁcant and sustained increase in oil revenues. Large oil price changes usually pre-date major shifts in economic policies in Iran. 5.3.1. National economic policy variables and growth: the evidence One way to model the inﬂuence of economic policies is to relate them to growth via their inﬂuence on TFP. This was the approach used in the growth accounting section earlier.37 An alternative is to use a number of institutional and policy factors that inﬂuence growth performance as growth correlates. We follow the literature on the growth correlates (Barro, 1995, 2001; Easterly and Levine, 2001). They include ﬁscal variables, the exchange rate, ﬁnancial development, measures of economic openness, proxies for the effect of macroeconomic imbalance and uncertainty, and institutional variables. In the case of Iran, the magnitude of oil revenues could be an important growth correlate. Owing to the

37

A more general approach is to model how institutions, markets being an important subset, affect the growth process. However, there is no comprehensive growth model to integrate even a subset of these variables in a uniﬁed framework.

Economic Growth and Economic Policy in Iran: 1950– 2003

293

lack of a sufﬁciently long time-series data on the quality of institutions in Iran that can be used to measure their impact on economic growth, we need to make special allowance for the effect of government policies and behavior, as the dominating social institution, on growth. Fiscal variables are the most reliable set of time-series data available to proxy for government policies. This is also warranted by the fact that the government’s command over resources, aggregate savings, and foreign exchange revenue in Iran is quite substantial. Our empirical model for GDP growth is: D ln yt ¼ a þ bfln yt1 ½a þ bðt 1Þg þ f0 X t þ t

ð1Þ

where y is the (non-oil) GDP per worker, X a vector of state variables, a+bt the deterministic trend growth, and the term in the brackets shows the deviation of output from trend growth in the previous period. If per-capita GDP tends to revert toward the trend, b is negative, else, it is equal to zero. The above equation can be rewritten as D ln yt ¼ g þ ct þ b ln yt1 þ f0 X t þ t

ð2Þ

where, g ¼ aba þ bb and c ¼ bb. Testing the restriction that b ¼ 0 means that GDP per-capita does not revert toward the mean, and if bo0, then per-capita output is trend reverting. If b ¼ 0 and if we ﬁnd that the state variables included in the vector X are signiﬁcantly correlated with output growth, the implication is that they have a long-lasting (‘‘permanent’’) effect on output. If bo0, then their effect on growth is temporary. Based on our previous discussion, we know that GDP series exhibits shock persistence and hence is not trend reverting. However, as shown in Figure 2, after 1976 both GDP and non-oil GDP per-capita seem to be mean reverting. The state variables in II are listed and described below. Oil, non-oil GDP, and fiscal variables. There are two approaches to capture the inﬂuence of oil. One way is to directly test for the effect of oil, e.g. via a Granger causality test. The growth rate of the oil sector can inﬂuence non-oil GDP growth rate, but changes in the value-added in the oil sector are driven by exogenous factors. Therefore, we expect that the Granger-causality runs one way from the oil sector to non-oil GDP. This is conﬁrmed by the F-test given in Table 8 for the 1959–2002 period. Another way to test the inﬂuence of oil is through its intermediate links with the economy. The effect of oil revenues in Iran is ﬁrst and foremost transmitted via the government budget which, in turn, inﬂuences economic growth through: (1) government investment and the size of the government and (2) the capacity to import investment goods. These inﬂuences enter as state variables in (II). Tables 9 and 10 show the results of the estimate of (II) for the 1960–2002 and 1978–2002 periods, respectively. The coefﬁcient b is negative and statistically signiﬁcant, particularly during 1978–2002, indicating reversion to a trend value of non-oil GDP per worker. The deterministic time trend is signiﬁcant in the 1960–2002 period, but not signiﬁcant at 5 and 10 percent levels for 1978–2002. Hence, it was omitted from the regression.

Ahmad R. Jalali-Naini

294

Table 8. Null hypothesis, one-lag Oil sector growth rate does not Granger cause non-oil GDP growth rate (1960–2002) Non-oil GDP growth rate does not Granger-cause oil sector growth rate (1960–2002)

Oil and non-oil GDP Granger-causality F-Statistics

Probability

4.476

0.040

0.419

0.521

Source: based on BMI data.

Total government expenditure per unit of GDP, a proxy for the effect of the relative size of the government, was included in the growth regression. The sign of this variable was negative, but was insigniﬁcant, hence dropped from regressions in Tables 9 and 10.38 The composition of government spending has been argued to have a signiﬁcant effect on economic growth (Tanzi and Zee, 1997). When expenditures are disaggregated, there is stronger evidence of a negative relationship between public consumption and growth than in the case of public investment. A number of studies (Aschauer, 1989; Knight et al., 1993) show that public investment positively affects growth. Empirical evidence cited in Barro (1991) points to a weak correlation between public investment and growth. He interprets this as indicating either that government investment is not a signiﬁcant determinant of growth or that governments are optimizing and investing up to the point where the marginal effect of such investments on growth is close to zero. There is also the more standard interpretation that, owing to a host of incentive and efﬁciency problems, public investment has a lower growth effect than private investment. To test for the effect of public investment, aggregate investment is broken into private and public investments. The coefﬁcient of private investment/GDP, as expected, is positive and highly signiﬁcant, and greater than that for government investment, in both the sample periods. The ratio of public investment to GDP is positive and signiﬁcant during both the sample periods.39 Easterly and Rebelo (1993) argue that government expenditures do not affect growth in the same way and certain categories of government expenditures may have stronger effect on growth. To capture the effect of different types of public investment on growth, government investment was

38

To capture the non-linearity of the effect of government size on growth the square of this variable was also used as a regressor, but it was insigniﬁcant. 39 As shown by Khan and Kumar (1993), private investments are usually more productive than public investment, so policies that result in a large reallocation away from the private sector might affect growth negatively.

Economic Growth and Economic Policy in Iran: 1950– 2003

295

Table 9. Growth rate of per worker non-oil GDP Variable

Coefﬁcient

Method: least squares, sample: 1960–2002 C One-period lagged non-oil GDP per worker Government investment per unit of non-oil GDP Private investment per unit of non-oil GDP Free-market premium Ratio of intermediate goods to non-oil GDP Dummy for 1977 Dummy for 1974 (ﬁrst oil shock) Time trend R2 Adjusted R2 Durbin–Watson statistic

0.719 0.132 0.048 0.087 0.004 0.064 0.094 0.107

t-Statistic

Probability

0.006 0.011 0.070 0.001 0.826 0.002 0.011 0.002

2.904 2.667 1.866 3.431 0.221 3.273 2.664 3.208

0.002 2.161 0.037 0.800 Akaike information criterion 3.91 0.753 Schwarz criterion 3.54 2.042 F-statistic 17.02

Note: data for all the variables are from BMI. All independent variables are in log units.

Table 10.

Growth rate of per worker non-oil GDP

Independent Variables

Coefﬁcient

Method: least squares, sample: 1978–2002 C 1.69 One-period lagged non-oil GDP per worker 0.444 Government investment per unit of non-oil GDP 0.062 Private investment per unit of non-oil GDP 0.082 Free-market premium 0.054 Ratio of intermediate goods to non-oil GDP 0.04 R2 Adjusted R2 Durbin–Watson statistic

t-Statistic

Probability

4.721 4.401 2.856 3.638 4.293 2.212

0.781 Akaike information criterion 0.723 Schwarz criterion 1.83 F-statistic

0.00 0.00 0.01 0.00 0.00 0.03 3.85 3.52 17.44

Note: Data for all the variables are from BMI. All independent variables are in log units.

divided into strategic and non-strategic sectors. The former consists of government investments in transportation, communications, water, and electricity. The estimated coefﬁcients were, positive, but not signiﬁcant. Public spending on education was used as a proxy for government investment in HC. The estimated coefﬁcient for this variable had the expected sign, but was not statistically signiﬁcant and therefore not reported. We included a measure of import compression, the black-market premium (on foreign exchange), and inﬂation to capture the effect of (oil-related) foreign

296

Ahmad R. Jalali-Naini

exchange shortage/balance of payments constraints, price distortions, and macroeconomic uncertainty on growth. The ﬁrst two variables reﬂect the indirect effects of the size of oil revenues. The ratio of imported intermediate goods to GDP was used as a proxy to capture the effect of this class of goods on capacity utilization rates, hence on growth. Imports were converted to domestic currency by the weighted (ofﬁcial and market rates) exchange rate. As indicated by the regression results, the coefﬁcient of this variable is positive indicating that (oilinduced) import contraction negatively inﬂuences economic growth and vice versa. This ﬁnding is consistent with the effect of the shift factor in the production function estimate presented in Table 5. The black-market premium reﬂects disequilibrium exchange rates, price distortions, and higher costs of acquiring investment goods by the private sector, so it negatively inﬂuences growth. During periods when oil revenues and foreign currency reserves are low, as the main supplier of foreign currency the government limited its supply through formal channels and resorted to bureaucratic allocation of foreign currency. Since the degree of foreign capital inﬂow, particularly during the 1978–2000 period, was limited, capital account ﬂows could not compensate for shortfalls in oil revenues. Ofﬁcial dual exchange rates existed in the Iranian economy during 1966–2001. In 2002, the various exchange rates were uniﬁed and because of higher oil prices foreign currency shortages have been absent and the black-market premium has been reduced to nearly zero. The coefﬁcient of the black-market premium is found negative and signiﬁcant in the 1978–2002 period. High rates of inﬂation reﬂect macroeconomic instability and hence economic uncertainty. A number of large-sample studies show that instability and uncertainty negatively inﬂuence growth. The threshold, above which inﬂation signiﬁcantly inﬂuences economic growth for industrial countries, is estimated to be between 1 and 3 percent and that for DCs 7–11 percent (Khan and Senhadji, 2000). The average annual inﬂation rate (measured by CPI) in Iran during the 1974–2000 period was 20 percent per annum. The inﬂation rate (CPI) was included in the regressions and its coefﬁcient was negative, but insigniﬁcant in both sample periods and hence was dropped. The contribution of ﬁnancial development to growth has been discussed extensively and a number of large-sample empirical studies provide evidence that more developed ﬁnancial institutions and ﬁnancial deepening are conducive to growth (Levine et al., 2000). We tested for the impact of ﬁnancial depth, measured by M2/GDP, on growth, but this variable was not statistically signiﬁcant and hence was dropped.40

40

For detailed discussions on the issue of ﬁnancial development and growth and related empirical tests in the case of Iran, see Jalali-Naini and Khalatbari (2003).

Economic Growth and Economic Policy in Iran: 1950– 2003

297

6. Oil, rent-seeking, and economic growth In the standard growth models with perfectly competitive markets, the effects of institutions and policies on growth are reduced to the ﬁrst-order conditions.41 Under this setting, factor accumulation and technical progress are just what are needed to explain growth.42 The bench mark optimal growth model generates results similar to the competitive model when the growth regulator is a benevolent government – a government devoid of self or group interest. Once the benevolence assumption is relaxed, a number of factors such as the quality of institutions, property rights, and rent seeking can affect growth performance. Hall and Jones (1999) and Stigliz (1989) argue that a signiﬁcant proportion of the per-capita income gap between LDCs and DCs can be attributed to qualitative differences in organizations and the institutions through which the individuals interact. Because of a historically weak private sector, missing markets, and monopoly of oil revenue, the government in Iran has had an enormous effect on the matrix of social institutions. The institutional matrix in an economy inﬂuences the allocation rules, development of non-governmental institutions, and economic incentives, hence it affects economic behavior and performance.43 Economic performance in state-led economies has varied quite signiﬁcantly ranging from ‘‘collapse’’ to ‘‘miracle,’’ depending on the institutional matrix and the governance structure. Governments that have undertaken to establish an institutional framework with credible rules; limited economic uncertainty; low transaction costs; efﬁcient provision of public goods; and conducted sensible economic policies, have been successful in promoting high rates of capital accumulation, higher productivity levels, and sustained growth.44 Under this model, the commitment and reputation of the state to uphold the rules of the game provide private agents with a frame of reference to form ‘‘social capital’’ which is progrowth.45 Few populated oil states have been successful in creating this type of

41

The government ﬁnds a place in the competitive model when externalities (learning, knowledge, public goods, provision, y) are taken into account. When competitive markets and prices, as regulators and disciplinarians of the economic process are totally or partially absent, a host of factors, e.g. transaction costs, information asymmetry, and credit rationing can inﬂuence growth. 42 If the differences were due to factor accumulation, then the rates of return on HC and physical capital would be signiﬁcantly higher in DCs and that massive foreign capital should have ﬂown to the poor instead of the rich countries. The evidence is clearly contrary to this pattern of factor reward and capital ﬂows based on differential factor endowments (Stiglitz, 1989). 43 Institutions are social constraints that structure incentives and human interaction. See North (1994) on institutions. 44 For more details, see World Bank Report (1997). 45 For more details, see Ismail Serageldin and Christiaan Grootaert, ‘‘Deﬁning Social Capital: An Integrating View,’’ Joseph Stiglitz, ‘‘Formal and Informal Institutions’’, Richard Rose, ‘‘Getting Things Done in an Antimodern Society: Social Capital Networks in Russia,’’ and Partha Dasgupta, ‘‘Economic Progress and the Idea of Social Capital,’’ in Dasgupta and Serageldin (2000).

298

Ahmad R. Jalali-Naini

economic environment. Gylfason (2001) argues that economic performance in oil-exporting countries is not satisfactory owing to widespread rent seeking, policy failures, and the Dutch disease. These tend to dampen the growth-inducing effects of government spending. A bureaucratic/rentier system limits the allocation of resources through markets and economic transparency. Price distortions in a bureaucratically controlled allocation process imply greater inducement for rent-seeking opportunities and behavior and designing private networks to extract rent become an attractive economic activity. In this environment, extracting beneﬁts from bureaucratic allocations of economic resources dominates the working agenda or the goals of business/formal organizations. ‘‘Social capital networks’’ can penetrate organizations and further affect allocation of resources through favoritism and rent seeking. Instead of following the laws and responding to price signals, rules are sidestepped by personal contacts and politics.46 The payoffs to rent seeking can be very signiﬁcant and in many instances greater than what could be obtained through competition and efﬁciency. Those who distribute rents and the ﬁrms collecting it can form interest groups to oppose reform and contest with outside groups for a share of the beneﬁts. The by-products of such an economic environment can be large unproductive expenditures such as military spending, large government payoffs to groups with political and military clout, and (numerous) government investment programs with low social rates of return (Gelb, 1988) and low TFP.47 Attempts to undertake some reforms in the economic sphere (in the context of the development plans) to foster non-government institutions, including incomplete and missing markets, to correct price signals, to reduce the degree of trade protection, and to improve economic-policy consistency have been made in Iran, but these were not accompanied by the complementary political and institutional reforms. These attempts have been resisted by the vested interests – and sometimes by the social groups that were not the direct beneﬁciaries – particularly during oil

46

See Rose (2000). This kind of model can also result in incredibility (or instability) of the rules Incredibility gives rise to contract uncertainty and low credibility. The effectiveness of the enforcement of contracts is one of the most important determinants of economic performance and a crucial difference between developing and developed economies (North, 1990). Uncertainty and low credibility of economic policies result in lower investment rates and adoption of costly strategies by the private sector. Countries with higher credibility tend to have a higher investment rate (as a percentage of GDP) and a higher growth rate (World Bank Report, 1997). As shown in Tabibian (2002) little or no growth in the average size of the private ﬁrms which can be due to economic uncertainity and property rights’ concerns. Salehi-Isfahani (2002) describes the relationship between contract uncertainty and growth in Iran. 47 The average duration of government investment projects in Iran is unusually high.

Economic Growth and Economic Policy in Iran: 1950– 2003

299

booms.48 Well-functioning economic institutions have a higher probability of developing in an environment in which small groups do not have privileged access to natural resource revenues. As mentioned in Sections 5.1–5.3, every major improvement in the terms of trade of oil, while improving the potential for rent seeking, has signiﬁcantly altered the posture of economic policies in Iran, making reform initiatives less credible. Based on the above discussions, we expect that economic rents made available through the government budget to be signiﬁcant and rent-seeking activities are likely to increase during an oil boom when the state distributes the windfall gains through bureaucratic channels. Karimi (2002) calculates the Katz and Rosenberg (KR) index of budgetary rent seeking and shows that this index is relatively high in Iran, as is the case for all populated oil-exporting countries. Jalali-Naini and Karimi (2003) show that the KR index of budgetary rent seeking is relatively high in Iran compared to a sample of developing and oil-exporting countries. Moreover, the value of the KR index increases after the oil boom of the mid-1970s and is highly correlated with changes in per-capita oil revenues and the size of the government. Furthermore, it is shown that the KR index negatively inﬂuences economic growth in Iran (Jalali-Naini and Karimi, 2003). This ﬁnding is consistent with the conjecture that a rentier system generates an incentive structure not conducive to growth (Easterly, 1993; Brixiova and Bulii, 2001). The above arguments also partially explain why Iran’s relatively rich resource endowments and high investment rates were not converted into high-growth performance.

7. Summary and conclusion Three distinct growth periods can be distinguished in Iran: 1959–1977, 1978–1988, and 1989–2003. The superior growth performance during the ﬁrst period reﬂected more favorable external and internal environments. During the ﬁrst period the share of the oil sector in GDP was not high until 1973, but oil revenues were growing. The growth of oil revenue and domestic expenditure prior to 1973, while boosting real aggregate demand and ﬁnancing imports of intermediate and capital goods and hence capacity output, were non-inﬂationary. Growth in domestic output was mainly due to growth in domestic

48

The masses and the intellectuals in Iran, as in a number of the MENA countries, have historically looked to the state as a medium for bringing about equity and social justice without sufﬁcient scrutiny under what kind of governance structure and policy regime the state can deliver the results. At various times there has been a signiﬁcant degree of resentment or opposition to the growth of private institutions and market-oriented economic policies in Iran. Not enough attention is given to the possibility that bureaucratic allocations and pervasive rent seeking associated with big governments can contribute to both inequality and inefﬁciency.

300

Ahmad R. Jalali-Naini

demand and non-oil exports grew at moderate to low rates. In spite of an extensive state presence in the economy, stable property rights, low uncertainty, and market-friendly policies allowed for the growth of the private sector. The sharp rise of the price of crude oil in 1974 fueled economic growth and inﬂation; it also induced rent-seeking activities and behavior. At the end of this period, growth came to an abrupt end owing to a political crisis, in part caused by income distribution concerns and inadequate social representation and participation in the socioeconomic system. The poor growth record in the second period is due to the debilitating effects of the war with Iraq and overbureaucratization of the economy. Moreover, a higher level of uncertainty at the macro as well as micro levels and adverse incentive/price signals were the main reasons for lower rates of the private sector participation in the economy. Per-capita GDP growth rates were negative in this period. During the 1989–2003 period, economic growth resumed and TFP rose. The improved economic growth during the third period partly reﬂects the unwinding of government controls, a more conducive environment for private sector participation in the economy, and a partial correction of the price system. However, the weight of a large and inefﬁcient bureaucracy, an institutional setup that limits internal competition and openness and encourages rent seeking, has limited the growth potential of the economy. Like other populated oil-exporting countries, the positive wealth and saving effects of the mid-1970s did not have a long-lasting effect on economic performance. Economic growth in Iran has been relatively volatile and factor-intensive during most of the 1959–2003 period and TFP did not signiﬁcantly contribute to economic growth. TFP’s contribution to growth was small, but positive during the 1961–1973 and 1989–2000 but negative during the oil boom of 1974–1977 and even worse during the 1978–1988 period. HC accumulation, as in the other MENA countries, did not make a signiﬁcant direct contribution to economic growth (reﬂecting inefﬁcient labor allocation mechanisms and markets); it made an indirect contribution through TFP growth. We used a number of policy and economic variables to explain movements of TFP. The rate of inﬂation and government size had a negative inﬂuence on TFP. The results from growth regressions indicate that economic growth is responsive to greater participation by the private sector in the economy, as measured by the ratio of private investment to (non-oil) GDP. They also show it to be negatively associated with proxies for price distortions, but no deﬁnitive effect could be ascertained with regard to the size of the government. Coordination and rent-seeking problems have contributed to a signiﬁcant increase of the average gestation period of government investments, which constitutes a large share of gross domestic ﬁxed capital formation in Iran. The international experience suggests that an economy characterized by the combination of a very substantial natural resource income; big government; extensive control of the economy; large price distortions as a by-product of inefﬁcient bureaucratic allocation schemes; an underdeveloped ﬁnancial market;

Economic Growth and Economic Policy in Iran: 1950– 2003

301

and limited domestic and foreign competition49, has a high probability of having an inferior growth performance. Economic performance is further undermined when coordination problems and rent seeking become pervasive. Such an environment is in most likelihood associated with low rates of increase in TFP and unsustainably high economic growth rates. Properly functioning institutions tend to ﬂourish when the rules limit rent seeking as well as impose credible checks and balances on those who control resource allocation via the political process. Economic growth in Iran has been adversely affected by large shifts in policy stance, motivated politically and/or by large shifts in the terms of trade of oil, resulting in uncertainty and high transaction costs. The consequence of a large and active government in both production and distribution spheres, relying on administrative allocation with disequilibrium prices, has been to undermine the development of social institutions for efﬁcient allocations of credit, capital, and labor, and the protection of property rights. Short-term jumps in growth rates owing to higher oil prices are not sustainable within the current economic structure. The experience suggests that the current oil boom will not have a longlasting positive growth effect unless institutional reforms to encourage wider participation of non-government institutions in the economic process, credible price/incentive reforms policies to improve resource utilization and allocation, and improved governance standards are in place. These reforms have a higher likelihood of success in an open economic and political environment. Acknowledgments I would like to thank Hashem Pesaran, Jeff Nugent, and two anonymous referees for their helpful comments on earlier drafts of this paper. References Arrow, K.J. (1962), ‘‘The economic implications of learning by doing’’, Review of Economic Studies, Vol. 29(3), pp. 155–173. Aschauer, D.A. (1989), ‘‘Is public expenditure productive?’’, Journal of Monetary Economics, Vol. 23(2), pp. 177–2000. Barro, R.J. (1991), ‘‘Economic growth in a cross-section of countries’’, Quarterly Journal of Economics, Vol. 106(2), pp. 407–443. Barro, R.J. (1995), Economic Growth, New York: McGraw–Hill.

49

The estimated Herﬁndahl indices demonstrate a high concentration ratio in the Iranian manufacturing industry and are indicative of a signiﬁcant degree of market power by government-owned enterprises. Moreover, government-owned companies account for a very large share in sales, valueadded, and employment. As the scale of operations becomes larger, the number of private ﬁrms decline, see Salehi and Mozhdehi (1999).

302

Ahmad R. Jalali-Naini

Barro, R.J. (2001), ‘‘Human Capital and growth’’, American Economic Review, Vol. 91(2), pp. 12–17. Benhabib, J. and M. Speigel (1994), ‘‘The role of human capital in economic development: evidence for aggregate cross-country data’’, Journal of Monetary Economics, Vol. 34(2), pp. 143–173. Bharier, J. (1971), Economic Development in Iran 1900– 1970, London: Oxford University Press. Brixiova, Z. and A. Bulii (2001), ‘‘Growth slowdown in bureaucratic economic systems: an issue revisited,’’ IMF Working Paper, WP/01/6, IMF, Washington, DC. Chenery, H.B. and M. Syrquin (1975), Patterns of Development 1950– 1970, London: Oxford University Press. Dasgupta, P. and I. Serageldin (editors) (2000), Social Capital: a Multifaceted Perspective, Washington DC: The World Bank. Dos Santos, T. (1971), ‘‘The structure of dependence’’, American Economic Review, Vol. 60, pp. 231–236. Easterly, W. (1993), ‘‘How much do distortions affect growth?’’, Journal of Monetary Economics, Vol. 32(2), pp. 187–212. Easterly, W., R. Islam and J. Stiglitz (2000), ‘‘Shaken and stirred: explaining growth volatility’’, Annual World Bank Conference on Development Economics, Washington, DC 2000, Oxford University Press. Easterly, W. and R. Levine (2001), ‘‘Its not factor accumulation: stylized facts and growth’’, World Bank Economic Review, Vol. 15(2), pp. 177–219. Easterly, W. and S. Rebelo (1993), ‘‘Fiscal policy and economic growth: an empirical investigation’’, Journal of Monetary Economics, Vol. 32(3), pp. 417–458. Easterly, W., C.A. Rodriguez and K. Schmidt-Hebbel (editors) (1994), Public Sector Deficits and Macroeconomic Performance, New York: Oxford University Press. Fischer, S. (1993), ‘‘The role of macroeconomic factors in growth’’, NBER Working Papers 4565. Frank, A.G. (1967), Capitalism and Underdevelopment in Latin America, New York: Monthly Review Press. Gelb, A. (editor) (1988), Oil Windfalls: Blessing or curse? New York: Oxford University Press. Ghosh, A. and H. Wolf (1998), ‘‘Threshold and context dependence in growth’’, NBER Working Papers 6480. Gylfason, T. (2001), ‘‘Nature, power, and growth’’, Manuscript, Center for Economic Studies (CESIFO). Hall, R.E. and C.I. Jones (1999), ‘‘Why do some countries produce so much more output than others?’’, Quarterly Journal of Economics, Vol. 114(1), pp. 83–116. Hirschman, A.O. (1968), ‘‘The political economy of import substituting industrialization in Latin America’’, Quarterly Journal of Economics, Vol. 82(1), pp. 2–32.

Economic Growth and Economic Policy in Iran: 1950– 2003

303

Iran Energy Report (2005), International Institute for Energy Studies, Tehran, Iran, 2005 edition. Islam, N. (1995), ‘‘Growth empirics: a panel data approach’’, The Quarterly Journal of Economics, Vol. 110(4), pp. 1127–1170. Issawi, C. (editor) (1971), The Economic History of Iran 1800– 1914, Chicago: University of Chicago Press. Jalali-Naini, A.R. (1999), ‘‘Business cycles in selected MENA countries’’, Research Project, ERF, www.erf.org.eg. Jalali-Naini, A.R. and M. Karimi (2003), ‘‘Rent seeking and economic growth’’, ERF 10th Annual Conference, December 2003, Marrakech, Morocco. Jalali-Naini A.R. and F. Khalatbari (2003), ‘‘Financial markets and growth in Iran’’, GDN, gdnet.org/pdf2/gdn_library/.../Iran_financial_final.pdf. Jalali-Naini and Naziﬁ, F. (2002), ‘‘Inﬂation-output tradeoff and asymmetric effects of monetary shocks’’, Plan and Development Quarterly,Vol. 3(2), June, pp. 13–41. Karimi, M. (2002), ‘‘Budgetary rent seeking in Iran’’, Unpublished M.S. dissertation, IRPD, Tehran. Karshenas, M. and M.H. Pesaran (1995), ‘‘Economic reform and the reconstruction of the Iranian economy’’, The Middle East Journal, Vol. 49(1), pp. 89–111. Khan M.S. and A.S. Senhadji (2000), ‘‘Threshold effects between inﬂation and growth’’, IMF Working Paper, WP/00/10. Khavari N.A. (2001), ‘‘An estimate of the gross domestic product of Iran: 1936–1958’’, Collected Monthly Lectures 2001, Monetary and Banking Research Institute, the Central Bank of Iran, Tehran, 2002. Klenow, P.J. and A. Rodrı´ guez–Clare (1997), ‘‘The neoclassical revival in growth economics: has it gone too far?’’, pp. 73–102 in: B. S. Bernanke and J. J. Rotemberg, editors, NBER Macroeconomics Annual 1997. Klenow, P.J. and A. Rodriguez-Clare (2004), ‘‘Externalities and growth’’, NBER Working Paper No. 11009. Knight, M.N., Loayza and D. Villanueva (1993), ‘‘Testing the neo-classical theory of economic growth’’, IMF Staff Papers, Vol. 40(3), pp. 412–441. Kongsamut, P., S. Rebelo, and D. Xie (1997), ‘‘Beyond balanced growth’’, NBER Working Paper 6159. Kuznets, S. (1971), Economic Growth of Nations: Total Output and Production Structure, Cambridge, MA: The Belknap Press of Harvard University Press. Levine, R., N. Loayza and T. Beck (2000), ‘‘Financial intermediation and growth: causality and causes’’, Journal of Monetary Economics, Vol. 46(1), pp. 31–77. Lucas, R.E. Jr. (1988), ‘‘On the mechanics of economic development’’, Journal of Monetary Economics, Vol. 22(1), pp. 3–42. Mankiw, N.G. (1995), ‘‘The growth of nations’’, Brookings Papers on Economic Activity, Vol. 1, pp. 275–310. Mankiw, N.G., D. Romer and D.N. Weil (2000), ‘‘A contribution to the empirics of growth’’, Quarterly Journal of Economics, Vol. 107(2), pp. 407–437.

304

Ahmad R. Jalali-Naini

Makdisi, S., Z. Fattah and I. Limam (2005), ‘‘Determinants of growth in the MENA countries’’, chapter 2 in: J.A. Nugent and M.H. Pesaran, editors, Explaining Growth in the Middle East, Amsterdam: Elsevier. Nelson, R.R. and E.S. Phelps (1966), ‘‘Investment in humans, technological diffusion, and economic growth’’, American Economic Review, Vol. 61(2), pp. 69–75. North, D.C. (1990), Institutional Change and Economic Performance, Cambridge: Cambrige University Press. North, D.C. (1994), ‘‘Economic performance through time’’, The American Economic Review, Vol. 84(3), pp. 369–395. Pesaran, M.H. (1992), ‘‘The Iranian foreign exchange policy and the black market for dollars’’, International Journal of Middle Eastern Studies, Vol. 24, pp. 101–125. Pesaran, M.H. (2000), ‘‘Economic trends and macroeconomic policies in postrevolutionary Iran’’, chapter 2 pp. 63–100 in: P. Alizadeh, editors, The Economy of Iran: Dilemmas of an Islamic State, London: I.B. Tauris. Pesaran, M.H. and B. Pesaran (1997), Microfit 4.0, New York: Oxford University Press. Phillips, P.C.B. and B.E. Hansen (1990), ‘‘Statistical inference in instrumental variables regression with I(1) processes’’, Review of Economic Studies, Vol. 57, pp. 99–125. Pritchett, L. (1996), ‘‘Where has all the education gone?’’, Policy Research Working Paper Series 1581, The World Bank, Washington, DC. Romer, P. (1986), ‘‘Increasing returns and long-run growth’’, Journal of Political economy, Vol. 94(5), pp. 1002–1037. Romer, P. (1990a), ‘‘Increasing returns and long-run growth’’, Journal of Political Economy, Vol. 94(5), pp. 1002–1037. Romer, P. (1990b), ‘‘Endogenous technological change’’, Journal of Political Economy, Vol. 98(5), pp. S71–S102 Part two. Rose, R. (2000), ‘‘Getting things done in an antimodern society: social capital networks in Russia’’, in: P. Dasgupta and I. Serageldin, editors, Social Capital: a Multifaceted Perspective, Washington D.C.: The World Bank. Saedi, A.S. (2002), ‘‘An analysis of the factors inﬂuencing wages in the Iranian services sector’’, Unpublished M.A. thesis, Tehran: IRPD. Salehi-Isfahani, D. (2005), ‘‘Microeconomics of growth in MENA: the role of households’’, chapter 6 in: J.B. Nugent and M.H. Pesaran, editors, Explaining Growth in the Middle East, Amsterdam: Elsevier. Salehi-Isfahani, H. (2002), ‘‘The political economy of growth in Iran’’, Paper prepared by GRP project, Global Development Network. ww.erf.org.eg/ html/grp/GRP_Sep03/Iran-Pol_Econ.pdf Salehi H. and A.J. Mozhdehi 1378 (1999), ‘‘Review of the market structure in Iran’’, The Iranian Economy, in: M. Tabibian, editor, Tehran: the Institute for Research in Planning and Development.

Economic Growth and Economic Policy in Iran: 1950– 2003

305

Senhadji A. (1999), ‘‘Sources of economic growth: an extensive growth accounting exercise’’, IMF Working Papers, WP/99/77. Solow, R.M. (1956), ‘‘A contribution to the theory of economic growth’’, Quarterly Journal of Economics, Vol. 70(1), pp. 65–94. Stigliz, E.J. (1989), ‘‘Markets, market failures, and development’’, American Economic Review, Vol. 79(2), pp. 197–203. Tabibian, M. (2002), ‘‘Manufacturing sector’s long-term strategy and development in Iran’’, GRP Country Paper, www.erf.org.eg. Tanzi, V. and H.H. Zee (1997), ‘‘Fiscal policy and long-run growth’’, IMF Staff Papers, Vol. 44(2), pp. 179–209. World Bank Report (1997), The State in a Changing World, Washington, DC: World Bank. Young, A. (1995), ‘‘The tyranny of numbers: confronting the statistical realities of Asian growth experience’’, Quarterly Journal of Economics, Vol. 110(3), pp. 641–680.

This page intentionally left blank

CHAPTER 10

The Political Economy of Development Policy in Tunisia Mohamed Z. Bechri and Sonia Naccache Abstract During the last five decades Tunisia managed to achieve a robust long-run GDP growth of about 5% per year, against 3.5% only, for the MENA region as a whole. However, Tunisia failed to reduce obvious inefficiencies such as high protection, financial repression, and a highly regulated economy in general. The present study attempts to explain why the country was unable to do better in this respect, even though it was a top performer in social modernization and it started its economic reforms relatively early. The study adopts the political economy approach to explain independent Tunisia’s development policy choices, based on the theory of collective action and an analysis of contracting problems of representation, coordination, and commitment. The paper draws on some lessons from Tunisia for the other MENA countries. It stresses the importance of women’s rights, population control, and modern education. In addition, it refers to the recent examples of Morocco and Kuwait regarding women status reform to argue that action is still possible. It argues that even though no efforts should be spared to get influential groups on board, the reforming governments in MENA today cannot afford to be paralyzed by anti-reform groups – such as Islamist groups in Parliaments – either. KEYWORDS: political economy, economic development, interest groups, collective action, Tunisia

JEL classiﬁcations: D78

Corresponding author. CONTRIBUTIONS TO ECONOMIC ANALYSIS VOLUME 278 ISSN: 0573-8555 DOI:10.1016/S0573-8555(06)78010-5

r 2007 ELSEVIER B.V. ALL RIGHTS RESERVED

308

Mohamed Z. Bechri and Sonia Naccache

1. Introduction When Tunisia gained independence from France in 1956, the economy was largely based on subsistence farming, food processing, and mining. In this respect, independence undoubtedly ushered in an unprecedented era of economic and social development. The policies pursued to that end shifted from State intervention (1956–1960) to socialism (1961–1969) to export promotion (1970–1985). Then following a severe balance of payments crisis in 1986, Tunisia adopted a structural adjustment program under the guidance of the International Monetary Fund (IMF). The liberal policies implemented under this framework were further boosted with the signing of a free-trade agreement with the European Union in 1995. The main features of Tunisia’s development during the past ﬁve decades are: (1) an early and sustained focus on social modernization and population control, (2) a heavy involvement of the State in the economy, even after the abandonment of collectivism in 1969, and (3) a robust long-run GDP growth of about 5% per year. But Tunisia’s performance remained well below that of Southeast Asia and other fast-growing economies, which means that the country could have done much better. Hence, the relevant questions for the present study are: how did Tunisia manage to achieve a robust growth over such a long period of time? And why was the country unable to do better? In the present study, we attempt to answer these questions, by explaining Tunisia’s policy choices as the outcome of interaction between institutions and interest groups, taking into account contracting problems and the way they impeded groups’ ability to implement efﬁcient policies. The policy variables we focus on are: the role of the private sector, foreign trade, and credit and exchange rate policies, in addition to gender issues and population control, which make Tunisia an exception by regional standards. The study is structured as follows: Section 2 summarizes the main policy episodes since 1956 and the political economy approach. Section 3 analyzes the establishment of a State-led economy, which paved the way to socialism in the early 1960s until its end in 1969. The adoption of a two-track strategy is analyzed in Section 4. Section 5 studies the populist experiment of 1981–1985, the crisis-triggered reforms of 1986, and the signing of the Association Agreement with the European Union. Section 6 contains the main conclusions. 2. Policy episodes and the political economy approach 2.1. Policy episodes: 1956– 2004 The main episodes of Tunisia’s development policy since independence are summarized in Table 1. The ﬁrst 5 years (1956–1960) were characterized by State intervention. Monopolies, inherited from the colonial rule, became State-owned.

1961–1969 A. Role of the state

B. Private sector development C. Trade policy

D. Credit policy

Number of public enterprises Share of the public sector in total investment (in %) Share of the private sector in total investment (in %)a Nominal protection rate (in %) Effective protection rate in (in %)b Real deposit interest rates Real lending interest rates

25 (1960)

Tunisia’s development policy episodes

1970–1980 185 (1970)

1981–1985 285 (1984)

1986–1991

1992–1995

1996–2000

n.a.

219 (1993)

203

2001–2004 187

68.08

59.37

54.05

50.89

50.79

47.80

43.26

31.92

40.63

45.95

49.11

49.21

52.20

56.74

n.a.

n.a.

34.33

[41.29]

37

46.66

36

n.a.

n.a.

75

[70.44]

56

67.66

50

1.15

–1.25

–3.92

–4.98

–5.73

–5.01

–4.08

6.7

5.33

4.28

7.77

7.22

4.83

4.49

The Political Economy of Development Policy in Tunisia

Table 1.

309

310

Table 1 (continued ) 1961–1969 E. Exchange rate (annual change in, for depreciation)

REERc Exchange rate equilibriumd Black-market premiume (%)

n.a. n.a.

81.0

1970–1980 n.a. 0.5 (1970–1979) 15.7

1981–1985

1986–1991

1992–1995

–1.39 2.0 (1980–1985)

–6.57 –6.0 (1986–1990)

0.297 4.0 (1991–1997)

102

5.0

9.2

1996–2000

2001–2004

–0.052 n.a.

–1.785 n.a.

0.7

0.7

(1) Nominal and effective rates relate to the whole economy, hydrocarbons excluded. Estimates began only in 1983 and were interrupted during the years 1987–1989, 1991–1994, 1996, and 1998. Therefore, average rates relate to the years for which estimates are available. (2) Average rate for the years 1983–1985. Source: Cahiers de 1’Institut d’Economie Quantitative, No. 5, March 1988, p. 92. (3) Between 1986 and 1990, respectively. Source: Institute d’Economie Quantitative. (4) Figure of 1995. (5) cf. footnote b. (6) Given different lending rates across industries, during the period 1962–1986, we take the taux du decouvert bancaire as a proxy. Since 1987, the rate is the money market rate+3 percentage points. (7) For nominal interest rates, the source is Statiques Financieres de la Banque Centrale de Tunisie. The inﬂation rate is the rate of growth of the GDP deﬂator (100 ¼ 1990). Sources: International Financial Statistics 2004, IMF and Statistiques Financieres de la Banque Centrale de Tunisie. (8) Since 1987, nominal rates on deposit have been freely set by banks.

a

Share of private ﬁrms and households in gross ﬁxed capital formation, at market prices. Sources: Annual Report of the Central Bank of Tunisia and National Accounts. b The preceding notes on nominal rates apply to the corresponding effective rates. c Real effective exchange rates, Source: International Financial Statistics, IMF. d Based on economic fundamentals. Source: Achy (2000) Appendix 2. e Black-market premium. Sources: Levine and Renelt: World’s Currency Yearbook (for 1985, 1990–1993), and Adrian Wood, Global trends in real exchange rates: 1960–1984, World Bank Discussion Paper No. 35, 1988.

Mohamed Z. Bechri and Sonia Naccache

n.a. Not available. Notes:

The Political Economy of Development Policy in Tunisia

311

And the creation of the Central Bank and of two government-owned banks in 1958 ensured public control of the ﬁnancial sector. At the social level, a radical reform of women’s status was adopted, population control was integrated in the development strategy of the country, and a modern educational system, based on the French curricula, was promoted all over the country, including the remote rural areas. Moreover, Zeitouna University, which had been the guardian of Sunni Islam since the Islamic conquest of the country, competing in inﬂuence with Al-Azhar in Cairo, was abolished in 1956 when its students integrated the public system. Meanwhile, at the economic level, State intervention was further strengthened, with the adoption of central planning in 1962 and collectivization in 1965, a disappointing experience that was abandoned in 1969. This was followed by the adoption of a ‘‘two-track’’ approach, with the establishment of an offshore industrial sector for exports, in addition to a traditional local and highly protected sector. With the passage of time, however, the system proved to be unsustainable. Despite the high protection of the onshore sector, the populist policies adopted in the early 1980s resulted in a severe balance of payments crisis. As a result, the country had ﬁnally no choice but to sign a Stand-by Agreement with the IMF in August 1986, and to adopt a stabilizationcum-structural adjustment program. As a result, macroeconomic stability was rapidly achieved, but institutional reforms remained hesitant, and private investment failed to recover. Under these circumstances, the authorities decided to take a bold decision by signing an association agreement with the European Union in 1995, being the ﬁrst country in the South-Mediterranean region to do so. In this framework, an ambitious program to upgrade domestic industries was launched in 1996. But again institutional reforms, especially regarding the civil service, the labor code, and banking and telecommunications, remained limited. A political economy approach will help to explain this failure.

2.2. Political economy approach Why do countries persist in adopting inefﬁcient policies? One explanation is that these inefﬁciencies may beneﬁt some powerful interest groups. In what follows we present the theory of collective action and its limits to explain reform, followed by a presentation of contracting problems as a potential source of inefﬁciencies. 2.2.1. Interest groups and collective action The main thrust of the theory of collective action is that policy choices are outcomes of interest groups’ pressures and lobbying (for a survey on the subject, see for instance, Nabli and Nugent (1989) and Isfahani (2002)).

312

Mohamed Z. Bechri and Sonia Naccache

First, success in collective action is enhanced with the small size of the group and its geographic concentration. Otherwise, it would be difﬁcult to mobilize its members. Farmers in the developing world, as an example, constitute a large share of the population and they are thinly spread in the rural areas, which makes it difﬁcult for them to mobilize. As a result, they fail to get the kind of subsidies that are prevalent in the industrial world, where it is easier for farmers to communicate and lobby for their common good. Second, sector concentration helps to do away with free riders, since a small number of large beneﬁciaries will be willing to undertake all the needed lobbying. Third, homogeneity of both background and goal enhances the focus on a speciﬁc set of claims. In a situation of rationing with price set below the market clearing one, as for example with regulated credit markets, business groups tend to focus more on credit availability than on preferential interest rates. Moreover, they may consider protection to be more sustainable than subsidies, since the tariffs can also serve as an important source of revenue for the government. 2.2.2. Limits of the theory of collective action in explaining reforms Despite its appeal, the theory presented so far does not do well in explaining the timing of reforms, since it is unlikely that relative strength of different interest groups’ inﬂuence would have changed dramatically at that particular period of time. We also observe that countries with similar economic conditions may adopt different policies. As we will see below, a failed collectivist experiment led to radical reforms in Tunisia in 1969, whereas neighboring countries with similar problems did not follow Tunisia’s lead. Several factors may explain the failure to reform. First, groups may not be able to ﬁgure out the type of policies that better serve their interests in the long run. For example, Rodrik (1998) argued that import-substitution industrialists in Korea and Taiwan resisted the reforms that these countries introduced in the mid-1960s. An alternative explanation is that high redistribution costs may be involved, as is generally the case in trade reform. In this case, politicians hesitate to reform precisely because they perceive a high ‘‘political cost–beneﬁt ratio,’’ i.e., a high ratio of redistribution to efﬁciency gains. Under these conditions, only a severe crisis that makes the status quo unsustainable may succeed in triggering reforms. Another alternative that may have lower political costs is to follow a two-track reform strategy, whereby new opportunities are created, without unduly altering the old system. Examples include the promotion of a private sector while maintaining State-owned enterprises and the creation of special processing zones for exports. 2.2.3. Institutional dimensions of governance: representation, coordination and commitment The failure to remove inefﬁciencies to have access to a larger surplus may also be explained by contracting problems, which depend on representation, coordination, and commitment (Castanheira and Isfahani (forthcoming)).

The Political Economy of Development Policy in Tunisia

313

Representation requires the government to have a power delegated to it by the people. This enhances the government’s ability to appropriately assess the needs of society to make the appropriate tradeoff between competing demands. Furthermore, groups will cooperate better in implementing government policies when they feel they are part of the decision-making process. But, a government setup aimed at enhancing representation may end up reducing efﬁciency at the implementation level. This is of particular importance in the Arab countries owing to the powerful impact of tradition, tribalism, and regionalism factors. Coordination is needed to prevent excesses like high taxes and overspending for government provision of goods and services that the private sector can better provide. In this respect, trade unions, as an example, could play a very constructive role in participating to wage negotiations and in ensuring social stability. The tradeoff between competing demands depends ﬁrst and foremost on the degree of government intervention in the economy. When the government plays an important role, regions will compete more to attract public projects and social groups lobby more for handouts (subsidies, access to bank credit, protection, etc.). That is why the ﬁrst-best solution is to have an institutional framework that leaves most choices to the market. The adoption of vouchers in education and private sector ﬁnancing of infrastructure, as an example, may dispense the government from having to make difﬁcult choices. Low taxes would also limit the amount the government will have to spend and, as a consequence, the coordination needed in that respect. Moreover, the same is true with the adoption of some ex-ante rules like a balanced budget law. In all circumstances, decentralization is very effective. Furthermore, decisions that could be made at the community or the regional level should not be left to the central government. Commitment indicates the extent to which announced policies are effectively implemented. Credibility is particularly difﬁcult to accomplish in developing countries known for weak institutions and past histories of failures by government ofﬁcials to fulﬁll promises. Some alternative solutions exist, however. Under these circumstances, the presence of a visionary leadership with historical legitimacy may be an effective mechanism to implement policies that improve signiﬁcantly the well being of the people, as pointed out by World Bank (2003) with a speciﬁc reference to Bourguiba in Tunisia and Lee Kuan Lew in Singapore. But, this should not be a justiﬁcation for dictatorship either. Many historical precedents show how one-man rule may not be sustainable and how democracies can indeed achieve reform. Recent successful MENA cases in social modernization include Morocco and Kuwait. Morocco failed for years to adopt a modern legal code for women owing to Islamist opposition. The latter mobilized hundreds of thousands of followers in street protests during the late 1990s to sabotage the socialist government’s attempts in this respect. However, the king seized the aftermath of the May 16, 2003 terrorist bombings of Casa Blanca, to present the Moudawana, which made polygamy impossible to practise and better protected women rights. Cornered and

314

Mohamed Z. Bechri and Sonia Naccache

isolated, the Islamist deputies in Parliament had no choice but to vote for it. Moreover, for years, tribal leaders in Kuwait’s Parliament turned down the Emiri decree of 1999 that gave women the right to vote. But, after long negotiations, Islamist terror that struck neighboring Saudi Arabia and Kuwait itself, and several democratic steps in neighboring Bahrain, Qatar, and Oman, the government ﬁnally succeeded in obtaining a vote in favor of the law on May 16, 2005. Once the political backing is secured to overcome local opposition, the selection of a competent technocratic team can signiﬁcantly enhance the probability that the reforms are implemented. In the case of economic liberalization, for example, the appointment of a technocratic pro-market team like the ‘‘Chicago boys’’ in Chile, can signiﬁcantly enhance the chances of success.

3. Establishment of a State-led economy (1956–1969) In addition to its agricultural potential, Tunisia has succeeded for centuries in producing and even in exporting some manufacturing goods such as textiles, ceramics, leather, and other handicrafts. Production processes remained, nevertheless, artisanal in nature (Bellin, 1989). In the mid-nineteenth century, the establishment by Ahmed Bey of light weaponry and food processing plants gave a temporary boost to industrialization. But this attempt was halted by the French invasion of 1881. In 1904, a customs’ union with France was established but within a colonial vision in which Tunisia was to be kept as a safe market for French consumer products. At the time when the country gained independence in 1956, the economy remained based largely on subsistence farming, a few other importsubstitution industries, and non-tradables.

3.1. Political setup and interest groups After a long and protracted struggle, taking the form of both peaceful resistance and military insurgencies, Tunisia won its independence from France in the mid1950s. The resistance was ﬁrst organized by the Destour Party, which was founded by Sheikh Abdelaziz Thaalbi, a prominent scholar at Zeitouna University. But the movement failed to outline a clear strategy for independence. A French-educated young lawyer, Habib Bourguiba, seized the opportunity to break ranks with the old guard and created the Neo-Destour in 1934. It was a preventive action that would gradually marginalize the historic inﬂuence of the traditional Zeitouna-based establishment. Later on, this gave Bourguiba a free hand in negotiating independence with France, and in implementing his modernization agenda, once the country became independent. The ﬁrst French offer of ‘‘internal independence’’ of Tunisia came in 1955. Bourguiba accepted it, but a nationalist wing of the Neo-Destour, headed by its

The Political Economy of Development Policy in Tunisia

315

Secretary General Salah Ben Youssef, vehemently opposed it. In a speech delivered at the Zeitouna Mosque, Mr. Ben Youssef considered the offer as ‘‘a step backward.’’ The conﬂict took a violent turn and brought the country to the brink of civil war. But an extraordinary congress of the Neo-Destour fully supported Bourguiba, and Mr. Ben Youssef was sacked. One year later, in 1956, Tunisia was granted full independence. This event proved Bourguiba right and gave a ﬁnal blow to the rebellion. Tunisia’s accession to independence greatly enhanced the legitimacy of its leadership. Moreover, Bourguiba’s hold on power was further strengthened with the co-option of the national associations and the proclamation of the Republic in 1957. At the interest group level, the groups expected to play leading roles in the policy debate during this period were producers’ associations, labor, and the ruling Party elite. Producers’ associations were limited to the Union Tunisienne de l’Artisanat et du Commerce (UTAC, to become UTICA later on) for manufacturing and commerce, and the Union Nationale des Agriculteurs de Tunisie (UNAT), for farmers. UTAC was founded in 1947, by the Neo-Destour, in reaction to the communist party’s Federation des Artisans et Petits Commercants de Tunisie, which was created in 1945. But while UTAC’s leadership was elected during the colonial era, since independence, the organization’s president was handpicked by the authorities. In addition, UTAC was asked to encompass all types of industries, handicrafts, and commerce as well as the service sector since the mid-1960s. This strategy of amalgamation further weakened the organization’s ability to mount collective action (Nugent, 1989). Moreover, UNAT was in no better position. Large farms, formerly in the hands of European nationals, became State-owned. This left the union membership limited to small rural farmers among whom organization was difﬁcult and to a handful of urban-based members more interested in (non-agricultural) lucrative activities such as trade, handicrafts, and manufacturing. As a result, the ﬁnancially squeezed organization had no choice but to rely on staff hired from the Agriculture Ministry and on government handouts. With their leadership neutralized, producers’ organizations became a mere instrument of State inﬂuence. This left organized labor and the ruling Party elite as the only players with the potential to effect coherent government policy. The national labor union Union Tunisienne des Travailleurs de Tunisie (UGTT) was created in 1946 by Farhat Hached, a popular ﬁgure who was later assassinated by the French. Its nationalist stand and large membership base made it the only organization really capable of challenging Bourguiba for the Neo-Destour Party political leadership. Some UGTT activists started the ideological struggle well before the country’s access to independence. For example, in 1955 at a congress of the Neo-Destour that was held in Sfax to give its support to Bourguiba, a young UGTT leader, Ahmed Ben Salah, seized the opportunity to present an ‘‘Economic Report,’’ which detailed his vision for a socialist development strategy for independent Tunisia.

316

Mohamed Z. Bechri and Sonia Naccache

UGTT’s strategy from the beginning was to inﬁltrate the ruling Party and thereby to orient economic policy from within. This became clear in 1956 when UGTT leadership called for the constitution of an ‘‘organic union’’ with the Neo-Destour in an attempt to turn the latter into a labor Party. Since the ruling Party’s strategy was to co-opt national associations so as to dilute dissent, it was easy for UGTT to increase its inﬂuence within the party. Indeed, at the ﬁrst legislative election of independent Tunisia, a ‘‘National Front’’ was constituted, including UTICA, UNAT, and UGTT, under the leadership of the NeoDestour. The UGTT was the main beneﬁciary from this participation since it succeeded in securing 35 out of the 98 seats in the new Parliament. However, it is worth noticing that the Neo-Destour’s natural inclination remained in favor of a liberal policy. Its membership base came mainly from the rural areas, where the basic activity is subsistence farming and in urban areas, merchants constituted its main support. Furthermore, the market economy is the natural choice of its Western-educated leadership. Indeed, even before the country’s access to independence, Habib Bourguiba expressed on many occasions his apprehension about UGTT-advocated socialism. 3.2. Representation, coordination, and commitment Tunisia’s independence raised expectations about a better popular representation. Indeed, the debates that took place at the National Assembly during the early years of independence reﬂected to a large degree the concerns of the population and gave the appearance of representative rule. The real power, however, was concentrated in the hands of Habib Bourguiba, who was so conﬁdent about his status that he dared to announce on August 13, 1956 revolutionary measures in favor of women, and on July 20, 1957 the abolition of the monarchy. In both cases, the rubber-stamp parliament was urged to adopt the changes. But Bourguiba was the real driving force behind this. Indeed, it is a well-known fact that he personally drafted the Women Code and for the announcement of the republic, the long sessions organized by the parliament played no more than a ceremonial role, since no serious debate took place, and no referendum was organized. From Bourguiba’s speeches and behavior during this period, one could easily sense how, like any other autocrat, he saw himself, from the start, as having a better grasp of the country’s priorities than the elected representatives. Moreover, his action in favor of women, education, and the overall development and his ability to co-opt labor and business associations, raised his status as the supreme leader of the country. Furthermore, his insistence on ‘‘national unity,’’ indicates just how he views representation as a threat, a view that was probably shared by a large section of the population at that time. In addition, the centralized nature of the system enhanced the coordination process, and Bourguiba’s secure position gave him a strong incentive to commit himself to efﬁcient policies. However, the newly independent Tunisia clearly lacked the technical

The Political Economy of Development Policy in Tunisia

317

skills it needed to manage its economy. Tunisian reformists of the nineteenth century, once they became disillusioned about the ability of the Zeitouna Mosque to modernize, created a new model of French-curricula-based schools. Some of their graduates (like Bourguiba himself ) continued their studies in France. But they remained limited in number, especially in the ﬁeld of economics and business management. These weaknesses, together with the failure of the liberal wing of the Neo-Destour to come up with a consistent plan for economic reconstruction, explain to a large extent the chaotic situation that prevailed during 1956–1961. This provided an opportunity for the trade union leader Ahmed Ben Salah to advance his socialist agenda. 3.3. Policy choices Two basic choices were made during the post-independence period: State intervention and social modernization. 3.3.1. State intervention The post-independence years of 1956–1961 were generally, but mistakenly, considered a period of liberalism. In fact, State intervention was prevalent. In public utilities (electricity and water) and other lucrative businesses like tobacco, government monopolies were maintained. The rate of protection of domestic industries increased, with some tariffs reaching prohibitive levels. In addition, the two banks created in 1958 were government-owned banks, guaranteeing government dominance over the ﬁnancial sector. There was obviously no shortage of arguments to justify this option, for it could be argued that: (1) the private sector was, at that stage, in no position to own and manage large units like public utilities, (2) the government was in a better position to allocate the scarce resources to priority sectors, and (3) what Grissa (1989) had called ‘‘the need for the State to establish itself’’ all required State ownership of the means of production to provide jobs and distribute rents. Likewise, public-sector banks would enhance the government’s ability to allocate credit to priority activities. But while making political sense, these interventions sent a negative signal to private entrepreneurs. Business took a ‘‘wait and see’’ attitude, and in some instances it took refuge in real estate speculation such as buying abandoned French properties. This understandable response to the prevailing set of incentives at that time was, however, mistakenly perceived by some observers as a disappointing response, and even a proof of the ‘‘parasitic nature’’ of the Tunisian private sector (Bellin, 1989). 3.3.2. Social modernization As already indicated, President Bourguiba took the lead by adopting some extremely bold measures in favor of women’s rights. First, Bourguiba set up a special commission to review the status of women, and, when the latter could

318

Mohamed Z. Bechri and Sonia Naccache

not agree on the thorny issue of polygamy (though their consensus was to limit the right to two wives only instead of four), the President took the matter in his own hands, by introducing the needed changes to the draft law, which was ﬁrst adopted in August 1956, and integrated later into the Woman’s Code in 1958. This would be by far the most progressive reform of women’s status in the Islamic world. The new code outlawed polygamy, tightly regulated divorce, and ﬁxed the minimum age of marriage for girls at 18. The move to announce these measures was so swift that it caught the remnants of the traditional establishment off guard. But, Bourguiba knew that, for these reforms to be effective, he had to use his prestige, historical legitimacy, and oratory gifts in an attempt to change people’s attitudes toward women. Hence, the issues of women’s liberation and population control became a recurrent theme of his regular speeches to the nation. Social modernization required also modern education. Bourguiba himself was educated in the ‘‘Sadiki College’’ created in reaction to the failure of Zeitouna University’s traditional system. Bourguiba’s ﬁrst action, in this respect, was to generalize free public education, with the adoption of the French curriculum. This was obviously at the expense of the ‘‘Madrassas’’ (religious schools) system. Moreover, once the new system became popular among parents, Bourguiba took a second bold decision, namely, to abolish Zeitouna University altogether. It was only in the early 1990s that a new Zeitouna University was created, modeled on Western theology departments where the focus is on comparative religious studies. 3.4. Rise and fall of collectivism After a period of economic stagnation, Bourguiba appointed the trade union activist Ahmed Ben Salah as the Minister of Social Affairs and Economy, in 1961. The decision made political sense, since the move was meant to defuse the threat of a uniﬁed leadership by labor, which would include three ﬁgures: Habib Achour, Ahmed Tlili, and Ahmed Ben Salah and things worked out very well. For once he became minister, Ben Salah used his insider knowledge of the union, with the help of some of his former comrades, to neutralize UGTT (Khiari, 2000). This was a milestone in Bourguiba’s quest to destroy any potential source of organized resistance to his leadership. But this ‘‘cooption’’ also had a major drawback in that the integration of UGTT leadership into the Neo-Destour gave the former a precious opportunity to introduce policy changes from within. 3.4.1. Adoption of central planning With Bourguiba’s blessing secured, Ben Salah had obviously no reason to give himself an executive role. Soon, he started maneuvering to shift policy toward central planning, which was introduced in 1962 with the publication of the

The Political Economy of Development Policy in Tunisia

319

‘‘Perspectives Decennales.’’ Meanwhile, a populist policy led to a large increase in wages. As a result, the ﬁscal deﬁcit widened and a balance of payments crisis erupted. In 1964, Tunisia had no choice but to adopt a stabilization plan and to devalue the dinar, for the ﬁrst time since independence. The 1964 debacle was a sufﬁcient reason for the dismissal of Ben Salah. As pointed out above, Bourguiba needed him in the government to keep UGTT leadership divided. But there was no need to give him the portfolio of the economy either. Bourguiba’s failure to act at that time remains a puzzle. One reason may be that the technicalities of economic management were not a priority for the President who was known ﬁrst and foremost as a veteran politician. In addition, the failure of the liberal wing of the Neo-Destour to come up with a credible alternative may have left the President with no other choice. Furthermore, the economic mess may have been presented to the President as a technical matter to be corrected with the adoption of appropriate stabilization tools alone. In any case, Bourguiba seemed to have bought the argument. So much so, that as soon as the economy showed signs of recovery, the President gave a carte blanche to Mr. Ben Salah to go ahead with the collectivization of the economy, an adventure terminated only by an economic collapse that came to threaten Bourguiba’s own hold on power. 3.4.2. Population control and family planning Unlike the chaos on the economic front, social modernization continued unabated. The revolutionary measures promulgated in the late 1950s were further consolidated by setting up a family planning program, which was integrated to the National Development Plan, starting in 1962–1963. To this end, an Office National de Planning Familial (ONPF) was created to disseminate information. Important investments were made in reproductive health and incentives were structured so that social security beneﬁts were limited to those families with no more than three children. Population control is obviously a long-term endeavor. Nevertheless, as early as in 1970, Tunisia’s total fertility rate (the average number of births per woman) declined to 6.4 compared to 7.0 in Morocco and 7.4 in Algeria (Table 2). Meanwhile, Tunisia boasted the highest rate of girl’s education at the primary level in the region (61.9%, compared to 49.4% for Egypt and 33.9% for Iraq). More importantly, population control allowed Tunisia to enter a ‘‘virtuous circle.’’ Less demographic pressure enhanced socioeconomic development, which in turn helped reduce population growth. 3.4.3. Shift to collectivism The political decision in this respect came at the Neo-Destour Congress in 1964 where Ben Salah scored a clear success. This owed much to his personal qualities, particularly his dynamism, relative integrity, and commitment. But these qualities would not have been enough, had Bourguiba been better advised, and

320

Indicator Year Tunisia Algeria Jordan Egypt Morocco Syria

Expected Life at Birth

Social development indicators: Tunisia and MENA countries

Total Fertility Rate

1970

1980

1998

2002

1970

1980

1998

54 53 54 51 52 56

62 59

72 71 70 67 67 70

73 71 72 69 68 70

6.4 7.4

5.2 6.7 6.8 5.1 5.4 7.4

2.2 3.5 3.6 3.5 3.0 3.7

56 58 62

5.9 7.0 7.7

Source: Arab Monetary Fund, the Arab Uniﬁed Report, 2004.

Primary 1970

1980

1996

Schooling (in %) Secondary

Higher

2001

1970

1980

1996

2001

1970

1980

1996

2001

100.4 102.1 118.0 117.0 76.1 94.5 107.5 112.0 72.0 81.6 70.6 100.8 67.6 73.1 101.1 100.0 51.5 83.0 86.0 94.0 77.5 99.6 100.8 109.0

22.7 11.2 32.8 28.4 12.6 38.1

27.0 33.0 59.1 50.5 26.0 46.4

64.63 63.3 57.4 78.3 39.1 42.5

78.0 71.0 87.7 86.0 39.3 43.0

2.6 1.8 2.1 6.9 1.4 8.3

4.8 5.9 13.4 16.1 5.9 16.9

13.7 13.4 17.9 22.6 11.1 15.1

15.0 – 28.6 – 10.0 –

Mohamed Z. Bechri and Sonia Naccache

Table 2.

The Political Economy of Development Policy in Tunisia

321

the producer organizations in a better shape. Indeed, as pointed out by Nugent (1989, p. 295) ‘‘cooperativization of private property would not have been possible if producer organizations had been strong.’’ In addition, the liberal wing of the Neo-Destour failed to seize on the failure of collectivism experiments, which were showing disastrous results at that time, in other countries such as Egypt and some Sub-Saharan countries. With their associations in shambles, the only way for producers to effectively voice their concerns was outside the ofﬁcial organizations. But this was very risky. The government’s intolerance of any type of dissidence was made very clear by its decision to disband the Communist Party in 1962, and by responding with overwhelming use of force to quell the violence that erupted in some regions to protest against cooperatives. This does not mean, however, that producers’ interests were irrelevant to the government. Interestingly enough, it was the weakest groups in terms of collective action that were the ﬁrst to be targeted for collectivism (farming, transportation, and local commerce).

3.4.4. Fall of collectivism After 4 years of collectivism, Bourguiba dismissed Ben Salah and abolished the cooperatives. It was an abrupt move that changed dramatically the way Tunisia will develop during the following decades. Bourguiba’s decision followed a World Bank Report on Tunisia in 1969, which focused on the severe ﬁscal crisis that the country was facing. The experience of the year 1964 showed, nevertheless, that an economic crisis by itself was in no way sufﬁcient to bring down the government, and it is even less likely to engineer a radical shift in policy. The difference this time is that the crisis was aggravated by popular discontent about expropriations, shortages, and a severe drought for 3 years in a row. But, one could still argue that similar problems were encountered in other countries in the region, without having the same impact. To understand Bourguiba’s decision, we need to keep in mind that for the Western-educated, liberal-minded Bourguiba (at least as long as his own grip on power was not concerned), the puzzle remains his adoption of the socialist orientation in the early 1960s, not its abandonment in 1969. This is indeed in contrast to many other Third World countries, where dogmatic socialist leaders implemented collectivism. More importantly, the large popular discontent at that time required a strong response from the security apparatus and, eventually, a need to call in the army. The latter option would have been particularly horrifying for Bourguiba, given his well-known distrust of the military. As soon as Ben Salah was dismissed, the government started a reversal of the most damaging and unpopular aspects of the former policy, namely cooperatives. Private initiative was reinstated and hailed as an engine of growth. Furthermore, a two-track strategy was adopted to promote export industries.

322

Mohamed Z. Bechri and Sonia Naccache

4. The two-track strategy To implement his grand scheme of economic reconstruction, Bourguiba needed new competencies. In 1971, he appointed the then Governor of the Central Bank, Hedi Nouira as the new Prime Minister. The system that Mr. Nouira became in charge of had lost much in popular representation and support. The disastrous adventure of the previous decade showed clearly that the leadership was not immune from making terrible mistakes. But, the new government had no choice but to take some bold initiatives. In particular, an offshore company law was adopted in 1972, which facilitated the establishment of special industrial zones devoted to exports. 4.1. Offshore Company Law and export promotion With collectivism dead, the alternative could be either the adoption of laissezfaire or of a hybrid system combining both private initiative and government intervention. In practice, the Nouira team implemented a two-track strategy. The idea here was to create new opportunities, but without unduly destroying the pillars of the old system (Lau et al., 2001). The Mauritius experiment with the development of an offshore enclave for exporting industries in tandem with a protected onshore sector was receiving considerable attention and may have served as a model for Tunisian decision makers. The other alternative was to radically liberalize the system, as Chile successfully did in the mid-1970s. At that time, however, these reforms were met with suspicion in much of the rest of the world (Edwards, 1998). Hence, they could have been easily discarded as extremist or irrelevant for the Tunisian context, if they were at all contemplated. In addition, Tunisia lacked an economic team capable of providing a coherent master plan for economic liberalization. Unlike Chile’s ‘‘Chicago boys,’’ the members of the Nouira team were mostly from legal and engineering background. Hence, they lacked a formal training in economics in the Anglo-Saxon tradition that favors market mechanisms. 4.2. Financial repression and exchange controls The development strategy described so far was implemented in a context of ﬁnancial repression. Banks were tightly regulated. Interest rates were administratively set at below-equilibrium levels, which discouraged savings, and led to credit rationing and misallocation of resources. Exchange controls were also prevalent, but as indicated in Figure 1, overvaluation of the local currency was limited. The real effective exchange rate of the Tunisian dinar was quite stable during the 1970s. The new Tunisian strategy of amalgamation of private initiative and export promotion, on the one hand, with heavy government intervention, on the other, was viewed by some economists as fundamentally inconsistent (Moussa, 1995).

The Political Economy of Development Policy in Tunisia

Figure 1.

323

Nominal effective exchange rate (NEER) and real effective exchange rate (REER) of the dinar 200 150 100 50

NEER REER

19 70 19 74 19 78 19 82 19 86 19 90 19 94 19 98 20 02

0

Source: International Financial Statistics, the IMF

This ‘‘inconsistency’’ will be better understood with the ﬂowing analysis of interest groups. 4.3. Role of interest groups To assess this role, we start with an assessment of groups’ expected success in collective action, followed by an estimate of their actual success. 4.3.1. Expected success Producer groups are classiﬁed in the following three categories: (1) non-tradables, (2) import substitution, and (3) exportables. Based on the theory of collective action presented above, the expected degree of success of these groups in the Tunisian context, is presented in Table 3 (low, L; medium, M; and high, H). 4.3.2. Actual success Two indicators were selected to assess the actual success of different interest groups, namely, (1) the effective rate of protection and (2) the ratio of access to bank credit (Nugent, 1989). The latter indicates the ratio of the sector’s share in bank total credit to its share in GDP. Accordingly, a ratio higher than 1 means that the activity gets more credit than its contribution in economic activity would justify. This may just mean that the sector is more capital intensive or it gets less government funds. Nevertheless, the ratio would indicate, to a large extent, the political inﬂuence of the group. If an activity scores L according to one indicator and H according to the other, the score that appears in Table 4 will be M. Otherwise, the ﬁrst score refers to the ﬁrst indicator (protection) and the second refers to the credit ratio. M/H, as an example, indicates that the activity has an average effective protection, but higher than average credit ratio.

Mohamed Z. Bechri and Sonia Naccache

324

Table 3.

Group characteristics and expected success in collective action Size/Concentration

Non-tradables Construction

Non-tourism services Import substitution Agriculture

Food processing Wood and paper Construction products Exportables Textile and leather

Mechanical and electrical Chemicals Mining Tourism

Homogeneity

Expected Success in Collective Action

Sector dominated by a few large enterprises, concentrated in the main cities very large number, spread all over the country

Homogeneity of goals

H

Lacking homogeneity of background, difﬁcult to mobilize

L

Very large number, members spread all over the country Small number, relatively concentrated Limited number of ﬁrms Large ﬁrms are dominant

Heterogeneity of background and objectives Similar objectives

L

Concentrated activity

Small number Concentrated and public sector dominated Public sector activity Large number

H

Homogeneous goals State-owned enterprises (cement factories and others)

H H

Different goals according to whether the activity is import substitution or export oriented Shift overtime from import substitution to exports Homogeneous

M

Homogeneous Heterogeneous

H M

H H

The comparison between expected success (Table 3) and actual success (Table 4) shows some differences. Textile and leather turn out to have been more successful than expected during the 1970s. Textiles and leather ﬁnished products were highly protected in the local market, and the activity’s access to bank credit was higher than average. This is not surprising since the sector was boosted by the Offshore Company Law of 1972 after which it became the main exporting activity in manufacturing. For some other activities, however, actual success turned out to be lower than expected. This is the case for construction, food processing, mechanical and electrical equipment, chemicals, and mining. This result is surprising for construction since this activity is of low risk to banks and it enjoys natural protection from foreign competition. During this period, however, the ﬁnancing of

The Political Economy of Development Policy in Tunisia

Table 4.

325

Group strengtha

1971–1980

1981–1986

1987–1989

1990–1996

1997–2004

Non-tradable Construction Non-tourism services

L L

L L

L L

L L

L M

Import substitution Agriculture Food processing Wood and paper Construction materials

L L M H

L M L M

L M M M

L H M M/H

M M H/M H

Exportable Textile and leather Mechanical and electrical Chemicals Mining Tourism

H M M L M

M M M M L

M/L H H L M

M H M M H

L M M L H

a

Group’s actual success (H, high; M, medium; L, low) based on effective protection and bank credit access ratio, except for non-tradables, mining, and tourism, where the estimate is based on the second indicator only.

this activity was entrusted to the Caisse Nationale de l’Epargne Logement (CNEL). Regulation prevented other banks from ﬁnancing real estate projects. In addition, some State-owned companies were very active in this sector, and their reliance on government funds made them less dependent on banks. The weakness for food processing is due to low protection for social considerations (providing food at an affordable price to consumers) and to a low credit ratio, since banks’ preference was for export-oriented activities. As regards mechanical and electrical equipment, the activity was based originally on import substitution, but by the late 1970s it was diverted to exports. Its medium score on actual success in collective action is due mainly to its low protection. This openness to foreign competition and its relative success in gaining access to bank credit were instrumental in the sector’s successful transition into exports. Low protection was also observed for chemicals, which made sense given its export orientation (the main units in this sector are joint ventures between the Tunisian government and foreign parties based on locally produced phosphate). With respect to mining, its low credit ratio may underestimate its true political inﬂuence as a public monopoly of the Office des Mines, which gives it privileged access to government funds. Tunisia’s economic success of the 1970s allowed major stakeholders to develop vested interests in the new system. The expansion of the government funded ‘‘The Industrial Promotion and Decentralization Fund’’ (FOPRODI) since its inception in 1974, created a new class of entrepreneurs. Public-sector employees beneﬁted from the increase in public enterprises, the expansion in the

326

Mohamed Z. Bechri and Sonia Naccache

civil service and better wages. Moreover, farmers beneﬁted from the liberalization of agricultural prices and the general economic boom. With the passage of time, however, some drawbacks of the strategy became increasingly clear: a lack of integration between the onshore and offshore sectors, serious misallocation of resources, technological backwardness, poor productivity, and lack of consumer choice. But, as long as the two-track strategy seemed to work, the pressure to liberalize remained limited. In 1981, Mr. Nouira had to be replaced for health reasons. The new government, moreover, had to maneuver in a situation of political chaos owing to the President’s deteriorating health, aggravated by economic hardship arising from the country’s deteriorating terms of trade and economic mismanagement. 5. Economic populism and the crisis-triggered reforms of the 1980s The President appointed Minister Mohamed Mzali, a close conﬁdant, as the new head of the government. But, Mr. Mzali, a literary critic turned politician, lacked the management experience of his predecessor. Moreover, with the memories of the 1960s still present, Bourguiba could not afford to neglect the economy once again. For this reason, he instructed the appointment of Mansour Moalla, a particularly competent banker, to the position of Finance and Planning Minister. With good economic management in place, Bourguiba may have thought that Mzali would be able to take care of the rest. But he was wrong. A demanding national trade union, political unrest led by the Islamist movement, and Palace intrigue, all took their toll on what turned out to be a lame-duck government. 5.1. Political and economic setup Mr. Mzali had ﬁrst to face some political realities. To gain credibility, he needed to get out of jail the UGTT leadership and other political activists from leftist and Islamist groups. This was obviously no easy task, given the President’s increasingly volatile character. Step by step, however, Bourguiba’s amnesty was secured for all of them. But this was not enough. With no political power base of his own, Prime Minister Mzali was vulnerable from the beginning to all types of pressure from the Palace as well as from the national trade union. In particular, UGTT leader Habib Achour withdrew his membership from the ruling Party and refused to cooperate on the issue of food subsidy reform. The crisis of conﬁdence ended up in a violent confrontation in 1984 with the eruption of the so-called ‘‘bread riots.’’ Cornered, Mr. Mzali had no choice but to give in to populist demands. The government decreed a wage increase that resulted in an annual inﬂation of 14% in 1982. It was the ﬁrst time since the crisis of 1964 that Tunisia experienced a double-digit rate. And, both the ﬁscal and current account deﬁcits increased,

The Political Economy of Development Policy in Tunisia

327

reaching alarming levels of 6.6% and 7.1% of GDP, respectively, by 1985. The situation was aggravated with the deterioration of the terms of trade, which started to decline in 1983, and decreased by 6.2% in 1985 and 13.8 % in 1986. Despite the gravity of the situation, however, the government failed to respond with appropriate corrective actions, such as a devaluation of the dinar and ﬁscal discipline. The presence of two antagonistic points of view within the government itself made action impossible. ‘Super’ Finance and Planning Minister Mansour Moalla was pressing for the elimination of food subsidies. But another wing, headed by Mr. Azzouz Lasram, the Economics Minister, favored a gradual approach. With the so-called ‘‘bread riots’’ erupting in other countries in the region, Prime Minister Mzali should have been well aware of how risky the elimination of food subsidies would be for his government. But the continuing pressure on the budget left him with no other choice but to overhaul the system. To be sure, there are ways to minimize the risks of social disruptions in a period of food subsidy reform. According to IMF (2000), this could be achieved by: (1) adopting a gradual approach, (2) choosing the appropriate timing, (3) providing an adequate compensation for the poor, and (4) succeeding in a public relations campaign to sell the idea to the public. But the decision that the Mzali government announced on January 1984 did not fulﬁll any of these prerequisites. First, the price increase (of>100% for bread) was particularly high. Second, political tensions were rising at that time, which saw demonstrations by Islamists who were emboldened by the success of the Iranian revolution, and severed relations with labor, following the resignation of the UGTT leader from the Neo-Destour. Third, there was no compensation for the poor, and the announcement that was made, in this respect, lacked credibility. Fourth, the campaign that was launched to explain the change in policy to the public was unsuccessful. It should be made clear that the failure to take these factors into account was the primary responsibility of the political leadership (the ruling party and government political advisors), not the economic team. Street riots erupted at the announcement of the increase in prices. And Mzali opponents’ next move was to sabotage public security. The then Interior Minister Driss Guiga, who saw himself as a potential Prime Minister, owing to his close ties with Bourguiba’s wife, gave his instructions to the security forces not to intervene. As a result, street violence increased. Bourguiba’s immediate reaction was to cancel the price increase, blaming once again members of the government for bad counseling. But, he renewed his support to the Prime Minister and the anti-Mzali clan failed. The return to square one was met with apparent relief by the poorest strata of the population, which would bear the full brunt of the price increase. While the rich would not be much affected, the middle class was the main loser since it had to pay a disproportionate share of the subsidies. But, maintaining food subsidies was unquestionably bad news for the macroeconomic balances. Both the ﬁscal

328

Mohamed Z. Bechri and Sonia Naccache

and the current account deﬁcits widened and under the circumstances, the President appointed a new Prime Minister in the summer of 1986. The appointment of Mr. Rashid Sfar as the new Prime Minister was testament to Bourguiba’s ability to take appropriate decisions, even at an advanced age. Mr. Sfar was a man of integrity, with long experience in the Tunisian civil service. More importantly, his ﬁnancial background should have been of primary importance to the President. But, Sfar had arrived too late. In addition to the economic mess it inherited, the new government (like the previous one) had to take into account the conﬂicting demands from within Bourguiba’s entourage and street unrest. Meanwhile, the Central Bank’s foreign reserves declined to a critical level, which sent the government begging the International Monetary Fund (IMF) for help. The reform package to be adopted following the signing of a Stand-by Agreement with the IMF would soon reach the political arena. In a chaotic situation, the Interior Minister was appointed to head the government. Within weeks, he took advantage of his new position to grab power. On November 7, 1987, President Bourguiba was removed on grounds of ‘‘senility.’’

5.2. Crisis-triggered reforms of 1986 According to Bates and Krueger (cited in Edwards (1998)) ‘‘there is no recorded instance of the beginning of a reform program at a time when economic growth was satisfactory and when the price level and balance of payments situations were stable. Conditions of economic stagnation y or continued deterioration are evidently prerequisites for reform efforts.’’ In this respect, Tunisia and other MENA countries ﬁtted well the stereotype. Indeed, as indicated in Table 5, all adjusting MENA countries adopted reform in a situation of crisis. In Tunisia and Morocco, both the budget and current account deﬁcits reached alarming levels of >7% of GDP in the pre-adjustment period. Likewise, in Jordan, the current account deﬁcit was 10.5% of GDP and inﬂation rose to 25.7% in 1989. Furthermore, in Egypt, the budget deﬁcit reached 5% of GDP while inﬂation reached 20% in 1991. The peaceful transfer of power that took place three months after Tunisia adopted the structural adjustment plan provided ideal conditions for the authorities to go forward with economic reforms. Indeed, most of the 16 hypotheses about reforms listed in Isfahani (2000) were observed. The crisis of August 1986 was a powerful trigger for radical reforms, since it created a general consensus on the need for bold action. The emergence of winners, especially in export activities, enhanced the sustainability of reforms. The IMF was ready to provide the external support needed. The newly established regime beneﬁted from an exceptional honeymoon period and managed successfully to domesticate local opposition, especially from the national trade union. Finally, several attempts were made to compensate the losers.

The Political Economy of Development Policy in Tunisia

Table 5.

329

Adjustment programs in MENA countries

Country

Pre-adjustment

Adjustment Year

Postadjustment

Recent Situation

Morocco

1982

1983

1984–1985

Average: (2000–2003)

Budget deﬁcita/GDP (%) Current account deﬁcit/GDP (%) Inﬂation (%)

–11.4

–7.7

–7.0

–4.1

–12.5

–7.3

–5.4

–3.6

10.7

6.3

9.6

Tunisia Budget deﬁcita/GDP (%) Current account deﬁcit/GDP (%) Inﬂation (%)

1985 –5.0

1986 –7.1

–6.2

–7.1

–0.2

–2.7

7.2

6.2

7.7

2.6

Jordan Budget deﬁcita/GDP (%) Current account/ GDP (%) Inﬂation (%)

1988 –9.0

1989 5.8

1990–1999 +1.2

–6.1

+10.5

–9.6

6.5

25.7

6.8

Egypt Budget deﬁcita/GDP (%) Current account / GDP (%) Inﬂation (%)

1990 –5.7

1991 –4.9

1992–1997 +1.0

+0.4

+5.7

+1.7

0.0

16.7

19.7

10.2

3.05

1.64

1987–1989 –4.2

n.a.

–2.2 (2000–2002) +6.2 1.66 2.1 (2002–2002)

n.a. Not available. Source: International Financial Statistics, IMF, January 2005. a For both the Government budget and the current account, a sign (+) means a surplus.

One should also notice that the new President was in a privileged position, at the Interior Ministry, to monitor the deteriorating situation of the country since the early 1980s. Once in charge, he should have been aware of the need for institutional arrangements to ensure better governance. First, he turned down suggestions to create a new political Party, preferring instead to ‘‘rejuvenate’’ the Neo-Destour. Second, a ‘‘National Charter’’ was signed in 1988 by all major political parties, including the Islamists. Moreover, coordination was strengthened with centralized power, a domesticated trade union, and co-opted business associations. The latter had long grown used to protection but eventually realized that they had no other choice but to adapt to the new orientation. The newly appointed President of UTICA as well as the entire Bureau was among the most

330

Mohamed Z. Bechri and Sonia Naccache

successful Tunisian exporters and therefore favorable to greater openness even though several of them were also involved in import-substitution activities. As regards credibility, the country still lacked institutional enforcement mechanisms. But, the general consensus to open up the economy, the appointment of pro-export leaders within UTICA, and the signing of the Stand-by Agreement with the IMF in 1986, all made return to protection unlikely. The stabilization program succeeded rapidly. In particular, the strong devaluation was accompanied by restrictive monetary, ﬁscal, and income measures. As a result, the nominal devaluation translated into a real depreciation of the dinar by 14.5% in 1986 and an additional 8.6% in 1987. The current account turned into a surplus in 1988, followed by moderate deﬁcits in the subsequent period. Structural reforms were also adopted to liberalize the price, investment, and trade regimes. As a result, private investment, shifted from non-tradables to exportables (especially manufacturing) and to a lesser extent to importables (Bechri and Lahouel, [1999]). Nevertheless, deeper institutional reforms were hesitant and in some instances clearly lacking, especially in critical areas like public administration, education, banking, telecommunications and the labor market (Bechri, 1999). 5.3. Post-adjustment uncertainties (1992– 1994) The post-adjustment years (1992–1994) witnessed some policy reversals, mainly in foreign trade. Temporary surcharges were established in 1992, on some manufactured consumer goods like clothing and footwear, ‘‘technical controls’’ under the so-called Cahiers des Charges regime were imposed on other imports. These reversals sent an unfortunate signal about the government’s commitment to reforms by indicating to the private sector that protection can always be restored by one type of restriction or another. These reversals and the lack of institutional reforms help explain the decline in private investment from 13.5% of GDP in 1992 to 12.7% in 1993 and 11.6% in 1994. The decline in the volume of private investment was particularly pronounced in 1993 for food processing (–20.0%) and in 1994 for textiles and leather (–13.0%), mechanical and electrical equipment (–20.0%), and chemicals (–27.3%). Foreign direct investment (outside energy) was also weak, o1% of GDP. This decline occurred despite the adoption of a uniﬁed investment code in 1993 and the removal of import licensing in 1994. Thus, the authorities realized that additional bold measures were needed. In 1995, Tunisia was the ﬁrst SouthMediterranean country to sign an Association Agreement with the European Union, which provided for the establishment of a free-trade area. 5.4. The association agreement with the european union The Tunisian ofﬁcial rationale in this respect was that as the ﬁrst signatory from the South-Mediterranean region, Tunisia would beneﬁt the most from: (1) EU ﬁnancial support, (2) industrial relocation, (3) enhanced credibility, by locking

The Political Economy of Development Policy in Tunisia

331

the economy into an irreversible set of reforms, and (4) an advantage in terms of determining the content of the agreement (Bechri, 1999). But, one should notice that this rationale might also be valid for countries like Egypt, Morocco, and Jordan. So to understand Tunisia’s decision, we need to look at some differences in interest group characteristics. First, the three decades-old export-oriented strategy in Tunisia ended up creating a whole constituency that had a stake in openness. For example, according to the most recent available estimates for the textiles and clothing sector, out of a total of 1540 exporting ﬁrms, only one-third (536 ﬁrms) is foreignowned, 333 ﬁrms are partnership companies, and the remaining 671 ﬁrms are locally owned. Other sectors, like mechanical and electrical equipment shifted gradually from import substitution in the 1970s to export by the late 1980s. This is a sign of a better performance, but it also means that with the passage of time these activities became more interested in export promotion than in protection. The second major difference relates to the domesticated labor union. UGTT was allowed to renew its activities in 1988, but under the condition of a cooperative leadership. As a result, strikes and violent protests became rare. Nevertheless, both UTICA and UGTT were concerned about the costs of adjustment. These concerns may explain why Tunisia chose a long implementation period of 12 years for tariffs removal. So far, the implementation of the association agreement is proceeding as planned and without major disruptions. A technological upgrading program Programme de Mise de Niveau (PMN), funded by the European Union, started in 1996. As of April 2003, upgrading was approved for 1448 ﬁrms with investments amounting to 2.4 billion dinars, of which 24% was allocated to agribusiness, 20% to construction products, 18% to textiles and clothing, 14% to mechanical and electrical equipment, 8% to chemicals, and 5.2% to leather and other industries. From a labor point of view, things also went smoothly. No major bankruptcies took place, worker layoffs were rare and there have even been reported instances where restructuring led to job creation. The long implementation period of the free-trade agreement, however, also had major economic drawbacks. First, by announcing the opening up of the economy to European competition by 2008, the agreement sent a wrong signal to business about the urgent need to adjust. Second, restructuring in import-substitution activities remained limited. This is harmful to consumers and did not bode well for the competitiveness of these industries. Third, the bureaucracy and other entrenched interests (especially protected business and trade unions) managed to postpone badly needed reforms in crucial activities like trading boards, telecommunications, and banking. Trading boards in Tunisia have a monopoly on foreign trade in speciﬁc products like edible oil, cereals, tea, coffee, and sugar. The Office de l’huile clearly outlived its usefulness, since its monopoly on olive oil exports was lifted in the early 1990s. The same could also be said about the Office du commerce,

332

Mohamed Z. Bechri and Sonia Naccache

following the withdrawal of sugar from the list of subsidized items. But, the Office des Cereales continues to have a monopoly on the imports of cereals that is unlikely to be eliminated as long as food subsidies are in place. As regards telecommunications, this sector used to be managed by the government until the mid-1990s. This resulted in poor quality of services and high costs. In an attempt to modernize the sector, the authorities created TunisieTelecom, as an independent entity from the ministry. However, the public monopoly failed. Thus, Tunisia ended up lagging behind countries like Morocco, which successfully liberalized its telecommunications in the 1990s. In 2002, a second license for GSM services was granted to the Egyptian company ORASCOM and further opening of the sector to competition is expected, including the withdrawal of the State from Tunisie-Telecom. The banking sector, however, remains dominated by State-owned institutions. Costs for banking services remain high by international standards, and non-performing loans (NPLs) a major source of fragility of the system. Despite the introduction of prudential regulations in 1992, their ratio remained at 22% in 2002–2003, and increased to 24% in 2004 (compared to an international standard of 6%). There has also been a merger of three public banks. One public sector bank (UIB) was sold to the French Societe Generale in 2004, and the Banque du Sud is expected to follow suit in 2005. But these attempts remain insufﬁcient to create an internationally competitive banking sector.

6. Conclusion During the last ﬁve decades, Tunisia clearly outperformed other MENA countries. This exceptional performance in economic and social development owes much to its visionary leadership, its early emphasis on modernization and economic progress, and its better technical abilities to implement economic policy. The leadership averted Tunisia’s involvement in regional conﬂicts that proved costly to other countries in the region like Egypt, Syria, Iraq, Algeria, and others. The Tunisian experience showed, indeed, the instrumental role of leadership in isolating traditional groups, and in rallying the support of women, trade unions and the other liberal fringes of the population. Once freed from the shackles of backward teachings of the traditional establishment, constituted around Zeitouna, Tunisia was able to improve the status of women, modernize its education, and adopt population control. This is in contrast to the other MENA countries, such as Egypt, where Al Azhar and similar institutions continue to inﬂuence public opinion and to oppose serious attempts (by parliament and other institutions) to take action in favor of women and educational reform. The gains were very important: lower population growth and higher GDP growth, compared to MENA averages. The combined effect was an acceleration of annual per-capita GDP from 2.1% in the 1960s, 2.5% in the 1990s, and

The Political Economy of Development Policy in Tunisia

333

about 4% at present, which gives the country the status of a top performer in the region. Indeed, demographic explosion turned out to be a chief culprit in the other MENA countries, as Bourguiba has been arguing for decades. The decline in fertility was slow. As recently as 1998, the average for Jordan, Egypt, Morocco, and Syria, was 3.5, against only 2.2, for Tunisia. The implication was enormous. Egypt’s population, as an example, would have been about 54 millions only, against 68 millions in 2000, had it experienced the Tunisian demographic growth of the last ﬁve decades. Population control and social modernization, therefore, would seem to have played an important role in explaining Tunisia’s high annual average growth of 6.0%, compared to Egypt’s 4.6%, Morocco’s 4.5%, and Jordan’s 2.7% during 1985–2003. Better health and education created a more productive labor force, which was strengthened with better women participation. Tunisia also did particularly well in starting early its reforms, especially its policy shift from import substitution to export promotion as early as the early 1970s. Current laggards in the region currently face serious challenges, although the recent examples of Morocco and Kuwait regarding women status reforms indicate that action in this respect is still possible. The Tunisian case also shows the clear dangers of co-option. In the case of the trade union, the ruling Party ended up adopting UGTT policies not the other way around. In most MENA countries, Islamists groups in parliaments are nowadays the main obstacle to social and educational reforms. Even though no efforts should be spared in seeking cooperation from these groups, neither can reformist governments afford to be paralyzed by them. For years, Tunisia tried to bypass business opposition by adopting a twotrack strategy that is no longer feasible. The only option left for laggards in the MENA region today is to adopt a wide range of social and economic reforms. In this respect, due consideration should be given to sequencing. It may be prudent to start with actions that are expected to face less opposition. For reforms that involve huge redistribution, governments need to start by explaining the longterm beneﬁts to the public and provide appropriate compensation schemes for losers. References Achy, L. (2000), ‘‘Misalignment and exchange rate arrangement against the euro’’, Paper presented at the ERF Annual Conference, Amman, Jordan. Bechri, M.Z. (1999), ‘‘Institutional obstacles, reform uncertainty and Tunisia’s integration in the European Union’’, Paper presented at the Euro-Med partnership seminar in Cairo, February. Bechri, M.Z. and M.E.H. Lahouel (1999), ‘‘Trade liberalization and the behavior of private investment in Tunisia’’, Paper presented at the ERF Annual Conference, Cairo, October 28–31.

334

Mohamed Z. Bechri and Sonia Naccache

Bellin, E. (1989), ‘‘Tunisian industrialists and the state’’, World Development, Vol. 22(3), pp. 1994. Castanheira, M. and H.S. Isfahani (forthcoming), ‘‘Political economy of growth: lessons learned and challenges ahead’’, in: L. Squire and G. McMahon, editors, Explaining Growth, Washington, DC: International Economics Association. Edwards, S. (1998), ‘‘The political economy of unilateral trade reform in Chile’’, available on Professor Sebastian Edwards Website. Grissa, A. (1989), ‘‘The Tunisian state enterprises and privatization policy’’, Mimeo, unpublished. Isfahani, H.S. (2000), ‘‘Political economy of growth in MENA: a framework for country case studies’’, Global Development Network Working Paper. Isfahani, H.S. (2002), ‘‘Political economy of growth in Iran, 1963–2002’’, Department of Economics, University of Illinois. Khiari, S. (2000), ‘‘La place de l’UGTT dans le systeme politique Tunisien’’, Mimeo, unpublished. Lau, L., J.Y. Qiau and G. Roland (2001), ‘‘Reform without losers: an interpretation of China’s dual-track approach to transition’’, available on Stanford University website. Moussa, H. (1995), ‘‘Economic policy and economic development in Tunisia’’, Paper presented at the International Economic Association World Congress, December, Tunis, Tunisia. Nabli, M.K. and J.B. Nugent (1989), ‘‘Collective Action, Institutions and Development’’, in: M.K. Nabli and J.B. Nugent, editors, The New Institutional Economics and Development: Applications to Tunisia, Amsterdam: NorthHolland (Chapter 3). Nugent, J.B. (1989). Collective action in Tunisia’s producer’s organizations. in: M.K. Nabli and J.B. Nugent, editors, The New Institutional Economics and Development: Applications to Tunisia. Amsterdam: North-Holland. Rodrik, D. (1998), ‘‘Why is trade reform so difﬁcult in africa?’’, in: I. Zubair and M. Khan, editors, Trade Reform And Regional Integration in Africa, The International Monetary Fund. World Bank (2003), Better Governance for Development in the Middle East and North Africa.

CHAPTER 11

Algeria’s Macroeconomic Performances from 1962 to 2000 Mohamed Abdelbasset Chemingui and Moataz Mostafa El-Said Abstract The paper examines Algeria’s economic growth performance with an effort to explain its performance over the period 1962–2000. The paper considers prevailing economic policies and changes introduced to the economic environment over the same time period. In addition, the paper conducts a Solow’s accounting exercise to break down growth in output into growth in capital, labor, and total factor productivity (TFP) and at key determinants that influence Algeria’s growth performance. Over the past four decades, Algeria’s growth performance has been characterized by high fluctuations and generally slower growth than that in the neighboring developing countries. The poor economic growth performance that Algeria experienced is a result of a slow growth in productivity which can be attributed to the following factors: (i) the inefficient management of the country’s natural resources; (ii) the country’s macroeconomic policies, specifically price deregulation and economic reforms aimed at increasing investment; (iii) structural impediments limiting the private sector contribution to the country’s economic activities; and (iv) an overall economic environment dominated by the public sector, a weak financial sector, slow and non-transparent privatization process, and other factors that limited the progress of economic reforms. Keywords: economic growth, productivity, technological change, government policy, trade policy, foreign exchange policy JEL classifications: F43, O47, O33, 038, 024

Corresponding author. CONTRIBUTIONS TO ECONOMIC ANALYSIS VOLUME 278 ISSN: 0573-8555 DOI:10.1016/S0573-8555(06)78011-7

r 2007 ELSEVIER B.V. ALL RIGHTS RESERVED

336

Mohamed Abdelbasset Chemingui and Moataz Mostafa El-Said

1. Introduction Growth has been one of the most active ﬁelds of research in economics since the mid-1980s. Economists seek answers to explain why there are very rich countries at the same time as very poor countries. For example, why do rates of economic growth vary substantially across countries? Finding answers as to what are the sources of economic growth and what can be done to improve it can help explain how countries can grow from rags to riches, and how economic policies can play a role in achieving rapid economic growth. In the 1950s and 1960s, it was accepted that inventing massively was sufﬁcient for achieving high economic growth rates. However, the 1970s showed that investment alone was not enough, which led to a search for the sources and determinants of economic growth. The economics of ‘‘ideas’’ and of human capital and the economics of technology were introduced to growth theories and much of the 1990s was spent in quantifying and testing those theories. In this paper Algeria’s economic growth performance is examined over the period 1962–2000. The paper makes use of a Solow accounting exercise to break down growth into growth in capital, labor, and total factor productivity (TFP). Over the past four decades, Algeria’s growth performance has been characterized by high ﬂuctuations and generally slower growth than that in neighboring developing countries. At times, Algeria even experienced growth rates lower than those of poor Sub-Saharan African countries. Four different periods in Algeria’s growth over the past four decades can be distinguished: (a) a period from 1962 to the early 1970s characterized by high growth rates averaging 8.3 percent per year; (b) a period of more volatile growth from the early 1970s until 1986, during which growth rates averaged at 5 percent; (c) a period from 1987 to 1994 during which growth rates were negative in some years and averaged only 0.2 percent and (d) a fourth period since the mid-1990s when the economy exhibited modest but progressively positive growth (averaging 3.2 percent per annum). Analyzing the main determinants of the Algerian growth achievements was a ﬁrst step toward identifying what needs to be done to make growth more sustainable. The empirical literature suggests a wide range of growth correlates. This list includes research and development (R&D), technological transfer from abroad mainly through foreign direct investments (FDI), population growth and unemployment rates, the level of human capital development, level of physical investment, infrastructure development, degree of factor allocations, government size, ﬁnancial market development, natural resource endowment, institutions, politics, technical barriers, income distribution, economic base, rates of returns on investment in physical and human capital, level of integration in the world economy, and the level of development of market institutions. The empirical literature indicates that most of these determinants of growth are as important as the proximate factors of growth namely physical capital, labor, and the efﬁciency with which these factors are combined. In this respect, a detailed examination of the determinants of

Algeria’s Macroeconomic Performances from 1962 to 2000

337

Algeria’s slow growth performance was carried out. The list of factors that impede Algeria’s economic growth include low labor productivity, high inﬂation rates, high levels of external debt, weak institutional development, non-transparency in administrative practices, inadequate infrastructure, and underdeveloped ﬁnancial markets. Furthermore, growth patterns are believed to be inextricably linked to various economic policies such as ﬁscal policy, labor policy, trade policy, exchange rate policy, interest rate policy, government spending in human capital, and infrastructure development and competition policy, among others. Algeria introduced substantial reforms in several of these respects beginning in 1986, and has since then undertaken a number of policy and regulatory changes to liberalize what was previously a highly protected and centrally planned economy. These changes included progressive liberalization of trade policy, privatization of state-owned enterprises (SOEs), and ﬁscal policy reform. The format of the paper is as follows: Section 2 of the paper considers an overview of economic growth in Algeria, while Section 3 presents a growth accounting exercise to quantify the sources of economic growth and to determine the contribution of each factor to the growth of output. This is followed by Section 4, which looks at some structural and economic policy obstacles behind Algeria’s sharp fall in economic growth rate and TFP. Finally, Section Y5 concludes with an assessment of the Algerian experience. 2. Overview of Algeria’s economic growth Algeria represents a rather special case among countries of the Middle East and North Africa (MENA) region. Although endowed with abundant natural resources, mainly crude oil, Algeria was not successful in fully reaping the potential exempliﬁed by the country’s low GDP growth and high unemployment rates. Compared to Tunisia, a neighboring MENA country, Algeria’s income per capita was 60 percent higher than that of Tunisia in 1960. However, 30 years later, Tunisia’s per capita income is higher than that of Algeria by 7 percent even though Tunisia is not endowed with natural resources (Tunisia is not an oilexporting country) like Algeria. To provide an answer as to why this is the case, the paper looks at Algeria’s growth path over the period 1962–2000 and considers the prevailing economic policies and changes introduced to the economic environment over the same time period. Algeria’s growth record over the 1962–2000 period can be broken down into four subperiods in terms of the country’s economic growth and economic policy orientations: 1962–1973: 1974–1986:

A period of high and stable economic growth. A period of macroeconomic instability, difﬁcult adjustments, and more volatile growth rates, and falling GDP per capita growth rate (Figure 1).

Mohamed Abdelbasset Chemingui and Moataz Mostafa El-Said

338

Figure 1.

GDP per capita growth rate

40 30

Series1

10 1999

1997

1995

1993

1991

1989

1987

1985

1983

1981

1979

1977

1975

1973

1971

1969

1967

-20

1965

-10

1963

0 1961

Percent

20

-30 Year

1987–1994: 1995–2000:

A period of structural adjustment programs coupled with declining economic growth. A period of modest economic growth and a new wave of economic reforms.

2.1. 1962– 1973: High but unstable economic growth Since independence in 1962, Algeria’s economic orientation followed a model of a socialist economy dominated by public sector industrialization-based development strategy. In general, the Algerian government favored state monopolies in a centralized economic planning system. In terms of economic growth, by the end of the 1970s, Algeria was on a path of high and ﬂuctuating growth in per capita income, with total investment accounting for a large share of national income (about 45 percent). Hydrocarbon exports provided the bulk of the revenue that ﬁnanced these efforts, and helped create a large network for public enterprises (PEs) and an extensive infrastructure base. The period was marked by improving social indicators, falling illiteracy, increasing life expectancy, and high school enrollment. Following an industrialization-based development strategy led to the allocation of a large proportion of total public investment (about 54 percent) toward manufacturing, mining, energy, and hydrocarbons, while the share for agriculture remained relatively constant at around 10 percent. The heavy industry, notably hydrocarbons, played a major role in Algeria’s economic activity, as it contributed to the construction of the industrial sector by endowing it with raw materials and provided agriculture with inputs and energy. However, as is the case with most state-run enterprises, Algeria’s public sector quite often followed contradictory political, ideological, economic, and social objectives. In general, public ﬁrms lacked managerial autonomy and had to attend to other objectives like having to aid in reducing regional disparities, absorb the growth in the labor force, and take care of the social necessities of the

Algeria’s Macroeconomic Performances from 1962 to 2000

339

employees (consumer goods with low costs, habitation, health, transportation). To a large degree the private sector was crowded out by state-owned enterprises. However, some private sector activities existed, mainly relying on subcontracting for public sector activities. In summary, Algeria was able to achieve high growth rates during this period as a result of a mix of factors, mainly from an intensive public investment program carried out by the state and an expansion in oil and energy production that was coupled with increasing international oil prices. 2.2. 1974– 1986: Macroeconomic instability, difficult adjustments, and volatile economic growth By the late 1970s, the inefﬁciencies of the central planning system became apparent. For agriculture, state farms, which accounted for the major part of Algeria’s agricultural output, suffered from low productivity growth and stagnating crop yields. Inefﬁciencies in state management were evident as many large public investment projects took unnecessary long time for completion and a large number of new industrial plants were running substantially below capacity because of weak domestic and foreign demand. In addition, the inefﬁciency of investment was reﬂected in an abnormally high incremental capital/output ratio (Table 1) which, despite investment levels averaging 40 percent of GDP, resulted in annual real per capita output growth of slightly o2 percent on average. Since the beginning of the 1980s, Algeria’s macroeconomic balances deteriorated rapidly in large part because of falling oil prices. By 1986, the deterioration reached a climax as export revenues from oil collapsed as a result of a Saudi-inspired lowering of world prices and depreciation in the value of the US dollar. Algeria’s terms of trade declined by almost 50 percent. The Government of Algeria had to borrow to maintain consumption levels, while the volume of imports declined by almost one-third between 1985 and 1987. In the meantime, investment was reduced and the level of domestic industries activities decreased signiﬁcantly. This includes almost all sectors of the economy with the exception of hydrocarbon and agriculture. Toward the period 1986–1988, real GDP growth averagedo1 percent per year (compared to 5.4 percent per year in 1978–1985). Unemployment and inﬂation rates increased, and per capita consumption declined. An acute social and political crisis was brewing, which Table 1.

Investment shares, capital/output ratios, and per capita GDP growth 1975–1976 1977–1978 1979–1980 1981–1982 1983 1984 1985

Gross investment/GDP (percent) Capital/output ratio Per capita GDP growth (percent)

44.0 3.5 3.5

49.5 4.5 3.9

41.0 5.5 1.3

37.0 7.0 1.6

Source: Authors’ calculations on the basis of data from World Bank (1994).

38.0 35.0 33.0 8.0 8.0 6.0 1.9 2.2 2.3

340

Mohamed Abdelbasset Chemingui and Moataz Mostafa El-Said

prompted the authorities to initiate a broad-ranging program of macroeconomic stabilization and structural reform. 2.3. 1987– 1994: Structural adjustment programs and economic recession period The adoption of the structural adjustment and macroeconomic stabilization program became necessary to achieve macroeconomic stability. By 1987, the program was initiated with an aim to establish the ‘‘right’’ conditions for sustainable long-term growth and at the same time to correct the macroeconomic imbalances and price distortions, and to contain inﬂation in the short term.1 In addition, reforms that covered the legal and institutional framework of the productive sector and the factor markets were introduced. These aimed at creating an incentive structure to stimulate supply response and to provide a source of competition for PEs, especially from the private sector. The reforms also included a program to restructure PEs and the banking sector. The structural reforms encompassed an overhaul of the incentive framework, including price, tax, and trade reforms. In January 1992, a major tax reform was enacted; a value-added tax (VAT) was introduced to simplify the indirect tax structure, while uniﬁed corporate and individual income taxes replaced a series of tax schedules and special tax regimes. In addition, the tax reform included a reduction in the maximum tariff from 120 percent to 60 percent, and a uniﬁcation of the compensation tax rates on imports and domestically produced goods. Nevertheless, the reform efforts remained ineffective because of signiﬁcant distortions that remained. While the tariff structure became more neutral, the overall incentive structure continued to distort border prices and to accord a relatively high rate of protection to domestic PEs. In addition, two key distortions remained: an overvalued exchange rate and a negative real interest rate. Despite the progress achieved in implementing reforms, the economic crisis persisted. Weak economic growth lowered per capita consumption and exacerbated unemployment. Over the 1985–1992 period, aggregate GDP growth averaged only 0.4 percent annually, and 1993 and 1994 were two years marked by very poor economic performance.2 The only exception came from the agriculture sector, averaging an encouraging 5.4 percent growth rate per year. This was

1

The priority objectives of structural reform included:

Promoting private sector development; Reforming and restructuring PEs; Deepening agricultural reform; Developing a competitive ﬁnancial sector, and a well-functioning labor market; Strengthening the social safety net, and adapting it to the needs for a market economy; and Integrating the economy into world markets.

2 During the 1990–1994 period, annual real growth was negative, averaging 0.9 percent (IMF, 2000).

Algeria’s Macroeconomic Performances from 1962 to 2000

341

taking place while value added by the core ‘‘enterprise sector’’ (industry, construction, and services) was declining by an average of 3.2 percent per annum. Per capita private consumption declined annually by about 3 percent over the 1985–1992 period. The industrial sector continued to be dominated by large, inefﬁcient, and heavily protected PEs, enjoying monopoly power and access to ofﬁcial foreign exchange and domestic credit at subsidized rates. Privatization prospects remained difﬁcult as the economy stagnated and the unemployment record worsened (the number of unemployed workers tripled over that period) and the ofﬁcial open unemployment rate reached an estimated 21 percent. In summary, the adopted reforms had at best a short-lived positive impact on economic performance. 2.4. 1995– 2000: A new wave of economic reforms and economic growth With the implementation of the 1994 reform program supported by the World Bank and the IMF, Algeria’s economic performance started to show signs of improvement. The program was put in place following a severe external payment crisis, in the wake of the oil price drop in 1993. The Government introduced swift stabilization and adjustment measures, including a strong ﬁscal adjustment; tight monetary policy; an active exchange rate regime; and price liberalization. The program was accompanied by a debt rescheduling agreement with the Paris and London clubs, and the initiation of structural reforms, including privatization of a number of PEs. To date, the main parameters of the reform program remains in place. This reform program has been successful in bringing about macroeconomic stability. Inﬂation dropped from 39 percent in 1994 to 5 percent by the end of the 1990s. The ﬁscal deﬁcit was signiﬁcantly reduced (from a high of 8.7 percent of GDP in 1993 to a surplus of around 2.5 percent by 1997) with the exception of 1998 when the state budget recorded a deﬁcit of 4 percent of GDP because of the sharp drop in oil prices. However, as oil prices started to recover along with signiﬁcant expenditures cuts, the ﬁscal stance was reversed yielding a close-to-balanced budget by 1999. Regardless of the reform efforts, the supply response failed to materialize. The economy remained vulnerable to oil price changes. The reform program succeeded in reversing the past decline in GDP growth, but the recovery remained slow and subject to large ﬂuctuations provoked by weather and oil price variability. Real GDP growth rate in the 1995–1996 period averaged 4 percent (mostly driven by the hydrocarbons sector), followed by a sharp slowdown in 1997 to 1 percent, reﬂecting the drought-induced decline in agricultural production and a continued decline in the state-owned industrial sector. Manufacturing value added decreased by 2 percent in 1995, 13 percent in 1996, and 7 percent in 1997, implying a cumulative loss during the period 1990–1997 of 40 percent in real terms. Despite a second oil price shock in 1998, the Algerian economy achieved a real GDP growth rate of 5.1 percent, mainly owing to agricultural growth (an exceptionally high rainfall record and an overall good harvest) along with a slight

342

Mohamed Abdelbasset Chemingui and Moataz Mostafa El-Said

recovery in manufacturing in response to the restructuring efforts of the previous 4 years. However, in 1999 real GDP growth dropped to 3.5 percent owing to a substantial slowdown of the non-hydrocarbon sectors (mainly construction and services sectors), which grew at only 2.7 percent in real terms owing to the budgetary impact of the sharp drop in oil prices. This was compounded by a severe slowdown in agriculture caused by a drought that mostly affected the cereal sector. Still the hydrocarbon sector grew by 6 percent as a result of improving oil prices to compensate for the slowdown in the non-hydrocarbon sectors. 3. Sources of economic growth in Algeria: a growth accounting exercise The incomes of countries increase as they accumulate more factors of production (labor, capital, and human capital) and as they use their resources more efﬁciently. There is a substantial literature that considers the relative contribution of the different factors of production and their productivity to output growth across countries (Page and Underwood, 1996). Growth accounting is an approach to decompose the growth of output into the growth of individual factors of production and TFP growth. This exercise is carried out to quantify the contribution of the different factors of production to Algeria’s total growth and to better understand what has prevented the country from achieving the rates of economic growth needed since its independence.3 As explained by Keller and Nabli (2002), in Algeria, accumulation and productivity have often gone in opposite directions, as seen during the period of massive public sector investments. Examining growth alone will mask these very different effects and the somewhat anemic growth that has characterized the country, especially considering that the observed effects of reform may be more a reﬂection of signiﬁcantly lower investments than of poor productivity performance. Using a constant returns to scale Cobb–Douglas functional form and assuming that the capital share, and hence technology, is the same across countries

3

An alternative econometric estimation of the production function was performed with the data used for the growth accounting carried out in this study as well as the database established by UNIDO. The estimation results using both data sources were not fruitful and arriving at conclusions could not be drawn using the estimated results. The production function used has GDP as output and labor and capital as input; alternative data were taken from Penn World Tables 6.1 (UNIDO). Data on capital were generated by UNIDO staffs from investment data (using PPP investment deﬂators, assuming 13.3 percent depreciation rate following Leamer). To compute TFP growth, the Data Envelopment Analysis (DEA) was used to obtain the change in technical efﬁciency and technical change, and the Malmquist index to obtain TFP growth. The advantage of this method is that it does not assume any functional form and no assumptions about perfect competition, proﬁt maximization, etc. are needed. Technically, DEA involves the use of linear programming methods to construct a non-parametric piece-wise frontier. Furthermore, a global study on determinants of productivity in developing countries was carried out last year (2004) by UNIDO, and also their estimation for Algeria was not promising and consequently Algeria was dropped from the project while it was included in the MENA countries to be covered by the project.

Algeria’s Macroeconomic Performances from 1962 to 2000

343

and equal to about one-third, we perform a growth accounting exercise for the four different subperiods mentioned earlier. That is: Y t ¼ AK at Lla t where Y is output; A an index of TFP, K and L the stocks of physical capital and labor, respectively; and 0oao1 a positive constant that can be interpreted as the share of capital income in total national income or as often referred to as the elasticity of capital.4 Our estimate for the sources of growth in Algeria begins by decomposing total growth into capital, labor, and TFP growth for the period 1962–1999. The year 1962 was chosen as the starting year since it is the year of Algeria’s independence. In addition, the analysis was carried out using two additional values for capital share (other than one-third) in the accounting exercise–0.35 and 0.4. Moreover, in the absence of a time series for employed labor force, we used a series for the total labor force instead. Again for lack of data, human capital was not incorporated in the growth accounting exercise. Regarding the data sources used in this exercise, the capital stock data were taken from Nehru and Dhareshwar (1994) and the total labor force data from the World Bank’s World Development Indicators (World Bank, 2002). The real GDP and capital stock series are at the constant prices of 1987. Furthermore, since the capital stock series estimated by Nehru and Dhareshwar (1994) stop at the year 1990, we have used investment ﬁgures from the World Bank World Development Indicators to complete the capital series using the simple inventory accounting approach for the years 1998 and 1999.5 For the period 1991–1997 we used the series estimated by Makdisi et al. (2007). (The data used in this paper are extracted from the database prepared by Makdisi et al. for the purpose of this project on explaining growth in MENA countries.) 3.1. Contribution of labor and capital to economic growth During the period 1962–1999, real GDP grew by 5 percent per annum, while stock of labor and capital grew at annual average rates of 2.4 percent and 4.9 percent, respectively (see Table 2). Assuming a capital share equal to 0.3, we ﬁnd that the annual growth rate of labor productivity is about 1.7 percent and that of capital is about 1.5 percent during the same period. Looking at the respective contributions of labor and capital to total GDP growth for the same period we ﬁnd the share of labor growth and capital growth to be 34.5 percent and 29.8 percent, respectively.

4 5

It is assumed that a factor is paid its marginal product. The estimation of the capital stock is given by the following equation:

K t ¼ ð1 dÞK t1 þ I t1 where K is the aggregate capital stock, d the annual rate of depreciation, and I the level of real investment in the previous period. Similar to the estimation done by Nehru and Dhareshwar (1994) and Makdisi et al. in this volume, the depreciation rate for installed capital is ﬁxed at 4 percent.

344

Mohamed Abdelbasset Chemingui and Moataz Mostafa El-Said

Table 2. Contribution to growth

1962–1999 Annual growth rate (in percent) Capital share ¼ 0.3 Capital share ¼ 0.35 Capital share ¼ 0.4 1962–85 Annual growth rate (in percent) Capital share ¼ 0.3 Capital share ¼ 0.35 Capital share ¼ 0.4 1986–1988 Annual growth rate (in percent) Capital share ¼ 0.3 Capital share ¼ 0.35 Capital share ¼ 0.4 1989–1994 Annual growth rate (in percent) Capital share ¼ 0.3 Capital share ¼ 0.35 Capital share ¼ 0.4 1995–1999 Annual growth rate (in percent) Capital share ¼ 0.3 Capital share ¼ 0.35 Capital share ¼ 0.4

GDP

Capital Stock

Labor Force

5.00

4.87 1.49 1.74 1.99

2.37 1.72 1.60 1.48

1.79 1.66 1.53

5.86 1.85 1.59 2.12

2.34 1.24 1.34 1.15

3.96 4.12 3.78

2.67 0.90 0.77 1.03

3.24 2.02 2.17 1.86

–3.66 –3.69 –3.63

4.54 1.4 1.2 1.6

1.88 1.09 1.18 1.01

–2.58 –2.46 –2.69

1.52 0.52 0.45 0.60

2.22 1.45 1.57 1.34

1.38 134 1.42

7.05

–0.74

–0.08

3.36

TFP

Source: Our estimations.

Underlying these ﬁgures are some signiﬁcant ﬂuctuations that the Algerian economy went through. For example, during the high but unstable economic growth period of 1962–1973, the volume of labor increased at an annual rate of 2.3 percent with a 17.6 percent contribution to total GDP growth while capital increased at an annual rate of 5.9 percent with a contribution to total GDP growth of 26.3 percent. The annual productivity of capital increased at an annual rate of 1.2 percent and that for labor was 1.9 percent. Thus, during the period 1962–1987, Algeria’s economic growth performance was considerably higher than that for the entire 1962–1999 period.6

6

Note that with ﬁxed amount of labor and an additional stock of capital of l percent, the Algerian economy gained >2 percent in economic growth. In other words, for an increase of n percent in the primary factors used for production, the Algerian economy grew at a rate (n+m) percent, where m is positive. This is a result of the increase in the growth of TFP.

Algeria’s Macroeconomic Performances from 1962 to 2000

345

During the ﬁrst period (1962–1973), Algeria embarked on a massive program of public investment in infrastructure, health, and education, which was translated into high economic growth and an increase in the contribution of capital as well as labor and also an improvement in TFP. However, relative to the volume of investment during this period, the contribution of capital to growth is low. One explanation is that, while investment was carried out in large infrastructure projects, state industries were protected, limiting competition and the efﬁciency of resource allocation, especially with subsidized energy, water, and other inputs. Toward the end of the period (in 1986), and with relatively low international oil prices, Algeria’s economy entered a phase of declining economic growth. With macroeconomic imbalances and growing external debt pressures, investments were reduced, and capital contribution to growth declined compared to that of the previous period (1962–1973). Whereas for labor, its contribution to growth increased from 1.2 percent to 2 percent for a capital share equal to 0.3, the volume of labor continued rising at an annual rate of 3.2 percent. Over the period from 1987 to 1994, a period that witnessed the implementation of the structural adjustment program, total labor employed increased at an annual rate of 1.9 percent and its contribution to economic growth varied between 1.0 and 1.2 (under the same capital share). 3.2. TFP and its contribution to economic growth This indicator sometimes referred to as the Solow residual, can be interpreted as a measure of the contribution of technological progress to the growth of output. However, in fact, it reﬂects all sources of growth other than the contribution of capital and labor, which might include changes in the utilization of capital as well as labor; changes in schooling quality as well as changes in the overall efﬁciency with which factors are allocated in the production process (Keller and Nabli, 2002). TFP evolution since 1962 is characterized by signiﬁcant variations from one period to another. Between 1962 and 1999, the contribution of TFP to economic growth is estimated at 35.7 percent (for a capital share equal to 0.3) or 30.6 percent (for a capital share equal to 0.4). However, during the period of sustained high growth of Algeria’s economy between 1962 and 1973 and before the ﬁrst spectacular fall in international oil prices, this indicator recorded a remarkable performance with a contribution to economic growth estimated to 56.1 percent for a capital share of 0.3. Again, sharp ﬂuctuations underlie the average TFP growth reported over the entire time period. Over the periods 1986–1988 and 1989–1994 and for a capital share of 0.3, the contribution of TFP to economic growth ranged from 3.7 to 2.6. The situation was reversed and a positive TFP contribution to economic growth was realized after 1995, when it reached 1.4 per annum or a contribution of 42 percent to total economic growth between 1995 and 1999. To test the sensitivity of the results with respect to the assumed value of capital share, TFP growth was measured for two alternative values (0.35 and

346

Table 3.

Mohamed Abdelbasset Chemingui and Moataz Mostafa El-Said

Determinants of long-term growth in algeria and other comparative countries

Parameter GDP annual growth rate (1960–1994) Contribution to growth (percent) Capital accumulation Labor force TFP (human capital included)

Morocco

Tunisia

East Asia

Algeriaa

4.6

5.3

6.1

4.9

2.6 0.9 1.1

2.3 0.9 2.1

3 0.6 2.5

1.5 1.7 1.8

Source: Page and Underwood (1996) and our calculations. a Data for Algeria are related to the period 1962–1999.

0.40). The results showed that changes in capital shares within this range did not change the value of TFP growth or its contribution to growth for the 1962–1995 period to any appreciable degree (see Table 3). Compared to the TFP growth over the entire period 1962–1999, the growth observed during 1995–1999 was higher. Indeed, during this period, the Algerian economy used only 1 percent of additional investment to achieve 2 percent of additional economic growth. Since the growth rate of labor was the same in both these periods, and knowing the role of capital, TFP growth explains the remaining part of total output growth. Over the long-term period, Algeria achieved economic growth mainly through labor force growth and secondarily though capital accumulation.7 Compared to the negative levels of TFP during the third period (1987–1994), Algeria achieved better results in terms of TFP growth since 1995. This was the time when the structural adjustment program in Algeria was adopted, reducing the share of public investment in total investment and especially in non-competitive activities. However, the efﬁciency gains were insufﬁcient to achieve the high rates of growth required to improve the needs of the rapidly growing population. Recent empirical estimations done by Senhadji (1999) and Makdisi et al. (2007) for the capital share ﬁnd that capital shares differ signiﬁcantly across countries, and mainly between oil and non-oil countries. The coefﬁcient for Algeria is higher than that for other MENA countries (with the exception of Tunisia). With similar shares of capital in real output, both Tunisia and Algeria realized quite different economic growth performances. In Tunisia, economic growth was more sustainable than in Algeria with an average of 5 percent during the last decade. To the contrary, the situation in Algeria is completely different, with an average economic growth estimated at 0.9 percent. How can this be explained? As suggested by Senhadji (2000), the contribution of TFP to GDP growth depends crucially on

7

In East Asia, TFP and capital accumulation are the main determinants of growth, while in Tunisia and Morocco, capital accumulation is the main determinant of growth (Table 3).

Algeria’s Macroeconomic Performances from 1962 to 2000

347

the share of physical capital in real output. The higher this share, the higher is the contribution of capital to economic growth. In both countries, there is a signiﬁcant contribution of capital to economic growth, but the difference between both of them is in the contribution of labor and TFP to economic growth. For Algeria, the contribution of labor growth to GDP growth is higher than that in Tunisia whereas Tunisia realized a much higher TFP contribution to economic growth as a result of technological changes, improvements in human capital, and many other structural reforms both at the macro as well as the micro levels8. 4. Structural weaknesses in the Algerian economy The potential for growth depends to a large extent on the resources that a country has, but for a country’s economy to grow, the ‘‘right’’ conditions have to be created. In this respect and in some of the recent contributions to the growth literature (Hall and Jones, 1999; Rodrik, 2003) a distinction has been drawn between two different layers of factors underlying growth: the ‘‘proximate’’ sources of factor accumulation and of productivity, and a few ‘‘deep’’ determinants. The latter can be divided into three groups, namely, geography, integration, and institutions (Rodrik et al., 2002).9 The present section will focus on the proximate sources. As far as the deep determinants go, all three groups (and also human capital) will play important roles; however, in a way that is different from that encountered in the standard analysis of aggregate growth. In Algeria and during the ﬁrst period (1962–1973) of economic development the hydrocarbon sector remained the backbone of the economy, accounting for 57 percent of the government revenues, 25 percent of GDP, and almost 100 percent of export earnings. This is not surprising as Algeria has the ﬁfth-largest reserves of natural gas in the world, is the second-largest natural gas exporter, and ranks 14th in terms of oil reserves. In this context, the impact of the higher oil prices observed during the past few years has been blurred by the internal political unrest that characterized the country since 1988.10 In what follows we consider some of the other structural barriers to achieving high economic growth in Algeria. These include weak institutions and administrative capacity, high external debt and debt servicing, high inﬂation rates, low productivity, an underdeveloped banking sector, and inadequate infrastructure.

8

For a detailed discussion of the labor factor and human capital contribution to economic growth in Algeria, the reader may refer to Keller and Nabli (2002). 9 Another view (represented by Lucas, 2002), would move one of the proximate sources, namely human capital to the level of deep – in fact, deepest – determinants. 10 To isolate the impact of oil on the growth performance of the Algerian economy given its political instability, one could choose a similar oil country or a counterfactual that did not experience the same instability and estimate what would have happened to the Algerian economy in the absence of this political instability. This exercise, however, is beyond the scope of this paper.

348

Mohamed Abdelbasset Chemingui and Moataz Mostafa El-Said

4.1. Weak institutions and administrative capacity The recent growth literature strongly emphasizes the role of good (i.e., highquality) institutions for speeding up growth, and of poor ones for retarding it.11 Unfavorable institutional factors – such as the degree of regulation and corruption in the government – are recognized as critical factors that can stand in the way of sustained economic growth. In this context, an important dimension for creating an investment climate conducive to growth is to have a governance system that honors contracts and property rights and works on reducing corruption. Surveys of existing enterprises and potential investors consistently rank various elements of institutional capability as among the major factors determining the ability of developing economies to attract new private investment (Dasgupta et al., 2002). Along the same lines is the work by Gwartney et al., (2004) which shows a positive correlation between investment and the quality institutions.12 Institutions affect performance primarily by fostering better policy choices and enhancing the sustainability of policies. Moreover, weak institutions tend to undercut the resilience of economies to exogenous shocks. Hence, poor institutions may lead to more volatile, crisis-prone economies compared to situations where institutions are better developed. Good governance is fostered through regulatory institutions that promote competition, support efﬁcient resource allocation, and protect property rights. Furthermore, empirical studies show a robust effect of institutions on volatility: the higher the quality of institutions, the lower the volatility of growth. The role of institutions has become a stronger determinant of growth as investors have become better informed about other countries and therefore, can compare institutional performance more easily than they could in the past. In Algeria, the quality of institutions is considered one of the main determinants of growth mainly because of the pervasive role of the state in the economy. There is a need for improving transparency and fair competition so as to attract external ﬁnancing.13 Typically, countries characterized by high returns, transparent investment procedures and regulations, low political risk, and accountable economic regimes and institutions are better able to attract external ﬁnancing. The ten-point index of Economic Freedom published by the Heritage

11

Rodrik et al. (2002). According to Noth (1990), institutions are deﬁned as the formal and informal constraints on political, economic, and social interactions. From this perspective, good institutions are viewed as establishing an incentive structure that reduces uncertainty and promotes efﬁciency–hence contributing to stronger economic performance. 13 Many indicators of institutions’ quality and governance are available for such assessments. The World Bank suggests evaluative and quantitative measures which are available for a large selection of countries including Algeria. But given that reliability of these measures depends on data accuracy and whether or not subjective perceptions are at play, rating measures seem useful for assessing both actual and potential performance and risk. 12

Algeria’s Macroeconomic Performances from 1962 to 2000

349

Table 4. Index of economic freedom for selected Arab countries (2002) Country Overall Trade Fiscal Government Monetary Foreign Regulation Property Black Score Burden Intervention Policy Investment Rights Market Algeria Egypt Jordan Kuwait Lebanon Morocco Syria Tunisia UAE Yemen

3.10 3.55 2.70 2.75 3.15 3.05 4.10 2.85 2.15 3.75

4.0 4.0 4.0 3.0 4.0 5.0 5.0 5.0 2.0 3.0

4.0 5.0 4.0 2.5 3.5 4.5 5.0 4.0 1.5 4.5

4.0 3.0 3.0 4.0 3.0 3.0 4.0 3.0 3.0 3.0

1.0 2.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 3.0

2.0 3.0 2.0 4.0 3.0 2.0 4.0 2.0 3.0 4.0

3.0 4.0 3.0 3.0 4.0 3.0 4.0 3.0 2.0 4.0

4.0 3.0 3.0 2.0 4.0 4.0 4.0 3.0 2.0 4.0

3.0 3.5 3.0 2.0 5.0 3.0 5.0 2.5 1.0 5.0

Source: The Heritage Foundation (2003).

Foundation (2003) is a good indicator of the quality of institutions.14 Table 4 indicates that Algeria is among the Arab countries with the highest overall scores (meaning less economic freedom) along with Egypt, Lebanon, Morocco, Yemen, and Syria. Algeria’s scores for government intervention, property rights, ﬁscal burden, and trade are among the highest in the Arab region. Only monetary policy seems to be relatively sound now. 4.2. High external debt and debt servicing According to Were (2001), the current stock of external debt as a ratio of GDP, is an indicator of the likely impact of external debt to economic growth. Debt servicing reduces foreign exchange earnings and has a direct effect on economic growth. According to Elbadawi et al. (1996), the debt burden indirectly affects growth through its impact on public sector expenditures. As economic conditions worsen, governments ﬁnd themselves with fewer resources and end up reducing capital spending and expenditure destined for social programs (which directly affects the poor). Algeria’s substantial debt dates back to the 1970s, when the government borrowed heavily to ﬁnance development projects and to meet rising consumer needs (Figure 2). By the mid-1980s, the debt servicing payment pressures started to build up, the debt-service ratio as a percent of GDP more than doubled between 1985 and 1988 (from 35 percent to 80 percent). By the end of 1990, the country’s

14

The coverage of the index is high across countries with a low level of speciﬁcity. The overall country index is composed of 10 elements, namely: trade, ﬁscal burden, government intervention, monetary policy, foreign investment, banking/ﬁnance, wages/prices, property rights, regulation, and black market. These elements encompass measures of governance. The higher the score the greater is the perceived risk and weaker the institutional quality.

Mohamed Abdelbasset Chemingui and Moataz Mostafa El-Said

350

80.00 70.00 60.00 50.00 40.00 30.00 20.00 10.00 0.00

1997

1995

1993

1991

1989

1987

1985

1983

1981

1979

1977

1975

1973

Series1

1971

Percent

Figure 2. External debt as percent of GDP

Year Source: World Bank (2002)

external debt slightly exceeded US$ 26 billion, of which almost US$2 billion was in short-term loans. To reduce the debt-servicing burden, the government subsequently concentrated on obtaining medium and long-term loans and discouraged importers from borrowing from their suppliers to minimize ﬁnancing costs. 4.3. High inflation rate In a paper by Andres and Hernando (1997), which examines the negative effects of inﬂation, the uncertainty associated with a high and volatile unanticipated inﬂation has been found to be one of the main determinants of the rate of return to capital and investment. In addition, inﬂation undermines the conﬁdence of domestic and foreign investors about the future course of monetary policy. Inﬂation also affects the accumulation of other determinants of growth such as human capital or investment in R&D, a channel of inﬂuence known as the accumulation or investment effects of inﬂation on growth. However, over and above these effects, inﬂation also worsens the long-run macroeconomic performance of market economies by reducing TFP. This channel is also known as the efficiency channel, which highlights the importance of the transmission mechanism from inﬂation to growth. A high level of inﬂation induces a frequent change in prices, which may be expensive for ﬁrms. It also generates larger forecast errors by distorting the informational content of prices, encouraging economic agents to spend more time and resources in gathering information and protecting themselves against the damages caused by price instability, hence distorting the efﬁcient allocation of resources (cf. Andre´s and Hernando, 1997). For Algeria, the annual inﬂation rate increased between 1971 and 1978 with a level varying from 2.5 percent to 17 percent (Figure 3). In 1989 and after the economic crisis, the inﬂation rate showed a marked increase from 6 percent in 1988 to>30 percent in 1992 and 1995. With the adoption of the adjustment

Algeria’s Macroeconomic Performances from 1962 to 2000

Inflation rate trends in Algeria

80.00 70.00 60.00 50.00 40.00 30.00 20.00 10.00 0.00

1997

1995

1993

1991

1989

1987

1985

1983

1981

1979

1977

1975

1973

Series1

1971

Percent

Figure 3.

351

Year Source: World Bank (2002)

program supported by the IMF and the World Bank, the inﬂation rate was sharply reduced to 4 percent by 1997. 4.4. Low labor productivity and the mismatch between the education and training system The growth of labor productivity is affected by changes in the capital/labor ratio and the growth rate of TFP.15 An increase in capital deepening tends to improve labor productivity since capital productivity is relatively slow to change. This provides a direct link between physical investment and labor productivity. If investment in physical capital is low, this will eventually produce a decline in the growth rate of the capital stock and in capital deepening, particularly if the labor input is growing rapidly. For example, an extended version of the neoclassical growth model (the so-called augmented Solow model of Mankiw et al., 1992) foresees a role for human capital as a separate factor whose ‘‘accumulation’’ contributes to growth of output per worker in the same way as physical capital. Furthermore, these authors presented evidence that variations in human capital are an important determinant of cross-country differences in income per capita. There is also evidence to suggest that the very high rates of growth in some East Asian countries are primarily attributed to human capital accumulation (Lucas, 2002). Otto (1997) argued that individuals can acquire human capital in two basic ways: formal schooling and on-the-job training, e.g., learning by doing. While both ways are likely to be important, it is the latter that appears to

15

We can show mathematically that the growth in labor productivity depends negatively on the growth in the labor force, positively on growth in Solow Residual (TFP), and positively on growth in capital stock. Thus, capital deepening and TFP increase labor productivity.

Mohamed Abdelbasset Chemingui and Moataz Mostafa El-Said

352

Output per worker trends

14,000 12,000 10,000 8,000 6,000 4,000

series 1

2,000

1990

1988

1986

1984

1982

1980

1978

1976

1974

1972

1970

1968

1966

1964

1962

0 1960

Output (Algerian Dinars)

Figure 4.

Year Source: World Bank (2002)

provide the most likely explanation for persistently high rates of growth. With learning by doing, knowledge accumulates, not by deliberate effort, but as a byproduct of economic activity. If learning by doing is an empirically important phenomenon, then the rate of productivity growth depends on the amount of new knowledge that is generated by conventional activities. In Algeria labor productivity has increased substantially since the 1960s (Figure 4). However, such increases were relatively low compared to other regions like East Asia. In addition, in East Asia but unlike Algeria, labor force growth was accompanied by enormous increases in real output. According to Keller and Nabli (2002) real GDP growth in East Asia averaged 7.6 percent a year during 1970–1980 (more than doubling its labor force). For Algeria, economic growth during the 1990s averaged only 1.9 percent (one-half less than the growth of its labor force), implying a substantial decline in productivity per potential laborer in the country.16 A number of reasons explain the weak labor market outcome in Algeria. These include the decline in government spending on education since the mid1980s as well as the inefﬁciencies characterizing the incentive structure associated with public sector employment and various aspects of rent-seeking behavior. Overall, the quality of education renders many graduates poorly equipped with modern skills. Enrollment explosions, Arabization disputes, poorly trained teachers, and the lack of high-quality educational materials have all seriously weakened the educational system. Although the amount of learning by doing at the national or sectoral level is difﬁcult to determine, Lucas (2002) argues that if learning by doing is itself

16

For more explanation on the low contribution of labor markets to economic growth in Algeria, see Chemingui and Ayadi (2002).

Algeria’s Macroeconomic Performances from 1962 to 2000

353

subject to diminishing returns to scale, a sustained rise in the growth rate of productivity will require the continued introduction of new goods, not just continued learning on a given set of goods. One measure by which a small economy can expand the range of goods it can (potentially) produce, is by selling on the world market. Thus, having a relatively open economy seems to be an important precondition for learning-based growth. Professional training is always considered as an indirect means of learning by doing. In Algeria, the professional training system, which started less than about 20 years ago, was largely designed to meet the needs of an economy based on PEs providing agricultural products and assembled manufactured goods with low local value-added and entirely oriented to a highly protected local market. At the moment, the education and training system are not able to build a workforce with high skills that are able to meet the challenges of the Algerian economy which needs to be diversiﬁed. 4.5. Underdeveloped banking sector One of the main roles of the ﬁnancial system is to ensure that resources get transferred to the most efﬁcient use. Thus, a well-functioning ﬁnancial system is an important precondition for capital accumulation and hence economic growth as well as for a burgeoning private sector. By contrast, the more frictions there are, the less well can the ﬁnancial system operate as an intermediary in resource (re-) allocation. Thus, a sound ﬁnancial system is required to allow ﬁrms to enter the market and operate effectively as well as to help restructure failing ﬁrms. An important task of the ﬁnancial sector is to support entry of promising ﬁrms and reallocate resources away from failing or underperforming ﬁrms to more promising ones. A well-functioning ﬁnancial sector operating at arm’s length from political and corporate interests is crucial for competition and productivity growth. A large body of literature on ﬁnancial intermediation shows the crucial role played by the ﬁnancial sector to improve the economy’s savings, investment, productivity, and growth. In this regard, developing the ﬁnancial sector may improve the level of savings, which will likely raise investment levels and growth. In addition, a performing ﬁnancial sector contributes to higher growth through better allocation of resources. Well-functioning ﬁnancial markets will allocate resources to users with the highest return and, through the monitoring of creditor’s performance, ensure continued efﬁcient use of productive assets (Jbili et al., 1997). Until the early 1980s, Algeria pursued an inward-looking economic development strategy emphasizing a key role for the state in economic activity to accelerate economic development and ensure government control over ‘‘strategic’’ sectors. These objectives were to be achieved through direct government investment in key productive sectors, the provision of generous incentives (such as interest rate subsidies) to private investment in priority sectors, and through a complex system of trade and exchange controls designed to protect local industries. The strategy resulted in a buildup of a large public sector.

Mohamed Abdelbasset Chemingui and Moataz Mostafa El-Said

354

Accordingly, ﬁnancial resources were allocated to achieve the planning objectives. Thus, available ﬁnancial resources were ﬁrst given to government and PEs and secondly to private sector ﬁrms operating in key sectors, which are mostly producing for local market and beneﬁting from high border protection. The remaining ﬁnancial resources were allocated to private sector ﬁrms with high interest rates. During the period from 1996 to 2000, the public sector absorbed 80 percent of domestic credit (IMF, 2001). In addition, inefﬁciencies of direct government controls on credit allocation contributed to the buildup of non-performing loans in most banks in Algeria, mainly the development banks. This weak ﬁnancial system in pre-1980 Algeria did not facilitate private sector investment in activities in which the country had a comparative advantage, affecting the level of economic growth as well as productivity. Faced with the increasing economic difﬁculties of the mid-1980s and inﬂuenced by the worldwide trend toward ﬁnancial liberalization and deregulation, Algeria embarked on a wide-ranging structural reform program that included liberalization of its ﬁnancial system. The objectives behind the ﬁnancial sector reform were to reduce direct government intervention and strengthen the role of market forces in the allocation of ﬁnancial resources, improve the capacity of ﬁnancial institutions to mobilize domestic savings, and promote competition among banks. Thus, the liberalization of the banking sector (abolishing credit control in 1991, liberalizing interest rates on overdrafts in 1992, and on accounts in credit in 1996) and an expansionary monetary policy (reducing intervention rates of the central bank on the money market and reducing reserve requirements) have led to a substantial reduction in interest rates. Figure 5 shows the development of private sector credit as a share of total domestic credit. The sharp decline in the domestic credit allocated to the private sector corresponds closely to the structural adjustment program.

19 70 19 73 19 76 19 79 19 82 19 85 19 88 19 91 19 94 19 97 20 00

67

Credits to Private Sector/Total Domestic Credit

19

64

100.0% 90.0% 80.0% 70.0% 60.0% 50.0% 40.0% 30.0% 20.0% 10.0% 0.0% 19

Percentage of domestic credit allocated to the private sector

Figure 5. Credits to private sector/total domestic credit

Year

Algeria’s Macroeconomic Performances from 1962 to 2000

355

4.6. Inadequate infrastructure Equally important for economic growth is the quality of infrastructure that allows private entrepreneurs and their employees to operate effectively. De la Fuente (2002) surveyed available evidence and concluded that there are sufﬁcient indications that public infrastructure investment contributes signiﬁcantly to productivity growth and thereby to economic growth, at least in countries where a saturation point has not been reached. The returns to such investment are probably quite high when infrastructure is scarce and basic networks have not been completed, but fall sharply thereafter. Hence, appropriate infrastructure provision is probably a basic ingredient for growth, even if it does not hold the key to rapid productivity growth in advanced countries, where transportation and communications needs are already adequately served. For Algeria, there is an urgent need to provide producers, whether they belong to the primary, secondary, or tertiary sectors, with the infrastructure services they need to become or remain competitive at the international level. Thus, and while massive public investment in infrastructure has been happening during the last four decades, the country still suffers from a lack of appropriate infrastructure services. This explains the low level of economic diversiﬁcation and poor economic and productivity performance observed in the country mainly during the last two decades. Infrastructure in Algeria is still characterized by public ownership, monopolies, and stiﬂing regulations. The situation is particularly difﬁcult in Algeria because it needs not only to maintain, indeed often to replace, the existing infrastructure, but also to expand it so as to cope with the requirements of more rapid economic growth. During the last decades, inadequate infrastructure in Algeria has paralyzed economic activities owing to trafﬁc bottlenecks, port congestion, and a stalled manufacturing sector as a result of interruptions in energy supply breakdowns on overloaded electricity networks. These are linked to difﬁculties in complying with international quality standards, longer transit times, and lower productivity. Furthermore, because of ﬁnancial constraints and limited resources, the State underinvested in the transport infrastructure. Inadequate resources for its rehabilitation and maintenance were lacking. The resulting deterioration of the infrastructure generates additional costs, and certainly reduces competitiveness. All these factors tend to make Algeria’s industries and services less competitive both in local and foreign markets. 5. Conclusion and policy options The central focus of this paper was to explain the sources of Algeria’s economic growth during the past four decades, i.e., since its independence. In doing so, the paper revisits Algeria’s economic growth performance to examine the economy’s growth path and explain its economic performance over the period 1962–2000. To highlight the sources of economic growth, the paper adopts Solow’s

356

Mohamed Abdelbasset Chemingui and Moataz Mostafa El-Said

accounting exercise to disaggregate overall growth in output into growth in capital, labor, and TFP. The paper ﬁnds that the poor economic growth performance that Algeria experienced during the last four decades is mainly due to poor internal management of the economy and the absence of economic policies in response to external economic shocks. Algeria’s slow economic growth performance can be attributed to (i) the country’s management of its natural resources; (ii) the country’s macroeconomic policies, speciﬁcally price deregulation and economic reforms aimed at increasing investment; (iii) structural impediments limiting private sector contributions to the country’s economic activities; and (iv) an overall economic environment characterized by a dominant public sector, a weak ﬁnancial sector, a non-transparent privatization process, and number of other factors that limited the progress of economic reforms. In addition to the reasons discussed in the paper, other factors played a role as well. Mainly, the recent political context (civil war) and other structural factors continue to block all reform initiatives. What Algeria needs is a focus on supplyside policies to boost the economy’s growth capacity. These include measures to improve the competitiveness of existing activities, efforts to attract investment in non-energy sectors so as to diversify the economy. Other complementary supplyside policies may include promoting education and training, and increasing spending in the health sector. Additional economic policy efforts may include accelerating the trade liberalization process and encouraging domestic and foreign investment in export-oriented industries. Moreover, Algeria can aggressively move toward improving its investment climate. Actions under this option can be grouped under three broad headings: (i) deepening of ﬁnancial markets; (ii) improvements in the judicial and administrative systems governing private and public activities; and (iii) further expansion of privatization. Finally, Algeria can adopt policies intended to accelerate the rate of productivity by adopting an ‘‘export push’’ strategy consisting of three essential elements: (i) accelerating trade liberalization to reduce its anti-export bias; (ii) institutional reforms for trade liberalization; and (iii) targeted investments in trade-related infrastructure. Acknowledgments The authors would like to thank Imed Limam, Terry Roe, Magdy Mashaly, Joan Haddock, Jeff Nugent, Hashem Pesaran and two anonymous referees for their comments and useful suggestions. The usual disclaimer applies. References Andre´s, J. and Hernando, I. (1997), ‘‘Does inﬂation harm economic growth? Evidence for The OECD’’, Paper presented at the NBER conference on The Costs and Beneﬁts of Achieving Price Stability, February 20–21, 1997, Federal Reserve Bank of New York.

Algeria’s Macroeconomic Performances from 1962 to 2000

357

Dasgupta, D., J. Keller and T.G. Srinivasan (2002), ‘‘Reform and elusive growth in the Middle East: what has happened in the 1990s’’, Middle East and North Africa Working Paper Series No. 25, The World Bank, Washington, DC. De la Fuente, A. (2002), ‘‘Does cohesion work? Some general considerations and evidence from Spain’’, Paper presented at the Conference on Income Convergence and European Regional Policy: Impact, Priorities and Prospects for the EU Candidate Countries, October 10–12, jointly organized by the Bertelsmann Foundation and the World Bank, Barcelona. Elbadawi, A.I., J.B. Ndulu and N. Ndung’s (1996), ‘‘Debt overhang and economic growth in Sub-Saharan Africa’’, Paper presented at the IMF/World Bank conference on External Financing for Low Income Countries, December, Washington. Gwartney, J., Holcombe, R.G. and Lawson, R.A. (2004), ‘‘Institutions and the impact of investment on growth’’. Memo, Florida State University. 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, Vol. 114, pp. 83–116. Heritage Foundation (2003), 2002 Index of Economic Freedom, New York: The Heritage Foundation/Wall Street Journal. IMF (2000), ‘‘Algeria: recent economic developments’’, IMF Staff Country Report No. 00/105, Washington. IMF (2001), ‘‘Algeria: Statistical Appendix’’, IMF Country Report No. 01/163, Washington. Jbili, A., K. Enders and V. Treichel (1997), ‘‘Financial sector reform in Algeria, Morocco and Tunisia: a preliminary assessment’’, IMF Working Paper WP/ 97/81, IMF, Washington, DC. Keller, J. and M.K. Nabli (2002), ‘‘The macroeconomics of labor market outcomes in MENA over the 1990s: how growth has failed to keep pace with a burgeoning labor market’’, Paper presented at the MDF4, October 2002, Amman. Lucas, R. (2002), ‘‘Making a Miracle’’, Econometrica, Vol. 61(2), pp. 251–272. Makdisi, S., Z. Fattah, and I. Limam (2007), ‘‘Determinants of growth in the MENA countries’’, Regional Background Paper for the GRP on Explaining Growth in MENA Region. Mankiw, G., D. Romer and D. Weil (1992), ‘‘A contribution to the empirics of economic growth’’, Quarterly Journal of Economics, Vol. 107, pp 407–437. Nehru, V. and A. Dhareshwar (1994), ‘‘New estimates of total factor productivity growth for developing and industrial countries’’, Policy Research Working Paper No. 1313, the World Bank. Noth, D.C. (1990), Institutions, Institutional Change and Economic Performance, Cambridge: Cambridge University Press. Otto, G. (1997), ‘‘Productivity growth and economic policy in Australia’’, Research Paper 19, 1996–1997, Parliamentary Library of Australia.

358

Mohamed Abdelbasset Chemingui and Moataz Mostafa El-Said

Page, J. and J. Underwood (1996), ‘‘Growth, the Maghreb and free trade with the European Union’’, The Egyptian Center for Economic Studies, Working Paper Series N 07. Rodrik, D. (editor) (2003), Search of Prosperity: Analytic Narratives on Economic Growth, Princeton: Princeton University Press. Rodrik, D., Subramanian, A. and Trebbi, F. (2002), ‘‘Institutions rule: the primacy of institutions over integration and geography in economic development’’, IMF Working Paper WP/02/189, International Monetary Fund, Washington, DC. Senhadji, A. (1999), ‘‘Sources of economic growth: an extensive growth accounting exercice’’, IMF Working Paper 99/77, the IMF. Senhadji, A. (2000), ‘‘Sources of Economic Growth: an Extensive Accounting Exercise’’, IMF Staff Papers Vol. 47(1). Were, M. (2001), ‘‘The impact of external debt on economic growth and private investment in Kenya: an empirical assessment’’, Paper presented at the wider conference on Debt Relief, August, Helsinki. World Bank (1994), (The Democratic and Popular Republic of Algeria Country Economic Memorandum: The Transition to a Market Economy), Vol. I (In two volumes), Washington: Country Operations Division. World Bank (2002), World Development Indicators, Washington, DC.

CHAPTER 12

Explaining Growth in an Oil-Dependent Economy: The Case of the United Arab Emirates Adam B. Elhiraika and Annas H. Hamed Abstract This paper attempts to explain the determinants of economic growth in the United Arab Emirates (UAE) using growth accounting and regression models. The findings indicate the predominance of labour in aggregate growth and a negative contribution by total factor productivity growth (TFPG). In addition to institutional changes and liberal economic policies that encourage re-exports rather than genuine non-oil exports, the evidence on TFPG has been attributed to wide fluctuations in oil price and income that constitute the main omitted variables in TFPG, which is measured as a residual. Econometric results confirm that, unlike the case of countries with diverse resources, natural resource abundance, represented by oil exports, spurs growth in an oil-dependent country such as the UAE. Similarly, changes in oil prices have strong positive effects on economic growth, while all other variables have no important effects, indicating the relatively weak contribution of the non-oil sector. The study discusses policy options for productivity enhancement and for strengthening the trend of increasing genuine exports in recent years as means of achieving true economic diversification and sustainable development.

Keywords: growth, oil-dependent, United Arab Emirates, oil income, non-oil income, development policy JEL classifications: O11, E22

Corresponding author. CONTRIBUTIONS TO ECONOMIC ANALYSIS VOLUME 278 ISSN: 0573-8555 DOI:10.1016/S0573-8555(06)78012-9

r 2007 ELSEVIER B.V. ALL RIGHTS RESERVED

360

Adam B. Elhiraika and Annas H. Hamed

1. Introduction This paper aims at explaining the determinants of growth in the UAE with the objective of highlighting strategies and policies for promoting sustainable development in the country. The search for the determinants of growth has continued to be at the core of economics because of its important policy implications. The allocation and efﬁcient use of resources, income distribution and prospects for sustainable growth are among the main issues of interest in the growth literature. Both ﬁscal and monetary policy design can beneﬁt from growth analysis. The UAE is a federation of seven small Gulf emirates that was formed immediately after the withdrawal of Britain from the region in 1971. Before the discovery of oil in Abu Dhabi in 1958 and in Dubai in 1966, economic activity in the seven emirates was dominated by pearling and ﬁshing, which provided employment and income to about 70% of the population (Ghanem, 1992). The rest, 30%, of the population was involved in little agriculture, rural handicraft and herding. There was hardly any kind of industry except for construction of wooden boats and simple handicrafts. The UAE being virtually a desert, its people have been involved in trade since ancient times. Oil wealth has provided the necessary initial conditions for drastic economic and social transformation as well as a rapid increase in population, from 180,000 in 1968 to about 3.7 million in 2002. In addition to oil, four key factors appear to have contributed to rapid economic growth and increased diversiﬁcation in the UAE: huge government investment in physical and social infrastructure, a stable macroeconomic environment, availability of capital and absence of restrictions on capital movement together with a high degree of openness and availability of cheap labour from neighbouring Arab countries and the Indian subcontinent. However, like the other oil-rich small countries in the Middle East and North Africa (MENA) region, the UAE remains heavily dependent on oil income, despite government efforts to encourage private investment in the non-oil sectors. Nevertheless, the UAE economy has been recently classiﬁed as the most relatively well-diversiﬁed economy in the gulf region (Askari and Jaber, 1999) with an average real GDP growth rate of about 4.5% for the period 1975–2002 compared to a negative average growth for MENA oil-exporting countries during the same period. It is no surprise that only oil revenue and changes in oil prices appear to have signiﬁcant effects on the growth of UAE. Unlike the case in countries with a diversiﬁed resource base, natural resource abundance induces growth in an oildependent country such as the UAE. The study underscores the fact that much of the economic diversiﬁcation and growth in non-oil income relates to growth in re-exports that have little inﬂuence on genuine production of goods and technological progress. Therefore, economic policies and strategies are needed to

Explaining Growth in an Oil-Dependent Economy

361

foster real production and enhance productivity as prerequisites for achieving diversiﬁcation and sustainable growth. In the next section, we discuss the performance of the economy during 1975–2002, the period for which data are available. In Section 3, we use a growth accounting framework to estimate the contribution of factor inputs to the overall GDP growth and analyse the ﬁndings in a regional context. Also a model of the determinants of UAE growth is speciﬁed and estimated and the results discussed in this section. Section 4 gives a detailed analysis of the role of markets, especially the labour and capital markets, in the UAE growth process. Finally, Section 5 concludes with some policy remarks. 2. Performance of the UAE economy To analyse UAE economic performance during 1975–2002, two distinct subperiods may be identiﬁed. The period from the mid-1970s to the early 1980s was characterized by high growth performance. This was the time when the government directed surpluses from high oil prices into the physical and social infrastructure. The period from around the mid-1980s witnessed signiﬁcant reduction in economic growth owing to a sharp drop in oil prices. Subsequent government austerity measures were directed largely towards capital expenditures since most of the basic infrastructure projects had been completed by then, while most current expenditure categories have become long-term commitments. Overall, the performance of the economy ﬂuctuated widely over the period considered though aggregate real GDP rose from US$14.72 billion in 1975 to US$50 billion in 2002 (Table 1). Owing to rapid population growth and immigration, real per capita income declined from US$30,161 to US$16,400 during the same period. Economic growth has since the early 1970s relied on both oil and non-oil income, and by 1982 non-oil GDP became more important. The share of crude oil revenue declined from about 70% of GDP in the 1970s too40% in the 1990s (Table 2). Despite government efforts to diversify the economy in the face of highly volatile international oil markets and generally declining oil prices and revenues, this trend is mainly a reﬂection of expansion in the export sector, especially re-exports that amounted to about 30% of the total exports in recent years. The non-oil GDP originates largely from the services sector with a share of 50% in the last decade. The share of the non-oil goods sector in aggregate GDP has increased, though very slowly, from 21.3% in 1985 to 27% in 2002, reﬂecting the relatively poor commodity base of the economy. Accordingly, the momentum of economic diversiﬁcation away from the crude oil sector has been slow despite notable increases in the allocation of capital into the non-oil sector. The bulk of domestic investment in the 1990s and after went to the construction and real estate sector, followed by the transport and communication sector and manufacturing, while the share of oil investments in total domestic investment was almost halved

Adam B. Elhiraika and Annas H. Hamed

362

Table 1.

Some macroeconomic and financial indicators (base ¼ 1990)

Indicator Real GDP growth (%) Real non-oil GDP (% GDP) Real non-oil exports (% of GDP) Real non-oil exports (% of total exports) Real average GDP per capita (US$) Gross domestic investment (% GDP) Gross domestic saving (% GDP) Inﬂationa Real lending rate of interestb Bank credit to the private sector (% of total) Secondary school enrolment ratio

1975–1985

1986–1990

1991

1992–1995

1996–2000

2001

2002

6.85 43.5

3.94 63.8

–3.11 54.55

3.98 65.61

3.62 72.4

3.6 70.0

1.9 72.0

11.6

20.4

25.25

37.67

26.0

25.6

23.9

19.8

36.1

37.34

53.28

61.9

63.5

65.8

26.6

15.8

18.41

17.40

17.5

17.2

16.4

31.1

23.6

20.7

26.8

27.6

28.8

28.9

39.0

20.5

12.51

21.2

22.7

24.1

21.4

2.5 7.8

2.5 7.21

0 8.44

1.94 5.51

2.36 5.31

2.1 4.82

1.3 3.4

77.1

81.0

79.8

76.85

83.0

86.9

84.6

48.0

60.4

69

77

76.0

72.0

74.0

Source: UAE Economic Indicators, 1980, 1989, 1999, Ministry of Planning and the Central Bank’s Annual Bulletin (various editions). a Calculated as percentage change in GDP deﬂator. b Calculated as the difference between nominal fending rate and the rate of inﬂation.

since the turn of the 1990s (Table 3). The meagre share of manufacturing, the potentially most dynamic sector, in aggregate investment in UAE partly explains why its contribution to GDP is still quite low, at just 14.1% in 2002. Real GDP growth rates ﬂuctuated widely from one year to another mainly owing to changes in oil prices and revenues (Figure 1). At the same time, with the exception of the First Gulf War period (1984–1988), real non-oil GDP has always maintained positive growth rates. Since the early 1990s, the growth in non-oil income appears to have signiﬁcantly stabilized aggregate GDP growth. Indeed, growth in the good sectors (agriculture, manufacturing, water and electricity and construction) represented the most stable source of GDP growth in the period considered, followed by growth in the trade and services sectors. This suggests that while the oil sector underpinned the modem economic evolution in the UAE and still dominates overall growth, achieving sustainable growth in the long-run would depend on expansion in other sectors. While the UAE maintained real GDP growth rate of about 4.5% per annum during 1975–2002, the average growth rate was -0.6% for MENA oil-exporting

Explaining Growth in an Oil-Dependent Economy

Table 2.

363

GDP by sector, 1975–2002 (%)

Sector

1975

1980

1985

1990

1995

2000

2001

2002

Agriculture, livestock and ﬁshing 0.8 0.7 1.4 1.6 2.9 3.6 3.5 3.7 Manufacturing 0.9 3.8 9.1 7.7 10.4 13.5 13.9 14.1 Water and electricity 0.5 1.2 2.1 2.0 2.1 2.0 1.93 2.0 Construction 10.9 8.8 8.7 7.7 8.7 6.6 6.9 6.9 Total non-oil commodity sector 13.1 14.5 21.3 19.1 24.0 25.7 26.2 26.7 Oil and mining 66.8 63.5 44.1 46.3 30.9 34.0 30.0 28.0 Trade 8.2 8.2 8.5 9.0 11.7 8.7 9.0 9.4 Transport, storage and communication 3.2 3.3 4.1 5.0 6.7 6.7 7.7 7.8 Finance and insurance 1.6 1.9 5.1 4.1 5.6 5.8 6.6 6.9 Real estate and business services 4.0 3.6 5.1 5.5 10.0 7.5 7.8 7.8 Other services 1.0 0.7 1.6 2.0 1.6 2.1 3.3 3.6 Government services 3.4 5.4 10.8 10.4 10.6 10.5 10.6 11.1 Domestic (household) services 0.1 0.2 0.4 0.4 0.8 0.6 0.8 0.8 Less imputed bank services –1.4 –1.3 –1.0 –1.6 –2.0 –1.6 –2.0 –2.1 Total services sector 20.1 22.0 34.6 34.7 45.1. 40.3 43.8 45.3 Gross domestic product 100 100 100 100 100 100 100 100 Total non-oil sector 33.2 36.5 55.9 53.7 69.1 66.0 70.0 72.0 Source: Development indicators in the UAE (1999, 2003), Ministry of Planning, Abu Dhabi, UAE. Italicized ﬁgures represent main economic sectors (total non-oil sector; oil and mining sector; and total services sector) which add up to 100%.

Table 3.

Gross domestic investment by sector (%)

Economic Sector Petroleum Transport, storage and communication Construction and real estate Manufacturing (non-oil) Electricity and water Government services Others

1975–1989

1990–1999a

2000

2001

2002

35.78 11.36 11.92 15.71 6.21 9.9 9.12

17.3 16.2 20.0 15.0 10.8 9.8 11.2

13.0 17.0 24.6 16.7 8.0 7.0 13.7

12.8 17.9 21.4 16.0 8.9 8.1 14.9

13.1 16.7 21.2 16.8 8.0 8.5 15.7

Source: UAE Economic Indicators, 1980, 1989, 1999, 2003, Ministry of Planning, UAE. a Excluding 1991.

countries and 1.2% for all MENA countries. Many observers (e.g. Elhiraika and Hamed, 2001) argue that relatively better economic performance in the UAE was spurred by increased domestic investment associated with rapid expansion in foreign trade; a stable macroeconomic environment; liberal ﬁnancial, exchange rate and trade policies; increased investment in human capital; easy availability of skilled labour from neighbouring countries; and access to Middle Eastern and African product markets.

Adam B. Elhiraika and Annas H. Hamed

364

60

GDP per capita Oil Price Index

40 20 0 -20 1975 1978

1981

1984

1987

1990 1993

1996

1999

2002

-40 -60

Figure 1. Growth in real GDP per capita and the oil price index (%) 2.1. Domestic investment Gross domestic investment declined from 34.1% of GDP in the 1970s to 25.6% in the 1980s before rising to 29% by 2002. The UAE average investment rate for the whole period is well above the average rate of 26% for the oil-rich MENA countries and 25% for the entire MENA region. Although public investment continues to dominate, the share of private investment has risen remarkably, from 6.6% of GDP in 1975–1985 to 12% in the 1990s. Meanwhile, the share of petroleum investment in aggregate investment declined from 36% in 1975–1989 to 13% in the 2000s. While public investment is concentrated on infrastructure and services sectors, most of the private investment is in the services and real estate sectors. In explaining the private investment behaviour in the UAE, Elhiraika and Hamed (2001, p. 13) found that ‘‘in the long-run, GDP growth1 has the largest stimulating inﬂuence, followed by bank credit to the private sector and a human capital development variable. The real lending rate and government investments are found to have strong but adverse effects on private investment. In the short-run, GDP growth, bank credit and investment in human capital still have positive but weak effects on private investment behaviour, whereas the lending rate and government investment variables retain their signiﬁcant negative effects’’. The ﬁnding of a crowding-out effect of public investment on private investment is consistent with the argument of Hakura (2004) that large government size in Gulf Cooperation Council (GCC) countries adversely affects the incentive to accumulate capital per worker and appears to have stiﬂed private sector growth and impeded the diversiﬁcation of production. The public and private sectors compete with each other in terms of areas of investment rather than sources of ﬁnance. Expansion of industrial zones that provide a variety of

1

As Elhiraika and Hamed (2001) note, there can be two-way causation between growth and investment. Therefore, this result was obtained after accounting for endogeneity using the two-stage least squares method.

Explaining Growth in an Oil-Dependent Economy

365

facilities and services at attractive prices is believed to have encouraged both domestic and foreign private investment. Outside industrial zones, foreign ﬁrms are allowed to operate with local UAE participation of at least 51% and company law may vary across emirates. In the zones, both domestic and foreign ﬁrms operate under the same company law that has been harmonized across the emirates. In 2004, Jebel Ali, the largest free zone in the UAE, hosted 850 manufacturing, trade and other ﬁrms, 200 of which were owned by the UAE nationals.2 2.2. Domestic savings The growth of domestic investment is aided by the easy availability of funds with internationally competitive and positive real interest rates. In the 1980s, Gross Domestic Savings (GDS) fell dramatically from record high levels in the 1970s before rising again in recent years. Recent increases in GDS rates were largely due to increases in private savings, whereas the government sustained current ﬁscal deﬁcits since the start of the 1990s. The average UAE aggregate saving rate of 33% during 1975–2002 is greater than that of 20% for all MENA countries, but lower than the oil-rich MENA countries’ saving rate of 39%. This observation, together with the fact that Gross Domestic Investment rates in the UAE were higher than in the other MENA oil-exporting countries, suggests that the UAE has managed either to make a more productive use of its ﬁnancial resources or to attract more foreign capital compared to these countries. 2.3. Export growth As mentioned previously, good economic performance in the UAE, in comparison to other countries in the region, coincided with sharp rises in non-oil exports. However, increases in non-oil exports are mainly due to re-exports that made up 30% of the total exports between 1982 and 2002. Thus, the UAE became the third largest re-export centre in the world, after Singapore and Hong Kong. Whereas the trend of total non-oil exports suggests that the UAE is experiencing an export-led growth phenomenon, the dominance of re-exports points to an important caveat in this trend. With huge investments directed to the re-export sector, the domestic production base of the country remains weak, constraining opportunities for technology transfer and productivity enhancement. Meanwhile, the re-export sector is expected to face ﬁerce competition from the free-trade zones that are rapidly developing in the region, especially in Oman that has a relative advantage in terms of having seaports that are closer to major sea routes. Therefore, sustainable growth in the non-oil export sector

2

For further information see Jebel Ali Free Zone’s website: www.freeznoneuate.com/Jebel_fze.htm

366

Adam B. Elhiraika and Annas H. Hamed

would require increased domestic production of export goods rather than reexports. 2.4. Financial development Saving and investment performance have been associated with a notable increase in the level of ﬁnancial development in terms of a high ratio of broad money to GDP, increasing share of bank credit to the private sector in total credit, positive real interest rates and an unusually high ratio of time, saving and investment deposits in total deposits. This signiﬁes the easy availability of credit to ﬁnance investment in the country. The development and role of ﬁnancial markets and institutions in the UAE is discussed in detail in Section 4. 2.5. Human capital development Increased investment in human capital had led to notable increases in the primary and secondary school enrolment ratios, fromo40% in the 1970s to>80% in the 2000s. Besides the increased education of the local labour force, educated foreign labour is easily accessible given the relatively high wages paid in the UAE compared to other labour surplus countries in the Middle East and Asia. Immigrant labour accounts for about 70% of the labour force in the country and immigrants are generally better educated than the local population. The IMF (2003) argues that the abundant supply of expatriate workers at internationally competitive wages has been an important source of growth for the nonoil sector of the UAE. Nevertheless, since expatriate workers remit much of their earnings to their countries of origin, their contribution to growth would, other things being equal, be lower than that of local workers. Though important, empirical investigation of the question of how much growth can be created by foreign3 labour is beyond the scope of this paper. 2.6. Macroeconomic stability In spite of high ﬂuctuations in oil price and revenue that lead to similar, though smaller ﬂuctuations in real GDP growth, the UAE economy remained remarkably stable in terms of inﬂation rates and the exchange rate. Since 1981, the UAE dirham has been fully pegged to the US$ at the rate of 3.67 and the inﬂation rate varied around an average of 2.5% over the period considered. It is believed that, because oil is priced in US dollars and because the UAE has huge investments in the US, the beneﬁts from the peg in terms of economic stability and reduced macroeconomic uncertainty is greater than the cost arising from inability to use exchange rate policy to promote domestic investment and

3

It is worth noting that high export-oriented and high-growth countries such as Singapore and Hong Kong also rely signiﬁcantly on foreign labour.

Explaining Growth in an Oil-Dependent Economy

367

international competitiveness. There are, however, no hard statistics to support or negate this argument. Since the turn of the 1990s, the consolidated budget (including the federal government and emirates’ governments) has experienced sustained deﬁcits. But, as Hamed and Elhiraika (2001) argue, the UAE government relies mainly on monetary policy tools, particularly the link between the dirham and the US dollar, to maintain macroeconomic stability, and that the governments of the dominant emirates ﬁnance their budget deﬁcits by drawing down their own abundant overseas assets. This helps them to avoid domestic borrowing that can crowd out private sector activities and create macroeconomic instability. The structure and ﬁnancing of the budget deﬁcit raises two important questions in the context of long-run growth. The ﬁrst question relates to ﬁscal sustainability in the absence of taxes and the second question relates to the potential impact of cuts in government expenditure on growth. Unfortunately, we have insufﬁcient data, especially on accumulated domestic and foreign investment4 by the various emirates, to answer these questions. To the extent that deﬁcit ﬁnancing does not exceed the income generated by public assets, the government would be able to maintain ﬁscal stability. Otherwise, there will be a need for signiﬁcant adjustment of government expenditure and increases in government revenue. While such moves may adversely affect economic growth in the shortto-medium term, they are paramount for achieving ﬁscal sustainability, which is in turn a prerequisite for attaining sustainable growth in the long-run. 3. UAE growth in regional perspectives 3.1. Growth accounting framework In this section, we analyze the determinants of economic growth using the growth accounting framework by decomposing aggregate GDP growth into labour, capital and technical progress. The decomposition is based on the standard neoclassical production function that assumes competitive factor markets, constant returns to scale and neutral technical progress (see Solow, 1956). Accordingly, total output growth can be explained by input growth and improvement of productive efﬁciency. That is Y_ K_ L_ A_ ¼ þ a þ ð1 aÞ Y A K L

ð1Þ

where the dotted variables denote the growth rates of output (Y), capital (K) and labour (L). The rate of change of the constant (A) is a measure of total factor productivity growth (TFPG) and is obtained as a residual. TFPG is viewed as

4

The only available data relates to federal government-owned foreign assets, which have been generally increasing throughout the period considered.

Adam B. Elhiraika and Annas H. Hamed

368

Table 4.

UAE sources of growth in regional and international comparisons

Country

Period

Growth

Capital

Labour

TFPG

UAE Algeriaa Egypt Kuwaita Venezuelaa MENA countries

1975–2002 1960–1997 1960–1997 1960–1997 1990–1997 1980–2000

0.047 0.031 0.057 0.022 0.028 –0.07

0.015 0.034 0.035 –0.015 0.012 0.99

0.052 0.006 0.011 0.056 0.023 –0.06

–0.020 –0.009 0.011 –0.018 –0.006 –1.00

Source: Development Indicators in the UAE (various issues), Ministry of Planning, Abu Dhabi. a Makdisi and Limam (2000); Hakura (2004).

that part of GDP growth that is not explained by growth in factor inputs and is therefore attributable to technical progress. However, as we explain below, this variable may well reﬂect the effects of many factors that are not accounted for in the model. This fact explains why, as Barro (1999, p. 119) notes, the usefulness of the growth accounting framework, as a preliminary step for the analysis of fundamental determinants of factor growth, depends on the condition that these factors are substantially independent from those that matter for technological change. a is the elasticity of output with respect to capital, which is equal to the capital-income share. In addition to physical capital and labour, numerous studies (e.g. Hahn and Kim, 2000) include human capital, usually measured by average years of schooling of the adult population. Because such measures are not available for the UAE for the entire period under review, it is not possible to include a human development variable in the GDP decomposition. The results of the output growth rate’s decomposition5 for the UAE are presented in Table 4 along with the decomposition of GDP growth rates for some comparable countries (the latter from Makdisi and Limam (2000)). For the UAE, labour contribution amounted to 110% of the overall average GDP growth rate, whereas capital contribution was 32% and contribution by TFPG was 42%. The results clearly indicate the overwhelming predominance of labour over capital and that on average TFPG was negative. This ﬁnding is consistent with the results for oil-dependent countries such as Kuwait and Venezuela. In fact, all the ﬁve MENA oil-dependent countries included in the sample used by Makdisi and Limam (2000) recorded negative TFPG. But, for the entire sample of 92 developing countries, Makdisi and Limam (2000) ﬁnd evidence of predominance of capital contribution over that of labour and TFPG.

5

The analysis here is based on real output data obtained from the 2000 and 2003 editions of the International Financial Statistics of the IMF and physical capital and labour data taken from various issues of the Economic Development Indicators (Ministry of Planning). For alternative methods of estimating the growth accounting model see Bisat et al. (1997) and Hakura (2004).

Explaining Growth in an Oil-Dependent Economy

369

Using a different data set, econometric estimates by Bisat et al., (1997) also indicate that TFPG is negatively related to aggregate GDP growth in the UAE. However, they conﬁrm that on average TFPG has a positive contribution to non-oil GDP growth in Saudi Arabia and the UAE, and that the contribution of TFPG to oil GDP growth was negative. Bisat et al., (1997, p. 24) argue that the negative average TFPG rates obtained for the total GDP of oil-exporting countries is misleading since it mainly reﬂects ﬂuctuations in growth induced by wide changes in oil prices and revenues that constitute the main omitted variables. In other words, changes in GDP growth that are not accounted for by labour and capital inputs are by deﬁnition attributed to TFPG. But, TFPG is dominated by changes in oil prices and revenues, given the high share of oil output in total GDP. This is a measurement problem that exists even if TFPG has no causal relationship with these changes. In fact, should it be possible to isolate the oil price effect, the true production efﬁciency in oil-exporting countries is likely to be greater than the measured one. Unfortunately, we do not have enough data for the UAE to decompose labour and capital by the oil and non-oil sectors. This decomposition is especially relevant in view of observed relationship between changes in oil prices and measured TFPG. Aside from measurement problems, low TFPG in the UAE may be attributed to the fact that trade openness and liberal economic policies in this case promote reexports rather than genuine exports based on production of goods. Indeed, productivity increases have not been an important source of growth in some Arab countries owing to economic policies and external shocks that favoured investments that were often not conducive to improvements in total factor productivity (TFP) (Bisat et al., 1997). Declining investment rates in some countries then compounded the negative trend of productivity growth. In the UAE, however, domestic investment increased since the turn of the 1990s following a sharp decline in the 1980s, although huge price and subsidy distortions (discussed further below) remained important and may be blamed for inducing biases in private investment decisions in favour of housing, services and reexports. The ﬁnding regarding the contribution of labour to total output growth is in contrast with the Labour Report (Ministry of Labour, 2000) that asserts that the labour market is dominated by foreign labour that is largely unskilled and has low productivity. The results of this study suggest that labour seems to be adequate for doing the jobs they are supposed to do in view of their overwhelmingly high contribution to GDP growth. But, the labour report is perhaps correct as regards the lack of skills needed to increase the productivity of capital, and this partly explains the poor TFPG rate in the UAE. The abundance of relatively cheap foreign labour is another incentive for private investors to continue to favour labour-intensive investments and reexports at the expense of genuine capital-intensive investment. Meanwhile, because of price-ﬁxing policies and subsidies, public capital is likely to be inefﬁciently used. Price-ﬁxing and subsidies relate mainly to housing and land, local

370

Adam B. Elhiraika and Annas H. Hamed

farm production and energy cost. For instance, UAE citizens have access to lowcost farm and real estate loans and receive above market wages in government sector (see Section 4 for further discussion). While most of the direct and indirect subsidies, such as artiﬁcially low electricity cost,6 go to UAE nationals, these subsidies affect the relative cost of inputs and are likely to encourage low-return investment as well as the use of unskilled labour by domestic ﬁrms and households. Obviously, policies need to be introduced to eliminate price discrimination and increase factor productivity to ensure long-run competitiveness. The importance of education and training in this respect cannot be overstated. A well-trained labour force that is complementary to capital is particularly needed in a labour-scarce country like the UAE, where promoting sustainable growth might require more capital-intensive investments. 3.2. Econometric model of the determinants of UAE growth 3.2.1. Specification of the model The literature on growth proposes a great number of explanatory variables (see Barro, 1991; Gregorio and Guidotti, 1995). Besides theory, our choice of variables in this study is guided by the speciﬁc economic conditions and environment of the UAE as well as by data availability. The dependent variable is real GDP per capita growth. The independent variables are as follows. Although issues of causality between investment and growth may be theoretically debatable, empirical research points out to strong long-run relationship between the two variables. Empirical evidence indicates that capital accumulation, represented by gross domestic investment as a ratio of GDP, often induces faster output growth (see e.g. Batina, 1998). Growth in a country’s terms of trade means increased net exports and aggregate output. The opposite may also be true. Therefore the terms of trade variable can take either sign. Because of lack of data, this variable is represented by changes in the oil price index (DOPI), which should be a good proxy as oil represents more than 50% of UAE exports. Natural resource abundance as the literature suggests is supposed to have an adverse impact on GDP growth owing mainly to the ‘‘Dutch disease’’ (see Sacks and Warner, 1997). High natural resource price and revenues attract factors of production from other sectors that decline as the abundant resource sector expands. More importantly, high natural resource prices often lead to an overvalued exchange rate, which in turn hurts the export sector in general and the exports of other sectors in particular. This added to wasteful use of resources induced by the natural resource abundance and lack of urge for proper resource

6

For example, in 2003, expatriate households were charged DH 0.15/KW of electricity, whereas nationals paid DH 0.05. The same price discrimination also applied to domestic and foreign companies outside free zones.

Explaining Growth in an Oil-Dependent Economy

371

management can impact negatively on growth. Natural resource abundance is represented in the model by oil exports as a percentage of GDP (XOIL). We have also included the non-oil exports-GDP (XNOIL) rate in the model to assess the impact of economic diversiﬁcation on growth. Moreover, the sum of exports and imports as a percent of GDP was used to test whether trade openness stimulates growth by encouraging competition and international technology transfer. Changes in the income of trading partners are envisaged to have a direct impact on growth in a country that is heavily dependent on foreign trade. However, given the nature of demand for UAE exports, GDP growth in the UAE may not be signiﬁcantly responsive to changes in the incomes of its trading partners. That is the income elasticity of demand for oil, the main UAE export commodity, is believed to be low such that changes in the incomes of its trading partners have little effect on exports and income. A great deal of theoretical and empirical literature underscores the importance of ﬁnancial development in economic growth, but several transmission mechanisms have been identiﬁed and controversy exists in the literature as to how to measure ﬁnancial development7 (see Demirguc-Kunt and Levine, 1996; Levine and Zervos, 1996). Two measures of ﬁnancial development are adopted in this study. These are the M2/GDP ratio and bank credit to the private sector (BCPS) as a ratio of total credit. Both measures are postulated to have positive coefﬁcients on growth. Human capital, accumulated through education, has long been acknowledged as an important factor in facilitating economic growth. In most regressions this variable turns out with a positive coefﬁcient (Barro, 1991). Investment in human capital enhances technological progress and raises labour productivity. This is expected to stimulate growth directly as well as indirectly by augmenting the productivity of capital. Barro and Lee (2001) developed an educational attainment census method to construct estimates of human capital that are superior to enrolment ratios. However, owing to unavailability of these estimates, we have used the secondary school enrolment ratio (SSER) as a traditional measure of human capital. Macroeconomic stability ‘‘sends important signals to the private sector about the direction of economic policies and the credibility of the authorities’ commitment to manage the economy efﬁciently. Such stability, by facilitating long-term planning and investment decisions, encourages saving and capital accumulation by the private sector’’ (Hadjimichael and Ghura, 1995). The most frequently used proxies for macroeconomic stability are variability of the

7

It should also be noted that empirical studies of the relationship between ﬁnancial and real development include cross-country studies that lump together data from countries with different levels of ﬁnancial and/or real development as well as country-speciﬁc studies, which often use different time frames and model speciﬁcation rendering comparisons across countries difﬁcult.

372

Adam B. Elhiraika and Annas H. Hamed

inﬂation rate and of the ﬁscal deﬁcit/GDP ratio. Since budget deﬁcits are a relatively recent phenomenon in the country, we have used the standard deviation of inﬂation to account for the possible inﬂuence of macroeconomic instability and uncertainty in UAE growth. Measures of macroeconomic instability are expected to have no signiﬁcant deleterious impact as the UAE economy was very stable during the period considered. The effects of government consumption and deﬁcits on growth depend on the level of consumption (GCON) and how the deﬁcits (DEF) are ﬁnanced. Generally, government spending is believed to induce growth. But, through tax distortions and adverse effects on saving, high levels of government consumption and deﬁcit in particular may frustrate growth (Barro, 1991). Accordingly, the coefﬁcients of these variables can take either sign. There is no income or proﬁt tax (aside from taxes on the proﬁts of foreign ﬁnancial institutions operating in the country) in the UAE, and as mentioned previously deﬁcits are ﬁnanced mainly from external assets with no direct effect on the exchange rate and/or private saving. Therefore, the coefﬁcients of these variables can take either sign. 3.2.2. Model estimation and discussion of results The model was estimated using data from various issues of the International Financial Statistics of the IMF, the Annual Reports of UAE Central Bank and the Economic Development Indicators published by the Ministry of Finance, UAE. Prior to estimation, we have investigated the time-series properties of all the variables of the model using the most commonly used tests: an augmented Dickey–Fuller (ADF) and Phillips–Perron (PP) tests. The latter non-parametric test is known to have more power than the parametric ADF. But the power of the two tests decreases in the presence of structural breaks in the series, in which case the test may fail to reject the null hypothesis of unit root when the series is actually stationary (see Enders, 1995). The two tests conﬁrm that all the variables8 of the model are stationary at the 5% level or better with all normalized variables, such as the investment/GDP and the broad money/GDP ratios, being entered after taking the ﬁrst difference (Table 5). Since the stationarity tests reject the null hypothesis of unit root, it is indicative that there was no signiﬁcant structural change in the series. Since all variables are stationary, a dynamic form of the model was estimated using ordinary least squares (OLS) technique and a general to speciﬁc estimation strategy. Owing to high correlation among trade variables, each of them was entered separately. Likewise, oil price and exports were entered each at a time. In view of possible endogenity of domestic investment, we estimated the model using two-stage least squares (2SLS) technique with domestic investment entered

8

The only exception is the trading partners’ GDP per capita growth and the M2-GDP variables that are only stationary at the 10% level according to the ADF test, but at the 1% level according to the PP test.

Explaining Growth in an Oil-Dependent Economy

Table 5.

373

Unit root tests and structural change

Variable GDP per capita growth rate Trading partners’ GDP per capita growth DM2–GDP (%) Inﬂation rate DGross domestic investment (% of GDP) DOil export (XOIL) in % of GDP DNon-oil export in % of GDP (XNOIL) DOil price index (DOPI) (%) DSSER DGovernment deﬁcit (% of GDP) DGovernment consumption (% of GDP)

ADF-test

PP test

–4.51 –2.63 –2.83 –5.80 –4.43 –4.44 –3.50 –5.98 –3.70 –4.88 –5.57

–4.75 –3.75 –3.92 –4.10 –4.46 –4.43 –3.88 –6.00 –3.88 –6.01 –4.69

with various sets of explanatory variables including trade openness, growth in the income of trading partners, standard deviation of inﬂation, ﬁnancial deepening9, human developments and government consumption and ﬁscal deﬁcit. Generally, the coefﬁcients of all explanatory variables had the expected signs, but only two of them were statistically signiﬁcant: natural resource abundance, as represented by changes in the oil revenue/GDP ratio, and the terms of trade variable, as represented by changes in oil prices (equations (1) and (2) of Table 6). All other variables were found to have no signiﬁcant impact on UAE growth and as equation (3), for example, suggests the estimated 2SLS model is statistically weak. In line with early discussion, these results raise the question of why the non-oil sector and related variables appear to have no important effect on UAE growth so far. To shed light on this question, we concentrate here on the implications of the statistically robust results of the model. The results for equations (1) and (2) are generally strong in terms of R2 and the F-statistics. The constant term has always turned out with a negative coefﬁcient, which is consistent with the earlier ﬁnding of a negative average TFPG rate. Moreover, two dummies, one for the First Gulf War (1984–1988) and the other for the Second Gulf War (1991) always had negative, but insigniﬁcant effects. The lag dependent variable always had the right positive and signiﬁcant coefﬁcient. Natural resource abundance (XOIL) produced the most interesting result. As mentioned previously, the literature postulates a negative relationship between natural resource abundance and growth. However, the case of the UAE, and perhaps that of all oil-dependent Gulf States, is clearly different. Before the discovery of oil, economic activity was extremely limited owing to the virtually

9

Represented by the broad money to GDP ratio and bank credit to the private sector as a ratio of total credit.

374

Adam B. Elhiraika and Annas H. Hamed

Table 6.

Econometric results Equation (3) (2SLS)a

Variable

Equation (1)

Equation (2)

Constant Lagged dependent variable Change in oil export (XOIL) in % of GDP Change in oil price index (DOPI) (%) Change in non-oil export in % of GDP (DXNOIL) DM2-GDP (%) Standard deviation of inﬂation rate DGross domestic investment (% of GDP) DSSER R2 F-statistic LM test (F-Form)

–0.104 (0.091) 0.190 (1.85)b 0.96 (5.89)c

–3.49 (2.01)b 0.26 (1.84)b –

–0.951 (-0.40) – –

–

0.22 (5.44)c

–

–

0.295 (0.63)

–

– – –

– – –

–0.59 (–1.44) 0.012 (0.18) 0.86 (1.50)

–

0.20 (0.36) 0.02 0.40

– 0.65 18.6 1.16 [0.33]

1.26 [0.277]

Notes: Dependent variable is real GDP per capita growth rate; t-values within () brackets; LM test is the Lagrange Multiplier test for error autocorrelation with critical values within [ ] brackets. a Instruments consisted of lagged values of dependent and explanatory variables. b Signiﬁcance at the 55% level. c Signiﬁcance at the 1% level.

‘‘absolute’’ scarcity of other resources. There were no developed economic sectors to be adversely affected by increased investment and production in the oil sector. In fact, initial overall growth conditions relied on the growth of this sector, which stimulated other sectors by raising aggregate demand and by providing the funds needed for increased investment in infrastructure and other non-oil activities. At the same time, the UAE currency is pegged to the US dollar and the exchange rate policy is largely neutral as regards its relative effect on different economic sectors. For example, a weak dollar and hence weak dirham may boost oil as well as non-oil exports and vice versa. At the same time, the exchange rate peg also helps to create a stable macroeconomic environment that is conducive to private investment and growth. As explained before, the exchange rate policy is justiﬁed on the ground that, oil, the largest export commodity, is priced in terms of US dollars. The UAE would not be able to maintain the peg without the huge foreign reserves accumulated from sales of the abundant crude oil. In short, whereas the abundance of a natural resource such as oil may be a curse in a country with diverse resources, it is deﬁnitely a blessing for a country with little other resources. Changes in the oil price index (as a proxy for terms of trade) have a highly statistically signiﬁcant effect on UAE growth. Oil represents about 50% of GDP and about 50% of total exports over the entire period considered, while exports

Explaining Growth in an Oil-Dependent Economy

375

accounted for about 70% of GDP. Therefore, growth in the UAE is predominantly export-led and oil prices play a key role. International oil markets exogenously determine oil prices. Because of OPEC’s (Organization of Petroleum Exporting Countries) policies and agreements that ﬁx production quotas and the nature of competition in the oil market, changes in the oil price have an immediate effect on oil GDP since individual countries may not be able to adjust quantities in such a way as to offset the price effect. Thus, increases in oil prices or the terms of trade will normally have a robust positive impact on GDP growth. The insigniﬁcant coefﬁcient of the standard deviation of inﬂation is a testimony to the stability of the economy and the absence of adverse growth effects owing to economic uncertainty. The ﬁndings of positive but weak coefﬁcients for gross domestic investment and non-oil exports seem to reﬂect the fact that domestic investment is largely in the construction and services sectors, while non-oil exports are dominated by re-exports rather than genuine goods output. However, the contribution of the non-oil goods sector to domestic output and exports has been rising in recent years. Therefore, the non-oil sector is likely to play a more important role in the future should this trend be sustained. The ﬁnding of no signiﬁcant role for ﬁnancial deepening, and by implication monetary policy, is also conceivable since the UAE economy is capital abundant and investment and growth in the country are limited by the availability of proﬁtable investment opportunities rather than the availability of capital (see Farzin, 1993). It is also not surprising that government consumption and deﬁcits appear to have no important effect on growth in view of the nature of these variables as discussed previously. To sum up, econometric evidence conﬁrms the heavy dependence of the UAE economy on oil exports and income. Therefore, more and concerted efforts need to be done to accelerate economic diversiﬁcation and achieve sustainable nonoil-based growth in the future.

4. Markets and institutions In this section, we brieﬂy discuss the role of key markets and institutions in the UAE growth process, focusing on the labour and capital markets that present the greatest policy challenges and require well-designed institutional interventions. We also discuss the role of institutions10 in terms of macroeconomic environment and investment climate.

10

Owing to lack of information, the behaviour of agents–namely ﬁrms and households–is not discussed here.

376

Adam B. Elhiraika and Annas H. Hamed

4.1. Labour market Economic progress engendered rapid population growth and signiﬁcant changes in the structure of the UAE labour market, which is dominated by immigrant labour. The immigrant population rose from 63.9% of total population in 1975 to 75.6% in 1995 before declining to about 70% in 2002. At the same time, the share of foreign labour increased from 85% in 1975 to 90.9% in 1995 and then decreased to about 87% in 2002. The rate of UAE women participation11 in the labour force decreased from 11.3% in 1975 to 10.1% in 1995. This is probably due to the fact that many women used to participate in traditional economic activities that have disappeared with the advent of modern activities that require education. However, with increased female education, the participation rate of UAE women in the labour force rose to 26.5% by 2002 compared to 76.6% for males. The average rate of local UAE population participation in the labour force of 60.4% as estimated by the Ministry of Planning12 in 2002 is quite low by international standards. The labour market is segmented along different lines with limited labour mobility between them owing to wage rigidities, skills differences and institutional and cultural factors. In the public sector, UAE nationals are more highly paid than expatriates with the same qualiﬁcations and who perform the same jobs. For example, using a 1997 survey data from the Emirate of Abu Dhabi, Al-Awad and Elhiraika (2002) ﬁnd that the income difference between citizens and foreigners is 62% in the case of skilled males and 55% in the case of unskilled men. For skilled and unskilled women, the respective difference is 81% and 55%. Moreover, women are consistently less paid than their male counterparts, irrespective of skills and nationality: the average income of a man is higher than that of a woman by 54% in the case of citizens and by 104% in the case of immigrants. This contrasts sharply with the distribution of skills, which is relatively higher among women (both local and immigrants). In addition to the nature of jobs that are normally preferred by women, gender income inequality in the UAE is attributable to labour regulation. Men are entitled to a variety of fringe beneﬁts, such as housing support, to which women are not entitled. Al-Awad and Elhiraika (2002) conclude that gender is the most important factor behind income inequality in the UAE, followed by skill level and nationality. Relatively high government wages together with job security, social status and other beneﬁts, make the government the most attractive employer for UAE nationals. The private sector relies on foreign labour that is less well paid and works for ﬁxed contracts. Foreign workers account for 99% of labour in the private sector (Ministry of Labour, 2000). On average, the level of education is

11

This rate refers to the number of working UAE women as a percentage of economically active UAE women population (aged 15 years and above) or UAE women labour force. 12 See Ministry of Planning ‘‘UAE in Figures 2002’’, Abu Dhabi, UAE.

Explaining Growth in an Oil-Dependent Economy

377

higher among immigrants compared to nationals, but still the overall average education level is low reﬂecting the dominance of unskilled labour. This fact is also attributable to the nature of investment that is dominated by labour-intensive projects in the services, transportation and other sectors. In terms of gender and owing largely to social factors, women participation in the labour force is very low and because female mobility is very restricted, they are on average paid far less than men are. Labour market problems have aggravated in recent years. First, with the government unable to provide enough employment to national entrants, there is a growing problem of unemployment among them. This may be considered as transitional or voluntary unemployment because UAE nationals prefer to wait for a government job rather than taking low-paid and demanding private sector jobs. Therefore, government policy reform is needed to remove the unrealistically high government wages for nationals that lead to such unemployment. In Saudi Arabia, for instance, the reforms consisted of ‘‘an array of measures affecting the quantity (quotas) and price (wage subsidies to private sector, government wage restraint and fees and charges on foreign labour) as well as the quality (education and training) of labour’’ (Sassanpour et al., 1997, p. 26). However, employment quotas and subsidies cannot provide the long-term solution and an effective labour market policy should be based on removal of distortions created by government wage and employment polices besides increasing the market ability to match labour skills with employers’ requirements in the private sector. The absence of clear labour market policies as regards labour compensation in the face of continuous excess supply of labour from neighbouring countries created incentives for investors to favour labour-intensive projects. Government subsidies in terms of low-cost public services are often blamed for encouraging the supply and use of relatively cheap and low-educated foreign labour. Until the early 2000s, in addition to tax-free wages and salaries, these subsidies included free health services to expatriate workers and their families. However, many of the subsidies have been cut recently. Labour market reforms are required to reduce discrepancy between public and private sector wages and encourage more nationals to participate in the private sector; to encourage recruitment and retention of skilled labour; the development of labour skills through training and to induce increased women participation, given the clear domination of women in the country’s tertiary institutions and the low current rate of women participation in the labour force. In line with these arguments, the 2000 Labour Report of The Ministry of Labour proposes that the UAE government should attempt to develop a model of growth similar to that of Singapore. The hallmark of the Singaporean model is that the country managed in less than 15 years to make a drastic move away from low-skill labour-intensive industries to high-skill high-tech industries. Singapore deliberately implemented strategies that made labour relatively expensive such that businesses began to favour capital-intensive investment that eventually

378

Adam B. Elhiraika and Annas H. Hamed

spurred rapid technology transfer and adoption, creativity and innovation and faster technical progress. The Government of Singapore invested heavily on education and training to ensure that industry’s demand for skilled labour is met. Many of the elements of the model of Singapore may be replicated by UAE, especially in view of relative abundance of capital in the country and its proximity to the Middle East and African markets including the oil-rich GCC countries. Like Singapore, the UAE has a relatively favourable business environment, a strong ﬁnancial sector and a modern physical infrastructure. However, it remains to be seen if the UAE could effectively implement human resource development strategies similar to those that underpinned market-oriented reforms in Singapore and at the same time attract high-tech investments. 4.2. Financial markets and institutions The ﬁnancial system of the UAE is a bank-based system. There are 21 national banks with 310 branches and 26 foreign banks with 112 branches spread throughout the country. This extensive network of banking services contributed to a high degree of ﬁnancial deepening. Currently, the ratio of quasi money to GDP is about 45% and M2/GDP ratio was about 67% in 2002. These ratios reﬂect the ability of banks to mobilize sizeable savings that are often far greater than the domestic demand for credit. Consequently, large amounts of deposits are invested abroad. Nonetheless, ﬁnancial deepening has enhanced the availability of capital to private domestic investment. In general, banks in the UAE are well capitalized, and the Central Bank regulatory framework meets the international standards. However, until 2000 the country lacked formal capital markets. Owing to the absence of formal markets, investors used the over-the-counter (OTC) market to trade shares. Since the OTC was fragmented and not well regulated, some market participants subjected it to harmful speculations and manipulations. These manipulations inﬂated share prices, in some cases prices went up 10 times, in a way that was not consistent with companies’ performance or their expected future earnings. Such high share prices enticed many ill-informed investors to invest a great deal of their wealth in stocks, and even some borrowed to buy stocks. Suddenly, the bubble burst and many lost a great part of their investments, while others defaulted on their bank loans. As a result, the public has lost conﬁdence in stocks and this has brought the initial public offerings (IPOs) to a halt. This highlights the need for effective regulatory institutions and government support to create a regulated and ofﬁcial stock market. Following the approval of a law regulating the issuance and exchange of securities in 2000, two stock exchanges have been launched in Dubai and Abu Dhabi. The operations of these exchanges are still limited with the combined value of traded shares of DH 3.87 billion by the end of 2002. This value accounts foro1% of the total ﬁnancial assets in the country. It is necessary for the country to develop its capital markets, because in the future these markets will

Explaining Growth in an Oil-Dependent Economy

379

be relied upon for the mobilization of resources and equity ﬁnancing for the private sector. 4.3. Investment climate We analyse the investment climate in the UAE in terms of infrastructure, the macroeconomic environment including monetary, ﬁscal and external policies, beside the laws that are relevant to investment decisions. Irrespective of the efﬁciency questions concerning provision of public services, as discussed earlier, the UAE has a modern physical and communication infrastructure that puts it on par with developed countries. In general, public investment in infrastructure creates an environment conducive for private sector activity, which can be relied upon to achieve sustainable economic growth in the long-run. In addition, public investment in roads, ports, airports, electricity and water utilities, education and training of work force raises productivity and increases economic efﬁciency. The most important feature of UAE monetary policy is that it is essentially targeted at maintaining the stability of the currency by pegging the dirham to the US dollar at a ﬁxed exchange rate of 3.67 AED/$. This exchange rate regime has been adopted since 1981 with great success owing to the huge foreign reserves the country holds. In addition to maintaining the stability of the currency, the policy of linking the dirham to the dollar has also contributed to a low inﬂation rate. However, the exchange rate policy can have a deleterious effect on growth depending on how import and export demand respond to changes in major currencies relative to the dollar and the size of the UAE trade outside the dollar area. Fiscal policy has played an active role in developing the economy, transferring wealth and promoting and supporting the non-oil sector. In the absence of income or consumption taxes in the UAE, ﬁscal performance of the country is largely dependent on oil revenue and income generated from foreign assets. Because not all oil revenues feed directly into current government revenues and owing to rapid population increases, the government’s growing wage bill and increasing maintenance costs, the economy has experienced budget deﬁcits since the early 1990s. These deﬁcits have become structural in nature and largely ﬁnanced by drawing down foreign assets and the earnings they generate. This method of ﬁnancing the deﬁcit is difﬁcult to sustain in the long run. Therefore, instead of deﬁcit ﬁnancing, ﬁscal policy should focus on structural adjustment including widening the non-oil revenue base, privatizing public enterprises, rationalizing government expenditure and reassessing subsidies and incentive programs. The UAE maintains a long tradition of liberal foreign trade. There are no capital controls or restrictions on foreign trade as is the case in many developing countries. The highest custom tax is 4%. The external sector of the country is solid. Since the ﬁrst oil boom in 1973–1974 the country has recorded a continual

380

Adam B. Elhiraika and Annas H. Hamed

external account surplus, which enabled it to accumulate substantial foreign reserves. In addition to the ﬁnancing of budget deﬁcits, these reserves act as a cushion that protects the economy from exogenous shocks. Well-designed laws are important for creating a climate conducive for investment and the smooth functioning of markets. The UAE government seeks to establish a legal framework to cover all aspects of business enterprises in the country. The most important laws in this framework are the federal company law, the commercial agency law and the intellectual rights protection law. In general, the legal framework favours local investors over foreign investors. This can be clearly seen in the following provisions of the company law (see Hamed, 2001). First, the proportion of national ownership of companies established in the UAE must not be lower than 51%. Second, branch ofﬁces of foreign companies must have a national sponsor, unless the foreign company establishes its branch ofﬁce based on an agreement with the federal government. Finally, the law forbids foreigners from owning shares in public joint stock companies directly, but allows ownership through mutual funds that are offered by commercial banks. However, the UAE as a federation allows considerable variations in company law across emirates. In some cases, for example Abu Dhabi company law, there is discrimination even against citizens of other emirates. As mentioned earlier, company law is more harmonized across emirates regarding free zones. The harmonization process should be extended to cover the laws regulating business outside the free zones. Overall, with the creation of the free zones, the UAE seems to have created a relatively facilitative business environment that is conducive to local and foreign investors. The main features of this environment are the creation of world-class industrial facilities and business support services as well as the updating of commercial laws and regulations to meet international standards and ensure effective protection for investors. For example, the Jebel Ali Free Zone has become one of the largest industrial complexes in the world, providing a wide range of options to investors including a 100% ownership of investment. Since 1995, the UAE has become a member of the World Trade Organization and this will have a direct impact on the nature and efﬁciency of investments that have to compete internationally. It is also important to note that favourable tax laws and political stability make the investment climate in the UAE attractive. 5. Conclusion and policy implications The objective of this paper was to explain the determinants of growth in the UAE as an oil-dependent economy. Like the case of other oil-dependent countries in the region, overall GDP performance in the UAE ﬂuctuated widely over the period 1975–2002. However, the UAE seems to have fared better than most of these countries in terms of a relatively high and positive average growth rate during this period. The UAE economy has been very stable, remarkably open

Explaining Growth in an Oil-Dependent Economy

381

and has the highest share of non-oil income in total GDP. Since the early 1990s, non-oil exports and income have generally exceeded their oil counterparts. However, non-oil exports are dominated by re-exports, reﬂecting the narrow domestic production base and continued heavy reliance on oil revenues. Achieving sustainable growth in the long run requires greater investment in the production of genuine export goods and increased production efﬁciency in various sectors. In analysing GDP growth by source, we found labour to have the most important contribution, followed by capital, while TFP had a negative average contribution. It has been argued that the negative TFPG is perhaps a reﬂection of the quality of domestic investment, which is dominated by public investment and low productivity private investment in the services and re-export sectors. In addition, TFPG, being measured as a residual might reﬂect the effect of omitted variables; speciﬁcally huge changes in oil prices that tend to dominate changes in growth factors. Unfortunately, owing to data limitations we were not able to isolate the oil price effect within the growth accounting framework that provided preliminary analysis of the UAE growth in this study. Econometric results indicate that unlike the case of countries with diverse resources, natural resource abundance spurs growth in an oil-dependent economy such as the UAE. Changes in terms of trade, represented by changes in oil prices, are directly related to growth. These ﬁndings conﬁrm the heavy dependence of the UAE economy on the oil sector. Increased non-oil investment and exports in recent years appear to have positive but still weak inﬂuence on growth. It is important that these trends be accelerated to promote sustainable growth. In this context, the study draws the following main policy recommendations: 1. Institutional reform is called for to address labour market distortions that create an over supply of labour in the public sector and raise the demand for low-paid, low-educated labour in the private sector. Thus, market segmentation by skill, nationality or public versus private sectors should be abolished and subsidies and incentives rationalized. In view of the rapid expansion of tertiary education, relative scarcity of labour and the abundance of capital, the UAE may well emulate the model of Singapore that has successfully moved from low-skill, labour-intensive industries to high-skill, capital-intensive industries. Investment in human capital should continue to focus on improving necessary skills to augment capital investment. 2. To reduce exposure to exogenous shocks, efforts to diversify the economy ought to be broadened by encouraging increased domestic, especially private investment in genuine goods production and expanding output markets regionally and internationally. This will reduce reliance on crude oil exports and re-exports that provide only limited room for productivity improvement. 3. The domestic ﬁnancial system should be widened in such a way as to be able to ﬁnance increased real domestic investment. In particular, the development

382

Adam B. Elhiraika and Annas H. Hamed

of a well-functioning equity market should be seen as an opportunity for encouraging the return of capital to the country, broader participation of domestic investors and inducement of foreign investors who bring in new technology. 4. Public deﬁcits ought to be addressed in a framework that takes into account future ﬁscal sustainability as a prerequisite for achieving sustainable growth in the long run. Fiscal policy should focus on structural adjustments that include widening the non-oil revenue base, privatization of public enterprises, rationalization of government expenditure and reassessing subsidies and incentive programs.

Acknowledgments The authors would like to acknowledge ﬁnancial support from the Economic Research Form (ERF) for Arab Countries, Iran and Turkey, and express their gratitude to the editors of this book and two anonymous referees for their valuable comments. References Al-Awad, M. and A. Elhiraika (2002), ‘‘Gender, skills, nationality and income inequality in the United Arab Emirate’’, Journal of Social Affairs, Vol. 19(76), pp. 314–328. Askari, H. and M. Jaber (1999), ‘‘Oil-exporting countries of the Persian Gulf: what happened to all that money?’’, Journal of Energy Finance and Development, Vol. 4, pp. 185–218. Barro, R. (1991), ‘‘Economic growth in a cross-section of countries’’, Quarterly Journal of Economics, Vol. 106(2), pp. 407–433. Barro, R. (1999), ‘‘Notes on growth accounting’’, Journal of Economic Growth, Vol. 4(2), pp. 119–137. Barro, R.J. and J.W. Lee (2001), ‘‘International data on educational attainment: updates and implications’’, Oxford Economic Papers, Vol. 53(3), pp. 541–563. Batina, R. (1998), ‘‘On the long-run effects of public capital and disaggregated public capital on aggregate output’’, International Tax and Public Finance, Vol. 5(3), pp. 263–281. Bisat, A., M. El-Erian and T. Helbing (1997), ‘‘Growth investment and saving in the Arab countries’’, IMF Working Paper WP/97/85. Demirguc-Kunt, A. and R. Levine (1996), ‘‘Stock market development and ﬁnancial intermediaries: stylized facts’’, The World Bank Economic Review, Vol. 10(2), pp. 291–321. Elhiraika, A. and Hamed, A. (2001), ‘‘Private investment and private sector development in the UAE: empirical and policy analysis’’, Paper presented at the ERF Annual Conference, October 2001, Amman, Jordan.

Explaining Growth in an Oil-Dependent Economy

383

Enders, W. (1995), Applied Econometric Time Series, New York: Wiley. Farzin, Y.H. (1993), ‘‘Importance of foreign investment for the long-run economic development of the United Arab Emirates’’, World Development, Vol. 12(4), pp. 509–521. Ghanem, S.M. (1992), Industrialization in the United Arab Emirates, New Castle Upon Tyne, UK: Averbury. Gregorio, J. and P. Guidotti (1995), ‘‘Financial development and economic growth’’, World Development, Vol. 23(3), pp. 433–448. Hadjimichael, M. and Ghura, D. (1995), ‘‘Public policies and private savings and investment in Sub-Saharan Africa: an empirical investigation’’, IMF Working Paper 95/19. Hahn, C. H. and Kim, J. (2000), ‘‘Sources of East Asian growth: some evidence from cross-country studies’’, Paper prepared fro the Global Development Project Explaining Growth initiated by the World Bank. Hakura, D. (2004), ‘‘Growth in the Middle East and North Africa’’, IMF Working Paper No. 04/56. Hamed, A. (2001), ‘‘Public and private investment in Abu Dhabi’’, in: F. Al-Faris, editor, The Economy of Abu Dhabi, A. Crown Prince Court, Abu Dhabi: Research and Studies Division Publications. Hamed, A. and A. Elhiraika (2001), ‘‘The budget institutions and procedures in the UAE’’, Paper prepared for the ERF project on Budgetary Processes in MENA countries, coordinated by Hadi Esfahani. IMF (2003), United Arab Emirates: selected issues and statistical appendix, Washington, DC, USA: IMF. Levine, R. and S. Zervos (1996), ‘‘Stock market development and long-run growth’’, The World Bank Economic Review, Vol. 10(2), pp. 323–339. Makdisi, S. and Limam, I. (2000), ‘‘Determinants of growth in the MENA countries’’, Paper presented at ERF workshop 6–7 April, 2001, Cairo. Ministry of Labour (2000), The labour report: moving from cheap unskilled labour to skilled labour and advance technology, Abu Dhabi, UAE: Ministry of Labour, (in Arabic). Sachs, J. and A. Warner (1997), ‘‘Natural resource abundance and economic growth’’, Harvard Institute for International Development Working Paper, pp. 1–150. Sassanpour, C., Joharji, G., Kireyev, A. and Petri, M. (1997), Labour market challenges and policies in the Gulf cooperation council countries. In: A. Jibili, V. Galbis and A. Bisat, editors, ‘‘Financial systems and reforms in the Gulf cooperation council countries,’’ Washington, DC: IMF, 1997, pp. 25–38. Solow, R. (1956), ‘‘A contribution to the theory of economic growth’’, Quarterly Journal of Economics, Vol. 70(1), pp. 65–94.

This page intentionally left blank

CHAPTER 13

Institutions, Household Decisions, and Economic Growth in Egypt Ragui Assaad Abstract I argue in this paper that the Egyptian government’s longstanding policy to guarantee employment in the government for upper secondary and university graduates, combined with educational policies that strongly favor these levels of education, has had a highly distorting impact on the production and deployment of human capital in the Egyptian economy. By trapping significant portions of existing human capital in unproductive government employment and inducing households to invest in the wrong kind of human capital, this combination of policies in ultimately responsible for the low productivity of human capital in the Egyptian economy.

Keywords: human capital, household decisions, institutions, social norms, gender, returns to schooling, Egypt JEL classifications: O12, J24, I21, J45, H52 1. Introduction Macroeconomic analyses of growth in Egypt indicate that labor contributes very little to growth, and that human capital, and in particular female human capital, contributes little to increases in total factor productivity (Kheir-El-Din and Moursi, 2006). These puzzling results come at a time when the role of human capital in generating growth is being increasingly stressed in the international literature and when investment in female human capital, internationally, is deemed to be one of the activities with the highest social returns (cf. Schultz, 1995). One of the main objectives of this paper is to shed light on these seemingly CONTRIBUTIONS TO ECONOMIC ANALYSIS VOLUME 278 ISSN: 0573-8555 DOI:10.1016/S0573-8555(06)78013-0

r 2007 ELSEVIER B.V. ALL RIGHTS RESERVED

386

Ragui Assaad

puzzling results by analyzing the microeconomics of the production and deployment of human capital in the Egyptian economy, with a particular emphasis on the institutional contexts in which these decisions are made. Making productive use of human resources is particularly critical at this stage of Egypt’s development as the country enters into its ‘‘demographic window of opportunity.’’ This is the phase in the demographic transition when the ratio of the working age to the dependent population rises, providing at least a potential for more rapid growth, through a reduction in the dependency ratio and opportunities to deepen investments in both human and physical capital. For this potential to be realized, institutional impediments to the efﬁcient production and deployment of human capital must be reduced. Moreover, because women constitute at least one-half of the working age population, achieving this potential will in part depend on the extent to which they are able to participate in productive employment. The main argument of the paper is that the longstanding policy of the Egyptian government to guarantee employment in the government for upper secondary and university graduates has distorted household decisions as to the level and type of human capital to acquire and has resulted in the entrapment of signiﬁcant portions of the human capital in unproductive government employment. With the distorted signals they receive from the labor market, the educational decisions of households are strongly shaped by the desire to ensure access to lifetime employment in the government, an objective that might well be at odds with realizing the productivity beneﬁts of education. Accordingly, they invest heavily in forms of education – such as technical secondary and higher technical institute education – that have very low returns in the private sector. Faced with strong demand for such education from the population, the government supplies it, at the expense of being able to guarantee basic education of decent quality to all those who are eligible for it. It is easy to see how this combination of educational and hiring policies, when applied over a long period of time, results in the distortion of households’ human capital decisions and the misallocation of human capital in unproductive activities, ultimately leading to the observed low productivity of human capital in the economy. In the case of women, the situation is further compounded by social norms deﬁning what constitutes appropriate female employment. These norms translate into signiﬁcant barriers to entry for women into the private sector, and a strong preference for the more egalitarian and family-friendly government employment among educated women. Norms that restrict women’s employment to a limited range of labor market segments deﬁned by occupations, types of workplaces, and locations, result in overcrowding in these segments and depressed wages. There is also considerable evidence that suggests that the private sector in Egypt, in stark contrast to the public sector, is highly inhospitable to married women. Working women’s careers are thus artiﬁcially truncated at marriage unless they are able to obtain employment in the public sector.

Institutions, Household Decisions, and Economic Growth in Egypt

387

In Section 2, I examine the growth implications of the demographic transition in Egypt. In Section 3, I examine the institutional setting within which household decisions on human capital investments are made, with a particular focus on government policies on education and public sector hiring. Section 4 examines how a move away from excessive reliance on public sector employment is hampered by social norms about what constitutes appropriate employment for women. Section 5 concludes the paper. 2. Growth implications of the demographic transition Salehi-Isfahani (2006) identiﬁes three main decisions households make that have direct impact on economic growth: the decision to save, to accumulate human capital, and to procreate. He also discusses in some detail the time allocation decision by which the household determines how much time and effort to allocate to market work, home production, and leisure. In this section, I focus on the changing patterns of procreation in Egypt in the recent past and their implications for future economic growth. In subsequent sections, I address the decision to accumulate human capital and to allocate time between market and home work. The literature in household economics has clearly demonstrated that parental decisions about child quantity (fertility) cannot be dissociated from decisions about child quality (human capital investments) (Becker, 1991). Moreover, decisions about education are clearly dependent on the returns to such education, which is closely related to opportunities for, and constraints on, labor market participation. For the sake of expositional clarity however, I will tackle each of these areas sequentially, making sure to make the necessary linkages when appropriate. Fertility rates have declined signiﬁcantly in Egypt in recent years. As shown in Table 1, the national total fertility rate (TFR) declined from a high of 6.7 in Table 1. Egypt: TFR 15–49 Year 1946–1947 1959–1960 1975–1976 19881 19921 19951 20001 20051

Urban

Rural

National

– – 4.6 3.6 2.9 3.0 3.1 2.7

– – 6.1 5.6 4.9 4.2 3.9 3.4

6.3 6.7 5.5 4.5 3.9 3.6 3.5 3.1

Source: For 1946–1947 to 1975–1976, CAPMAS, Panel on Egypt of the Committee on Population and Demography (1982, p.17). For 1988–2005, El-Zanaty and Way (2006), Egypt Demographic and Health Survey 2005. 1 Rates are for the 36 months period preceding the survey

388

Ragui Assaad

1959–1960 to 3.6 children per woman in 1995. Fertility decline slowed signiﬁcantly in the subsequent decade, especially in urban areas (Table 1). Fertility declines were driven in part by rapid urbanization in the 1960s and the 1970s. Since the mid-1970s, urbanization has slowed signiﬁcantly. The proportion of the population living in urban areas went from 38 percent in 1960 to 44 percent in 1976, only to stabilize at that level until 1996. Although fertility has continued to decline in rural areas from 5.4 births per woman in 1988 to 3.9 in 2000, the decline has stalled in urban areas from 1992 to 2000 at about 3 births per woman (El-Zanaty and Way, 2006). From 2000 to 2005, the decline in fertility appears to have resumed in both urban and rural areas. Another driver of fertility decline was the decline in the age at which women marry. In 2005, the median age at ﬁrst marriage among women aged 25–29 years was 21.3, about 1.5 years higher than what it was among women aged 45–49 years (El-Zanaty and Way, 2006, p. 93). This delay in marriage reﬂects the general increase in educational attainment and a general change in attitudes about early marriage (Singerman and Ibrahim, 2003). Perhaps the most important driver of fertility decline in Egypt was the substantial increase in survival rates among infants and children in the past three decades (Table 2). Infant and child mortality rates began declining slowly in the 1970s, but the decline accelerated signiﬁcantly in the 1980s as widespread use of oral rehydration therapy signiﬁcantly reduced deaths from diarrheal diseases (Miller and Hirschhorn, 1995). With some lag, the increased chances of survival of children translated into fertility decline, as parents increasingly adopted modern family planning practices. Contraceptive use has increased from 24 percent in 1980 to 59 percent in 2005 (El-Zanaty and Way, 2006). As a result of these developments, Egypt is now at the early stage of an important phase in its demographic transition referred to by some as ‘‘the demographic window of opportunity’’ (Barlow, 1994; Bloom and Williamson, 1998). According to the proponents of this approach, the implications of the recent Table 2. Mid-point 1967 1972 1976 1980 1983 1986 1988 1990 1993 1998 2003

Egypt: infant and child mortality (number of deaths per thousand) Infant Mortality

Under-ﬁve Mortality

141 146 124 108 98 73 74 62 66 44 33

243 238 203 157 140 102 103 85 84 54 41

Source: El-Zanaty and Way (2006), Egypt Demographic and Health Survey 2005.

Institutions, Household Decisions, and Economic Growth in Egypt

Figure 1.

389

Evolution of the age structure in Egypt (1950–2050)

60 50

Percent

40

0-14 15-24

30

25-64

Projection

65+

20 10 0 1940

1960

1980

2000

2020

2040

2060

Year Source: UN Population Division (2003).

fertility declines for the age structure of the population are even more important than its impact on population size. Population growth attributable to a fall in infant and child mortality that is not immediately matched by lower fertility has an immediate negative effect on economic growth by increasing the number of dependents relative to producers. Once fertility starts declining, however, there can potentially be a delayed positive impact on economic growth, as the share of the economically active population begins to grow two decades later.1 The impact of the demographic transition on the age structure of the Egyptian population can be clearly ascertained from Figure 1. The proportion of children from 0 to 14 years rose rapidly in the 1960s. It stabilized in the 1980s after a small decline in the 1970s. However, by the 1990s it began to decline steadily and is expected to continue to do so through 2025. The absolute number of those under 15 is expected to stabilize by 2010 and then begin declining by 2015. This suggests that even by sustaining the share of national resources devoted to basic education at present levels, there is now an opportunity to signiﬁcantly increase educational spending per child, and thus signiﬁcantly improve the quality of education. This sort of human capital deepening applies to both public resources devoted to education as well as to private resources under household control. The proportion of those in the prime working age, those between the ages of 25 and 54 years, began increasing in 1975 and will continue to increase steadily through 2025. To the extent that a stable or growing proportion of people in this

1

See also Williamson and Yousef (2002) and Tunali (1996) for a discussion of this phenomenon in the context of the MENA region.

390

Ragui Assaad

age group become economically active, this trend has the potential of signiﬁcantly increasing the number of workers and savers per capita and in turn the per-capita rate of economic growth, subject to labor productivity and savings rates remaining constant or rising. At the early stages, the increase in the number of adults in their prime working ages is accompanied by an increase in the numbers of those who are 15 to 24 years, those who constitute the bulk of new entrants to the labor market. For Egypt, such an increase occurred brieﬂy in the 1970s and resumed in the 1990s, both periods of severe pressure on the labor market and rising rates of youth unemployment. Projections indicate that the proportion of those in the 15–24 years age group will begin declining starting in 2005, relieving some of the acute labor supply pressure. Eventually, the number of older dependents will rise sufﬁciently to reduce the number of workers per capita, but that is still a relatively distant prospect for Egypt. The decline in the proportion of school-age children in the population, which has already begun, provides the country with an immediate opportunity to greatly enhance the quality of human capital investments. 3. Institutional context and investments in educational human capital In a paper in this volume, Pissarides argues that if we are to understand growth and development, we need to understand the creation and deployment of human capital (Pissarides and Ve´ganzone`s-Varoudakis, 2007, p. 140). This is precisely what I attempt to do in this section for the case of Egypt. Since the early 1950s, the Egyptian government has been investing strongly in expanding access to education. The effort was meant to substantially increase enrollment rates while at the same time ensuring access to the rapidly growing number of children. The evidence also suggests that parents are making signiﬁcant investments in the education of their children, which primarily take the form of private tutoring and enrollment in supplemental evening classes. Despite impressive growth in the number of graduates at all education levels, the question remains as to why Egypt has not reaped a signiﬁcant growth reward for its educational investments. The answer lies in two different areas. First, demographic pressures on the schooling system forced educational authorities to sacriﬁce quality to increase quantity. As demographic pressures begin to let up, there is now an opportunity to refocus efforts on educational quality. Second, by providing the wrong signals, the institutional setting in the labor market (essentially the public employment guarantee) has encouraged investment in essentially obsolete human capital and has entrapped much of that human capital in unproductive employment in the government. 3.1. Expenditures on education and school enrollment Egypt’s rising expenditures in education have done a great deal to expand access and increase enrollment rates at all educational levels. However, these positive

Institutions, Household Decisions, and Economic Growth in Egypt

391

results are limited by their comparatively small economic impact. The fact that Egypt has increased educational expenditure so hugely while not reaping enormous returns on this investment indicates that there may be some misallocation of funds or lack of attention to quality in pre-university or higher education. Exploration of expenditures per student over a period of 20 years reveal some signiﬁcant ﬂaws in the way Egypt spends money on education. First, there is a signiﬁcant discrepancy in spending per student between pre-university and higher education. In 2003, per student expenditures in higher education was up to four times as much as spending at the pre-university level. Second, the spending per student appears to increase with rising enrollment and decrease with lower enrollment, suggesting that overall spending varies more than proportionately with enrollment. Over the past 20 years, the government made special efforts to substantially increase school enrollments while accommodating the substantial increase in the pre-university-age population. From 1980 to 2002, the number of school-age students has grown by close to 115 percent, with the overall expenditures increasing by an astonishing 320 percent in real terms. Despite continued enrollment pressures, Egypt has increased its per-capita spending on primary, preparatory, and secondary students by 95 percent. Unlike growth rates in pre-university students, which have been fairly steady, university and higher institution enrollments have varied considerably during 1980–2002. After a slow decline in the late 1980s, enrollments in universities rose sharply beginning in 1992. Despite the period of decline, post-secondary enrollments had increased by 133 percent by 2002. Per capita student spending in higher education institutions varied enormously over this time period, declining when the government limited enrollment in the second half of the 1980s and early 1990s, and increasing when enrollments grew again thereafter. Per student spending rates were actually fairly similar in 2002 to what they had been in 1980, having increased by only 11 percent (Figure 2). It thus appears that funding for higher education in Egypt has varied more than proportionally with enrollment. The policy of equalizing per student expenditure at the university and pre-university levels by limiting university enrollment and decreasing university expenditures was fairly short-lived. It soon gave way to rapid increases in enrollment and a growing gap in per student funding between higher and pre-university education in the 1990s. These growth rates are somewhat deceptive; indeed, years of escalating education expenditure should have resulted in economic growth in Egypt as its human capital pool grew and improved in quality. The problem lies partly in the misallocation of both public and private spending on types of education that do not create the sort of workforce Egypt requires. While it is clear that Egypt is steadily making pre-university education a greater priority in its budget, there is still an enormous gap between higher education and pre-university expenditures per student. The Egypt Human Development Report of 1998/99 argues that there is still substantial misallocation of educational resources despite increasing

Ragui Assaad

392

2500

300

250

2000

200 1500 150 1000 100 500

50

0 1980

1985

1990

1995

2000

0 2005

Index (1980=100)

Per Student Spending (LE/yr in 1995 constant prices)

Figure 2. Real per-student spending and evolution of number of students enrolled, (pre-university and higher education) during 1980–2003

pre-university spending per student (left axis) higher education spending per student (left axis) index of pre-university students (right axis) index of higher education students (right axis)

Year

Sources: CAPMAS, Egyptian Statistical Yearbooks (various years) (number of student, all years); UNDP (2000) (educational expenditure: 1990/91 and 1996/97); Devarajan et al., (1996) (educational expenditure: 1980–1985).

expenditures (UNDP, 2000, pp. 77–78). While ﬁgures from the Egyptian Statistical Yearbook make it difﬁcult to differentiate between secondary, preparatory, and primary spending, UNDP assessments note that resources at the preuniversity level are biased toward the secondary level. Enrollments are far higher at the primary level, but it receives similar funding at the secondary level of education. By no means should one assume that the cost of primary education is comparable to secondary or higher education; however, Egypt’s lack of success at these lower levels of education denote the potential need for refocusing or reallocating funds toward these levels. In an attempt to supplement public educational expenditures that, despite steady increases, do not adequately prepare Egyptians for the job market, parents have signiﬁcantly increased their own spending on education. Since over 98 percent of Egyptian children are in public schools, this private spending usually takes the form of spending on private tutors or evening support groups for their children, which are often led by the children’s own school teacher. These personal investments are common across different regions and even levels of income. Although it is quite difﬁcult to obtain historical data on spending for private lessons, data from the Egypt Labor Market Survey of 1998 suggest that about one-half of primary school children and two-thirds of preparatory and secondary school children participated in some kind of private lessons in the year prior to the survey, with the majority taking the more expensive private lessons (Elbadawy et al., 2004). On average, the parents of a child who took private tutoring spent LE 350 per year on such lessons (ELMS, 1998). UNDP

Institutions, Household Decisions, and Economic Growth in Egypt

393

estimates that approximately LE 7 billion is spent on tutoring per year. This cost of private tutoring is only somewhat less than the average public expenditure per student (LE 393 in 1998), and negates the purported right of ‘‘free education’’ available in Egypt. The burden is particularly heavy on poorer families, who pay nearly 10 percent of their annual income in tutoring services (UNDP, 2000, pp. 84–85). Teachers can earn more from private tutoring than they can from their public school salaries, creating an incentive to focus their time and efforts more heavily on individual instruction. This certainly plays a role in the low quality of education offered at public institutions. To a certain extent, this is a self-perpetuating circle of demand for private tutoring and the low quality of education; the supply of teachers willing to spend more time in private lessons will not decrease until their public salaries are increased. The end result is that tutoring has become a substitute for basic education because of the low-quality instruction offered in schools (UNDP, 2000, pp. 84–85). Such individual family investments have become necessary because of the inadequacy of much of Egypt’s pre-university and post-secondary education. While educational expenditure is rising, Egypt’s Human Development Report rightly notes that it is not enough to focus on the acquisition of credentials, but that productive skills must increasingly become the focus of the education system (UNDP, 2000, p. 1). Increased expenditure on education has not been the solution in Egypt because of skewed demand from students seeking degrees guaranteeing public employment rather than market-based skills. The obsolete nature of the abilities acquired in Egypt’s educational system inhibit Egyptians from pursuing problem-solving skills that the private sector can use and beneﬁt from. Many students are encouraged to pursue commercial degrees, in which they are taught clerical skills that are of little use to the private sector. In the past, the demand for the more marketable analytical and problem-solving skills has not been overwhelming because of the safety net of public service. This policy, over 40 years, has altered the demand for education in such a way that many available degrees do little to improve students’ productive skills. The increasing expenditure on education needs to be focused more directly on skills that match the needs of the economy as Egypt gradually moves away from its government employment guarantee. 3.2. Role of the state as a dominant employer of educated workers The guarantee of public employment – part of an extensive nationalization drive in 1961 – has played an enormous role in shaping the demand for and the quality of Egyptian education at all levels. The length and impact of the policy must be understood before an adequate assessment of educational reform can be accomplished. Begun as a guarantee to publicly employ all university graduates, it was extended to vocational secondary school and technical institute (2-year post-secondary institutions) graduates and the policy was formalized in Law No.

394

Ragui Assaad

14 of 1964. It was later made permanent in Law No. 85 of 1973 (Assaad, 1997). Combined with the abolition of fees for higher education institutions in 1963, the employment guarantee for graduates provided a major boost to the demand for education. The employment guarantee greatly enhanced the private beneﬁts of university education and the abolition of fees signiﬁcantly lowered private costs (Richards, 1992, pp. 8–9). Since the policy was instituted in the early 1960s, the Ministry of Manpower has maintained a registry of applicants to public sector jobs where graduates are listed by year of graduation and type of degree. While graduates are removed from the registry as a matter of course if they obtain a public sector job, many do not give up their position in the queue if they get a private job, thus reserving their right to a public sector job when their turn comes up. By law, university graduates must wait at least 2 years and secondary school and technical institute graduates 3 years after graduation to apply to the Ministry of Manpower for a government job.2 By 1984, the waiting period between graduation and appointment had been extended to 31/2 years for university graduates and 4 years for vocational secondary and technical institute graduates and, by 1987, to 5 and 6 years, respectively (Handoussa, 1992). Although the policy is still on the books and the Ministry continues to receive applications from graduates, hiring through the centralized manpower allocation system of the Ministry of Manpower has been suspended in the early 1990s, but has not been formally abolished. As of 2004, the last cohorts of graduates to have been offered appointments in its entirety through the centralized manpower allocation system was the 1984 cohort of university graduates and the 1983 cohort of vocational secondary and technical institute graduates. Until 1978, graduates could either be placed in the government or in State-owned enterprises (SOEs), but since then the manpower allocation scheme has stopped applying to SOEs, which have been allowed to control their own hiring directly. Despite the suspension of he centralized manpower allocation system, graduates continued to be hired into the civil service. The Ministry of Education, whose hiring has always been independent of the centralized system continued to hire at a rapid pace and individual government agencies and ministries were also hiring graduates directly through a series of competitions and special arrangements. The graduate employment guarantee has had major implications for the size and composition of the public sector workforce in Egypt. As a result of the graduate employment guarantee scheme, Egypt has one of the highest proportions of total employment in the civil service in the world. The proportion of the civil service in total employment stood at 24 percent of total employment and 39

2

The waiting period was designed to allow male graduates to complete their military service. With the exception of the period from 1967 to 1973, when conscripts could serve indeﬁnitely, the duration of military service is normally 1 year longer for secondary and technical institute graduates than for university graduates.

Institutions, Household Decisions, and Economic Growth in Egypt

395

percent of non-agricultural employment in 1998, up from 19 percent and 32 percent, respectively in 1988. Despite the formal suspension of the graduate guarantee scheme, government employment continued to grow at 4.8 percent per year in the 1990s, twice the rate of overall employment growth (Assaad, 2002). However, even that rate of growth was signiﬁcantly below the 7.5–8 percent growth rates logged in the late 1970s and early 1980s (see Handoussa, 1992; Assaad, 1997). This surprisingly rapid increase in government employment during a period of structural adjustments and budget cuts requires some further explanation. The trend has to be seen against the backdrop of a long-term policy to guarantee employment for all university and vocational secondary school graduates that, although suspended during the relevant decade, was not abolished. Thus, the failure to slow the growth of government employment may have been the result of the very rapid growth in the ranks of those who were until recently eligible for the government employment guarantee, that is those with upper secondary and post-secondary education. Even at sharply lower rates of hiring among these ‘‘graduates,’’ the rapid growth of their ranks could be responsible for the continued expansion of government payrolls. Another reason for the more rapid than expected expansion of government payrolls is the changing behavior of existing civil servants. As government jobs became more difﬁcult to obtain for new entrants, existing female civil servants were increasingly more likely to hold on to their jobs longer rather than leaving at marriage to start a family. This trend is clearly apparent from the age–employment proﬁles shown in Figure 3, where there is a clear shift rightward in the age–employment proﬁle of female government workers. By reducing the rate of exits from government employment, this trend contributed to the faster than expected expansion in the 1990s.

3.3. Effects of the graduate employment guarantee on human capital acquisition and rates of return to schooling Years of promising a public job to anyone who obtains a secondary school certiﬁcate has skewed both public and private education spending in Egypt toward the upper secondary and university levels, at the detriment of basic education. A comparison between Egypt and Turkey is quite instructive in this regard. Both countries have achieved an average of about 6 years of schooling for their population 12 years and older, but the distribution of that schooling over the different levels of education looks very different. As shown in Table 3, Egypt has not only a much higher proportion of its population than Turkey that is still illiterate or literate without formal schooling, but also a much higher proportion that has higher secondary and university education. Turkey, in contrast has achieved low rates of illiteracy by investing heavily in elementary education. Egypt’s educational resources in comparison to Turkey’s are therefore heavily skewed toward the secondary and university levels.

Labor force participation and employment ratios by type of employment and age for women in urban areas

396

Figure 3.

Government Employment

Wage Employment

50

Labor Force Participation Percent

Percent

40 50.0 40.0 30.0 20.0 10.0 0.0

30 20 10 0

0

10 20 30 40 50

0

60 70 80

10 20 30 40 50

Age

60 70 80

Age

Private Sector Wage Employment

30 20 10

Percent

Percent

Percent

Public Enterprise Employment

50 40

40

30 20 10

0 10 20 30 40 50

60 70 80

30 20 10 0

0 0

50 40

0

10 20 30

Age

40

50

60

70 80

Age 1988

1998

Poly.(1988)

0 10 20

30 40 50

60 70 80

Age Poly.(1998)

Note: The denominator for all ratios shown is the population of the relevant age group. The black and white symbols on the ﬁgures indicate mean participation rates by age in 1988 and 1998, respectively. The gray and black solid lines are a ﬁfth-order polynomial ﬁt through these points, for 1988 and 1998 Source: LFSS, 1988 and ELMS, 1998

Ragui Assaad

Total Employment 50

Institutions, Household Decisions, and Economic Growth in Egypt

Table 3.

397

Education of non-institutional civilian population 12 years old and older, 1998 (in percentage) Total

Male

Female

Turkey Egypt Turkey Egypt Turkey Egypt Illiterate 14.3 31.9 6.0 20.9 22.3 42.9 Literate without any diploma 5.0 8.7 5.3 10.7 4.8 6.6 Elementary school 51.7 16.2 53.0 18.3 50.4 14.1 Middle school 11.9 13.0 14.9 14.5 9.0 11.4 General secondary school 9.8 4.6 11.3 5.3 8.3 4.0 Vocational secondary school 2.9 14.7 3.7 16.8 2.0 12.6 Universities and other higher educational institutuions 4.5 11.0 5.9 13.6 3.2 8.3 Total 100.0 100.0 100.0 100.0 100.0 100.0 Note: Illiteracy rates from these studies differ from those offered in UNDP’s Egypt Human Development Report (UNDP 2000), which give higher numbers of illiteracy (36.4 percent for males and 61.2 percent for females in 1998, and 35 percent and 58 percent in 2000 and 2001, respectively). The studies were conducted in different years (1995 for UNDP and 1998 for the ELMS), and used different age levels (12 years and above and 15 years and above, respectively). The variances as well as variations in the size and origins of the sample contribute to the different results. Source: HLFS, 1998 for Turkey and ELMS, 1998 for Egypt.

By making this comparison, I am not suggesting that Egypt must reduce its absolute investment in secondary or tertiary education, nor that Turkey has discovered the best possible equation in the allocation of education resources. However, as the results in these charts demonstrate, Egypt has been allocating a signiﬁcant share of its education resources to the secondary and higher education levels before achieving universal basic education. This misallocation in spending could be redressed by introducing means testing and greater cost recovery at the university level, especially considering the vast difference in per student spending between university students and primary students. Minimization of personal costs and the guarantee of employment have driven the demand for secondary and tertiary education in Egypt to laudable but economically unrealistic levels. The enormous demand for secondary and higher education has biased funding toward these educational levels to the detriment of basic education, while poorly serving the needs of the economy. The evolution of the number and composition of graduates at the secondary and tertiary levels over the past 30 years reveals a fairly erratic process of educational expansion. Table 4 shows the evolution of the number of graduates in various specializations by level of education over the past 40 years according to ofﬁcial government data. The data reveal the rapid growth in the number of of secondary and university education in the 1970s as the public employment guarantee grew more alluring. Most of the expansion in secondary education was in the technical secondary branch, which grew at nearly 10 percent per annum. The equally rapid growth in the number of university graduates was more skewed toward humanities and social sciences because it is easier for universities to absorb students

Number of Graduates Specialization

1973–1974

General and Azhari secondary All secondary General and Azhari preparatory General and Azhari primary

1990/1991

1995/1996

2000–2001

1973/1974 to 1981/1982

1981/1982 to 1990/1991

1990/1991 to 1995/1996

Statistical Statistical Statistical Statistical

Year Year Year Year

Book Book Book Book

of of of of

the the the the

1990/1991 to 2000/2001

86,841 8,340 6,318 5,813 3,699 5,943 30,113 20,502 8,588 10,717 16,921 56,728

108,276 6,675 5,845 5,806 4,460 4,219 27,005 22,378 14,258 14,990 29,645 81,271

126,353 5,938 2,856 5,226 3,869 4,800 22,689 30,694 12,657 26,645 33,668 103,664

248,451 14,539 4,354 10,653 5,289 12,787 47,622 47,586 23,089 44,875 85,279 200,829

10.5 7.2 2.2 5.0 10.9 29.3 7.9 12.5 15.3 17.3 8.2 12.1

2.5 2.5 0.9 0.0 2.1 3.8 1.2 1.0 5.6 3.7 6.2 4.0

3.1 2.3 14.3 2.1 2.8 2.6 3.5 6.3 2.4 11.5 2.5 4.9

13.5 17.9 8.4 14.2 6.3 19.6 14.8 8.8 12.0 10.4 18.6 13.2

8.3 7.8 2.9 6.1 1.7 11.1 5.7 7.5 4.8 11.0 10.6 9.0

6998 4424 2574

16,836 12,838 3,998

35,112 21,911 13,201

22,506 12,364 10,142

30,154 13,370 16,784

11.0 13.3 5.5

8.2 5.9 13.3

8.9 11.4 5.3

5.9 1.6 10.1

1.5 4.9 2.4

82,264 9317 47,517 25,430

179,585 17,918 111,169 50,498

281,912 35,339 114,234 132,339

547,831 62,884 247,321 237,626

509,750 50,084 219,469 240,197

9.8 8.2 10.6 8.6

5.0 7.5 0.3 10.7

13.3 11.5 15.4 11.7

1.4 4.6 2.4 0.2

5.9 3.5 6.5 6.0

124,258 206,522 274,113 399,565

171,627 351,212 431,125 582,534

176,946 458,858 631,811 984,181

310,803 858,634 952,772 1,259,201

377,622 887,372 1,288,994 1,377,455

4.0 6.6 5.7 4.7

0.3 3.0 4.2 5.8

11.3 12.5 8.2 4.9

3.9 0.7 6.0 1.8

7.6 6.6 7.1 3.4

Sources: (1) (2) (3) (4)

1995/1996 to 2000/2001

37,574 4683 5313 3910 1543 570 16,019 7549 2533 2694 8779 21,555

All higher technical institutes Commercial Industrial All technical secondary Agricultural Commercial Industrial

Average Annual Growth Rate (Percent)

Arab Arab Arab Arab

Republic Republic Republic Republic

of of of of

Egypt, Egypt, Egypt, Egypt,

1995–2002 1990–1995 1952–1986 1952–1978

(Cairo, (Cairo, (Cairo, (Cairo,

June 2003) for 1998–1999 and 2000–2001 ﬁgures. July 1996) for 1990–1991 ﬁgures. June 1987) for 1981–1982 ﬁgures. July 1979) for 1973–1974 ﬁgures.

Ragui Assaad

All university Health sciences Agriculture Engineering Sciences Other science-related ﬁelds All sciences Commerce Law Education Other humanities and social sciences All humanities and social sciences

1981–1982

398

Table 4. Number of graduates from the formal schooling system, based on number of students who passed general examinations

Institutions, Household Decisions, and Economic Growth in Egypt

399

in these departments. In the presence of the employment guarantee for technical secondary and university graduates, this explosive growth fueled a tremendous expansion of government employment in that decade, which in turn was made possible by the ﬂow of oil and oil-related revenues into government coffers (see El-Issawy, 1983; Handoussa, 1992). Alarmed by the rapid growth in the number of secondary and university graduates, educational authorities appear to have applied the brakes on educational expansion in the 1980s. The government’s attempt to slow down the increase in the number of university graduates in the 1980s and early 1990s began by controlling the intake into general secondary schools and directing the growing number of degree seekers to technical secondary education. As a result of this policy, the share of enrollment in general secondary schools declined steadily from a high of 60 percent in 1965/66 to a low of 31 percent in 1995/96 (Antoninis, 2002). The decline in general secondary enrollment resulted in an eventual decline in university graduates in the late 1980s and early 1990s (see Figure 2). The decline was particularly sharp in the science-related ﬁelds, which contracted in absolute terms. The ﬁrst half of the 1990s witnessed a tremendous acceleration of technical secondary education in all branches, but continued slow growth of university education. In particular, the industrial branch of technical secondary education grew rapidly in the ﬁrst half of the 1990s, possibly as a result of a belief on the part of education ofﬁcials that industrial skills would be in more demand in the private sector. This pattern was entirely reversed in the second half, as the number of technical secondary graduates shrunk in absolute terms and the number of university graduates exploded at a rate of 13.5 percent per annum. A similarly erratic pattern can be seen in the evolution of the number of graduates from higher technical institutes. These 2-year post-secondary institutions, whose graduates were also covered by public employment guarantee, grew rapidly in the 1970s and 1980s, but then contracted sharply in the ﬁrst half of the 1990s, only to recover somewhat in the second half. The explosive, if somewhat erratic growth of technical secondary education is in large part driven by the promise of government employment and the higher returns of such employment on these educational investments. In Table 5 below, we reproduce results on rates of return to schooling in Egypt in 1988 and 1998 for various levels and types of schooling in the private and public sectors presented in Zhang (2003). Zhang estimated a series of wage equations by sex and sector for data derived from the Egypt Labor Force Sample Survey 1988 and the Egypt Labor Market Survey of 1998. These equations are corrected for selectivity into the non-agricultural wage labor force and into the public and private sectors using the Heckman two-stage method. The most striking result is that rates of return for males to all levels of schooling in the private sector are falling. While there is a positive return for each consecutive year of schooling, with university educations offering the highest return at above 10 percent each year, the overall return in the private

Ragui Assaad

400

Table 5.

Incremental Mincerian returns to education per year of schooling, Egypt (1988, 1998), non-agricultural wage workers

Year

Males Government 1988 1998 Nongovernment 1988 1998 Females Government 1988 1998 Nongovernment 1988 1998

Read and Write vs. Illiterate (percent)

Primary vs. Vocational Illiterate Secondary (percent) vs. Primary (percent)

Post-secondary vs. Vocational Secondary (percent)

University vs. Vocational Secondary (percent)

Years of Schooling (percent)

6.9 9.1

6.4 6.3

9.7 9.5

9.1 12.5

11.7 14.0

9.0 9.2

4.4 4.9

2.8 2.2

5.1 4.3

7.8 7.3

12.4 11.0

5.3 4.4

7.3 10.4

0.1 7.9

24.4 19.7

7.8 17.6

11.6 14.4

10.5 13.3

16.4 4.3

8.9 9.5

8.5 3.3

2.0 15.6

14.7 18.3

9.6 9.4

Source: Zhang (2003).

sector for each year of education decreased from 1988 to 1998 (from 5.3 percent to 4.4 percent). There is not a great appreciation of even primary education in the private sector; the rates of return to primary schooling and vocational secondary schools are quite low. These results suggest that the government’s policy to restrict university education in the 1980s and then reverse that trend in the 1990s initially reduced the surplus of graduates at that level (giving them higher returns) and then returned to a surplus (decreasing the rates of return). However, rates of return to schooling are lower or even negative for higher technical institute graduates whose numbers were also rising rapidly in the 1980s. In all cases except university levels in 1988, rates of return to education in the public sector are higher than in the private sector. In addition, from 1988 to 1998, all rates of return increased for government work, while all private sector rates of return declined. This suggests public sector employment still looms large in the motivations of students to pursue secondary and higher education. The rapid expansion in the number of university graduates since 1995 is likely to have signiﬁcantly increased their supply to the labor market and to have further reduced the premium they are able to receive for their education outside the government sector. The graduate employment guarantee generated an artiﬁcial incentive to achieve the lowest educational threshold to qualify for the guarantee. That minimum threshold is the technical secondary degree, which, despite its name, is

Institutions, Household Decisions, and Economic Growth in Egypt

401

not very vocational or technical in nature. A signiﬁcant proportion of vocational secondary graduates are in fact in the ‘‘commercial’’ branch, which essentially imparts bureaucratic and clerical skills, which are in little demand in the private sector (Antoninis, 2002). With the suspension of the graduate employment guarantee and the relative slowdown of government hiring, there was a serious glut of graduates at that level of education in the late 1990s (Assaad, 2002). The government has since curtailed enrollment in technical secondary schools and expanded it signiﬁcantly at the university level. It remains to be seen what effect this will have on the returns to university education. What is readily apparent, however, is that Egypt’s attempts to control access to various levels of education and steer students toward the government’s institution of choice has not had the desired impact on reducing graduate unemployment rates or increasing rates of return to education outside the government. At the university, higher institute, and secondary level, the issue is not how many students to have at each level, but the quality of the education they are receiving. 3.4. ‘‘Entrapment’’ of graduates in the public sector The public employment guarantee for graduates has skewed demand for higher education while doing little to add to economic growth. An unfortunate consequence of this decades-old policy is that much of Egypt’s educated human capital is trapped in the public sector, particularly in the civil service component of the public sector, thus contributing little to raising productivity in the rest of the economy. The guarantee of employment offers little incentive to seek out skills more marketable in the private sector. In the long-term, the promise of lifetime employment with beneﬁts does not encourage transfer between the sectors. Although this is beginning to change as the growth of the government bureaucracy begins to slow, it will take a while before the stock of educational human capital shifts to the private sector. As shown in Table 6, nearly 60 percent of employed males and 80 percent of employed females with secondary education or above worked in the public sector in 1988, compared to 21 percent of employed males and 2 percent of employed females with a lesser education level.3 Although one would expect the public sector workforce to be more educated, these lopsided numbers suggest that the education system in Egypt is effectively geared toward meeting the demands of the bureaucracy. Although there have been efforts to move away from this dominant pattern in recent years, progress has been slow. The proportion of employed graduates working in the public sector was still 52 percent and 63 percent for males and females, respectively. Young graduates, in particular, were ﬁnding it increasingly difﬁcult to get these public sector jobs.

3

The vast majority (77 percent) of educated workers in the public sector are in the civil service component of that sector.

Ragui Assaad

402

Table 6.

Proportion of employed working in the public sector by education and age LFSS (1988)

Education Males Below secondary

Secondary and above

All

Age

15–34 35–64 All 15–34 35–64 All 15–34 35–64 All

No. of observations Females Below secondary

Secondary and above

Urban

Rural

14.1 41.2 29.8 47.7 72.9 59.2 29.8 52.3 41.7

8.0 22.9 15.8 53.9 88.3 63.1 19.4 29.5 24.2

3245 15-34 35-64 All 15-34 35-64 All 15-34 35-64 All

No. of observations

12.2 9.5 10.7 75.9 91.5 81.1 54.6 47.9 51.8 1157

3000 0.8 0.4 0.6 72.5 92.3 75.8 8.4 2.8 5.9 2173

ELMS (1998) Total

10.1 30.1 20.9 50.3 76.7 60.6 23.8 40.3 32.0 6245 2.4 2.0 2.2 74.9 91.6 79.9 22.2 15.2 19.1 3330

Urban

Rural

11.2 32.1 24.1 33.4 69.4 53.0 24.2 51.2 39.8

8.0 29.4 20.1 32.9 75.1 49.8 18.1 41.4 30.0

3341 4.0 9.7 7.7 54.4 90.1 74.1 38.9 57.2 49.7 1572

1861 1.0 0.8 0.9 34.7 76.6 45.4 9.2 7.9 8.6 1507

Total

9.1 30.4 21.4 33.1 71.5 51.6 20.6 46.0 34.3 5202 1.4 2.5 2.0 44.3 87.0 62.9 16.6 25.0 20.9 3079

Source: LFSS, 1988; ELMS, 1998.

Whereas 50 percent of male graduates in the age group of 15– 34 years was working in public sector in 1988, this proportion was down to 33 percent in 1998. Among young female graduates, the share in the public sector dropped equally fast from 75 to 44 percent. Older educated workers continued to be highly concentrated in the public sector, in both 1988 and 1998, however. For older educated females over 35, public sector work is nearly the only option. Those who are unable to get public sector work simply drop out of the labor market after marriage, since private sector employment is deemed incompatible with the domestic responsibilities of married women in Egypt.

4. Gender norms and barriers to entry for women in the private sector The institutional norms that entrap human capital in public sector employment are all the more limiting in the case of women. In addition to the demand for the security of public sector employment, norms that deﬁne what is appropriate

Institutions, Household Decisions, and Economic Growth in Egypt

403

female employment often prevent women from taking up employment in the private sector. These norms deﬁne appropriate gender roles regarding women’s participation in the public sphere. Women are expected to assume the primary responsibility for the domestic sphere and to prioritize that responsibility over their professional careers outside the home. Women’s safety from male harassment or molestation is often so highly prized as to place severe constraints on how far they can travel to go to work, how late they can remain outside their homes, and what kind of workplace is appropriate. These norms not only shape what women and their families perceive as an acceptable form of involvement in the labor market, but they also determine employers’ perceptions of female workers. For instance, the general tendency for women to stop working at marriage to devote themselves completely to domestic responsibilities discourages employers from investing in their training. Women are therefore often relegated to jobs that require little training and that can tolerate high levels of job turnover. Folbre (1996, p. 128), among others, has described development as involving the gradual transition of economic activities from the household economy to the market economy. Being at the interface of this divide between the household and the market, women see a signiﬁcant change in their roles as development proceeds. As fertility declines and the focus of economic activity shifts away from the family farm and the family enterprise, women have sought greater participation in paid work. Rising educational attainment among women further reinforces these trends, leading to a signiﬁcant association between educational attainment and work outside the home. These powerful forces are partially counteracted by the strong social norms discussed above that are much slower to change. Historically, government policies have had a major role in changing gender norms about women’s employment outside the home. In fact, the graduate employment guarantee mentioned above, by applying equally to male and female graduates, has resulted in large numbers of women joining the public sector workforce. Consequently, employment outside the home is now part of the normal expectations of young women who achieve a secondary level of education. This is clearly apparent when one examines the pattern of female labor force participation by educational attainment, where there is a sudden and large increase at the secondary education level, the point at which graduates become eligible for the employment guarantee (see Assaad, 2002). Besides making paid employment the norm for educated females, the employment guarantee scheme has also helped shape what is deemed female-appropriate employment. Public sector work is strongly preferred by female workers, not only because it pays better than the private sector, but also because it has other important characteristics that make it gender-appropriate.4 These characteristics include generous

4

The government does not discriminate along gender lines on pay, whereas there is a large unexplained wage gap between male and female workers in the private sector, giving females in the public sector a signiﬁcant pay advantage over their private sector counterparts (Assaad, 1997).

404

Ragui Assaad

medical and retirement beneﬁts, relatively short work hours (six hours a day, six days per week is the public sector norm in Egypt), low effort requirement, transportation provided by the employer in many instances, and the presence of a large number of other women in the workplace, providing a measure of sexual safety. With the inevitable slowdown in public sector employment, its role in ‘‘normalizing’’ female employment outside the home might see a signiﬁcant reversal. Although a relatively small proportion of Egyptian women work for pay in the private sector, few remain in that sector after marriage and many would rather queue for years for a public sector job rather than take one in the private sector. 4.1. Female participation in the Egyptian labor market in the 1990s A closer look at the evolution of female participation from 1988 to 1998 reveals that participation among older, married females was increasing in urban areas, but the increase can be attributed entirely to female government employees (who make up the bulk of female wage employees) holding on to their increasingly valuable public sector jobs. As shown in Figure 3, the employment ratio in the government for urban females has shifted noticeably to the right, not only indicating that older women were staying longer on government payrolls, but also that young women had reduced, or at least delayed, access to government employment. Employment in public enterprises, in contrast, was falling in all age groups, but the decline was the largest among younger females. The situation in rural areas (Figure 4) is less clear given the relatively limited role of wage employment among females, but it suggests that government employment continued to increase among prime working-age females there as well. Reduced access to public sector employment for young women was not being counteracted by increased access in the private sector. In fact, although private sector wage employment was primarily the domain of young women in their twenties and early thirties, access to it had not increased from 1988 to 1998. Employment ratios in that type of employment remained virtually unchanged for urban women and declined slightly in rural areas. There seems therefore to be some support for Moghadam’s contention that structural adjustment is crowding women out of the Egyptian labor market by reducing access to public employment without affecting the barriers to entry that women face in the private sector (Moghadam, 1998, pp. 109–112). The problem is not one of general lack of dynamism in the private sector. Wage employment in the private non-agricultural sector was the second-fastest growing segment of the labor market after government. It grew at an average of 4.3 percent per year from 1988 to 1998, a rate that is 72 percent higher than the 2.5 percent overall employment growth rate in that period. However, the growth of employment in that sector was much more rapid for males (4.7 percent per year) than for females (1.6 percent per year), despite the fact that overall female employment growth was more rapid than that of males. Given the almost certain

Figure. 4.

Labor force participation and employment ratios by type of employment and age for women in rural areas Labor Force Participation

Government Employment Wage Employment

80.0

20

20.0

Percent

Percent

10

0.0

0 0

10 20 30 40 50 60 70 80

0 0

Age

10 5

5

1988

10

20

1998

30

40

50

Poly.(1988)

60

70

0

80

Total Employment

80

Private Sector Wage Employment

Percent

20

40 20

15 10

10

20

30 40 50 Age

60 70

80

3 2

0

0 0

4

1

5

0

60 70 80

Public Enterprise Employment

5

60

10 20 30 40 50 Age

Poly.(1998)

Percent

Percent

40.0

15

15

0 1988

10 20 30 40 50 60 70 80 Age 1998

Poly.(1988)

0

10 20 30 40 50 Age

60 70 80

Poly.(1998)

Institutions, Household Decisions, and Economic Growth in Egypt

60.0

Percent

20

Note: The denominator for all ratios shown is the population of the relevant age group. The black and white symbols on the ﬁgures indicate mean participation rates by age in 1988 and 1998, respectively. The gray and black solid lines are a ﬁfth- order polynomial ﬁt through these points for 1988 and 1998 Source: LFSS, 1988 and ELMS, 1998 405

406

Ragui Assaad

prospects of a slowdown in the growth of public sector employment in Egypt, the barriers to female employment in the private sector may lead to continued high unemployment rates for women and a possible reversal in the growth of female labor force participation observed in recent years. Thus, it appears that the growth of the female labor force in Egypt in the foreseeable future will be more constrained by factors related to labor demand and labor market structure than to labor supply. 4.2. Barriers to private sector employment for women Women’s paid employment in the private sector is one area in which the prevalent social norms interact with government policies in powerful ways. The availability of government employment for female graduates has done a lot to attenuate strong social norms that discourage women’s employment outside the home. In 1998, nearly 32 percent of Egyptian women with secondary education or above (the group that is eligible for the public employment guarantee) worked for pay, compared to o3 percent of those who have less than a secondary education. Government employment has therefore ‘‘normalized’’ working for educated women in Egypt. Although most of them work for the public sector, their involvement in private sector wage employment is also signiﬁcantly higher than their less-educated counterparts. What government employment has also done, however, is deﬁne what is a gender-appropriate job in Egypt. If private sector work does not ﬁt into that category, it is virtually closed to women, both because employers do not consider them worthy candidates for such jobs and because the women and their families do not consider it suitable employment. Some have argued that excessive regulation, and in particular female-speciﬁc mandates, such as paid and unpaid maternity leave and childcare provisions, discourage private employers from hiring women (see discussion in Moghadam, 1998, p. 119). This argument appears to be somewhat exaggerated, however, given that the Egyptian government has been very lax about enforcing its own labor laws in recent years. A comparison of survey data from 1988 and 1998 shows that only 15 percent of new jobs obtained by women in the non-agricultural private sector in Egypt were protected by a legal employment contract (Assaad, 2002). One can safely assume that jobs that are not protected by a contract will not provide all the female-speciﬁc beneﬁts that the law stipulates. The main reason Egyptian employers are reluctant to hire women has to do with the widespread perception that women have a low attachment to the workforce. This perception manifests itself in several ways, including a sense that female workers have a high turnover rate, which makes them poor prospects for training, that they have high absenteeism rates, and that they are unwilling to work the long hours that the private sector in Egypt requires. Such low commitment to the workforce is a direct outcome of social norms that make the home women’s primary domain of responsibility. Although this applies to both married and unmarried women, the situation for married women is compounded

Institutions, Household Decisions, and Economic Growth in Egypt

407

to an extent that private sector wage work is seen as totally incompatible with marriage, leading to a virtual certainty of quitting such work at marriage. While I do not have direct evidence on the reluctance of private sector employers to hire female workers, gender wage differentials are fully consistent with such a hypothesis. The much lower wages that female employees receive in the private sector, after correcting for productivity-related characteristics, suggest that low rates of wage employment among females are not simply due to supply constraints (Assaad, 1997). In fact, female labor supplies studies indicate that female participation in paid work is quite responsive to wages (Assaad and El-Hamidi, 2002). Low female wages are most likely due to barriers to entry that force female workers to crowd into limited segments of the labor market, depressing their wages in these segments. These barriers are undoubtedly due to employer perceptions about the suitability, reliability, and productivity of female labor; perceptions that are themselves shaped by the predominant gender norms in the society. Besides employers’ reluctance to hire them, women face other sorts of constraints in the private labor market. Women tend to prefer working in relatively large establishments with many other female workers in them because of the protection against sexual harassment that the presence of other women provides. Unfortunately, the share of such large establishments in the Egyptian private sector is rather small. In 1998, nearly 48 percent of all private non-agricultural wageworkers in Egypt worked in establishments of less than 5 workers, and only 20 percent were in establishments of 30 or more workers. Women are also more geographically constrained than men. Their place of residence is generally determined by either parents or husbands, so that they cannot adjust where they live to where the jobs are (Moghadam, 1998, pp. 118–119). Moreover, they are less able to commute long distances to work because of time constraints and safety concerns. A comparison of commuting times in 1988 and 1998 for men working in the private sector in Egypt shows that men were having to commute signiﬁcantly more to get to private sector jobs. Women’s inability to make such commutes limits them to jobs that are locally available, and therefore is likely to be an important entry barrier to women in the private sector (Assaad and Arntz, 2005). To conclude, gender norms about the primacy of women’s reproductive role and the sexual division of labor constrain women’s geographical mobility. Women tend to be disproportionately concentrated in the public sector, partly because public sector jobs tend to be more compatible with women’s household responsibilities than paid employment in the private sector, and partly because public sector employers are less able to discriminate against women. 5. Conclusions The demographic transition presents Egypt with a window of opportunity to signiﬁcantly raise economic growth over the near future. A number of policy

408

Ragui Assaad

actions will be necessary to capitalize on this opportunity. The ﬁrst set of actions relates to the educational ﬁeld, where the government must at least maintain the share of national resources devoted to education at present levels to allow for a signiﬁcant increase in resources per child. Part of improving the quality of education is to increase its relevance to the needs of an increasingly marketoriented economy. Furthermore, improvements in the quality of education will provide additional incentives for households to increase their own human capital investments. A second set of actions relates to maintaining or increasing labor force participation rates among adults so that the changing age structure actually results in an increase in the economically active population. This appears to be occurring in Egypt, as female participation rates are on the rise (Assaad, 2002). However, the inevitable contraction of the public employment option without a concurrent improvement in opportunities in the private sector could jeopardize this trend. Finally, policies are necessary to ensure a sufﬁciently dynamic economic system so that productivity levels continue to rise, or at least do not fall, as more people join the work force. In assessing the micro-determinants of economic growth in Egypt, I emphasized the way in which household decisions are shaped and constrained by strong institutional forces deriving from government policy and gender norms. In the fertility realm, I discussed how declining fertility levels in recent years can contribute to economic growth if the additional human resources that are being made available by a growing share of the working-age population can be used effectively and if the declining number of children can be used as an opportunity to deepen investments in productive human capital. Both potentialities are present in Egypt, but their realization will depend on developments in the arenas of human capital accumulation and the labor market. Investments in human capital have been strongly shaped by the longstanding, albeit ofﬁcially suspended, policy of guaranteeing public employment for all technical secondary school, post–secondary, and university graduates. This policy has resulted in an overinvestment in the threshold level of education that can make a graduate eligible for the guarantee, namely technical secondary degrees. As a result, there is a large surplus of graduates at that level, leading to negative rates of return on the human capital investment in the private sector and to high unemployment rates among members of this group. The situation is similar for graduates from 2-year post-secondary institutions, but quite different for university graduates. Until the mid-1990s the government had followed a deliberate policy to limit the number admitted to that stage of education by means of controlling entry requirements in the predominantly public university education system. As a result, rates of return to university education have been relatively high and rose signiﬁcantly until the mid-1990s. Since then, however, the government reversed its policy and signiﬁcantly increased the number of university graduates. The labor market impact of this change has not yet been detected in the available data.

Institutions, Household Decisions, and Economic Growth in Egypt

409

Overall, however, Egypt has historically neglected basic education relative to secondary and tertiary education, but has made efforts to close the gap in recent years. The challenge now that demographic pressures on the education system are subsiding is to focus resources on improving the quality of education, both at the basic level as well as for vocational secondary schools. Another consequence of the graduate employment guarantee is the trapping of a large portion of Egypt’s educational human capital in the civil service, where its contribution to economic growth is dubious at best. The employment guarantee scheme has therefore not only had an impact on the sorts of human capital investments that were made, but has also starved the private sector of the human capital it needs. With regard to labor force participation, the employment guarantee, which is applied equally to male and female graduates, has made it the norm for women to work outside the home once they achieve the secondary level of education. As a result, female participation in paid work rises sharply once this level of education is reached. With the slowdown in government hiring in recent years, the expectation of work remains, judging from the high number of female graduates seeking paid employment. However, the opportunities to work are becoming increasingly limited owing to the presence of signiﬁcant barriers to female employment in the private sector. Unless these barriers can be reduced in the near future, the gains achieved from signiﬁcant increases in female participation in the labor market are liable to be reversed.

Acknowledgments The author would like to thank two anonymous referees for their excellent comments and suggestions and Jessica Soto and Loren DeJonge for their able research assistance.

References Antoninis, M. (2002), ‘‘An empirical investigation of the effects of the expansion of technical secondary education on the Egyptian Labor Market. A tale of two data sets’’, forthcoming at the Sixth Annual ERF Conference, January 15–17, 2002, Cairo, Egypt. Assaad, R. (1997), ‘‘The effects of public sector hiring and compensation policies on the Egyptian Labor Market’’, The World Bank Economic Review, Vol. 11(1), pp. 85–118. Assaad, R. (editor), (2002), ‘‘The transformation of the Egyptian Labor Market: 1988–1998’’, The Labor Market in a Reforming Economy, Cairo: American University Press.

410

Ragui Assaad

Assaad, R. and M. Arntz (2005), ‘‘Constrained geographical mobility and gendered labor market outcomes under structural adjustment: evidence from Egypt’’, World Development, Vol. 33(3), 431–454. Assaad, R. and F. El-Hamidi (2002), ‘‘Female labor supply in Egypt: participation and hours of work’’, forthcoming in: I. Sirageldin, editor, Population Challenges in the Middle East and North Africa: Towards the 21st Century, London: I.B. Tauris. Barlow, R. (1994), ‘‘Population growth and economic growth: some more correlations’’, Population and Development Review, Vol. 20(1), pp. 153–165. Becker, G.S. (1991), A Treatise on the Family, second edition, Cambridge, MA: Harvard University Press. Bloom, D. and J.G. Williamson (1998), ‘‘Demographic transition and economic miracles in emerging Asia’’, The World Bank Economic Review, Vol. 12(3), pp. 419–455. CAPMAS (various years), Egyptian Statistical Yearbook, Cairo: Central Agency for Public Mobilization and Statistics. Devarajan, S., V. Swaroop and H.-F. Zou (1996), ‘‘Public expenditures data – 1975–1985 for 69 countries: composition of public expenditure and economic growth’’, Journal of Monetary Economics, Vol. 37(1996), pp. 313–344. ELMS (1998), Egypt Labor Market Survey 1998, Cairo, Egypt: Economic Research Forum. Elbadawy, A., R. Assaad, D. Ahlburg and D. Levison (2004), ‘‘Private and group tutoring in Egypt: where is the gender inequality?’’, Paper presented at the ERF 11th Annual Conference, December 14–16, 2004, Beirut, Lebanon. El-Issawy, I. (1983), ‘‘Labour force, employment and unemployment’’, The Technical Papers of the ILO/UNDP comprehensive employment strategy mission to Egypt, No. 4. International Labour Ofﬁce (ILO), Geneva. El-Zanaty, F. and A. Way (2006), Egypt Demographic and Health Survey 2005, Cairo, Egypt: Ministry of Health and Population, National Population Council, EI-Zanaty and Associates, and ORC Macro. Folbre, N. (1996), ‘‘Engendering economics: new perspectives on women, work, and demographic change’’, in: M. Bruno and B. Pleskovic, editors, Annual World Bank Conference on Development Economics 1995, Washington, DC: The World Bank. Handoussa, H. (1992), ‘‘The burden of public sector employment and remuneration: the case of Egypt’’, in: W. Van Ginneken, editor, Government and its Employees: Case Studies of Developing Countries, Aldershot, Hants, England: Avebury. HLFS (1998), Household Labor Force Survey 1998, Ankara, Turkey: State Institute of Statistics. Kheir-El-Din, H. and T. Moursi (2006), ‘‘Sources of economic growth and technical progress in Egypt: an aggregate perspective’’, chapter 7 in: J. Nugent and H. Pesaran, editors, Explaining Growth in the Middle East, Contribution to Economic Analysis, Amsterdam: Elsevier.

Institutions, Household Decisions, and Economic Growth in Egypt

411

LFSS (1988), October 1988 Round of the Labor Force Sample Survey, Cairo, Egypt: Central Agency for Public Mobilization and Statistics. Miller, P. and N. Hirschhorn (1995), ‘‘The effect of a national control of diarrheal diseases program on mortality: the case of Egypt’’, Social Science And Medicine, Vol. 40(10), pp. S1–S30. Moghadam, V. (1998), Women, Work and Economic Reform in the Middle East and North Africa, Boulder, London: Lynne Rienner Publishers. Panel on Egypt, Committee on Population and Demography (1982), The Estimation of Recent Trends in Fertility and Mortality in Egypt, Committee on Population and Demography, Report No. 9, Washington, DC: National Academy Press. Pissarides, C. and M.A. Ve´ganzone`s-Varoudakis (2007), ‘‘Labor markets and economic growth in the MENA region’’, chapter 5 in: J. Nugent and H. Pesaran, editors, Explaining Growth in the Middle East, Contribution to Economic Analysis, Amsterdam: Elsevier. Richards, A. (1992), ‘‘Higher education in egypt’’, Policy Research Working Paper No. 862. The World Bank. Salehi-Isfahani, D. (2006), ‘‘Microeconomics of growth in MENA- the role of households’’, chapter 6 in: J. Nugent and H. Pesaran, editors, Explaining Growth in the Middle East, Contribution to Economic Analysis, Amsterdam: Elsevier. Schultz, T.P. (editor) (1995), Investment in Women’s Human Capital, Chicago: Chicago University Press. Singerman, D. and B. Ibrahim (2003), ‘‘The cost of marriage in Egypt: a hidden variable in the new Arab demography’’, in: N. Hopkins, editor, The New Arab Family, Cairo Papers in Social Science, Vol. 24 (1–2) pp. 80–116. Tunali, I. (1996). ‘‘Labor market implications of the demographic window of opportunity’’, The Forum, Vol. 3(2), ERF: Cairo. UNDP (2000), Egypt Human Development Report (1998/1999), Cairo: Institute of National Planning and UNDP. UN Population Division (2003), World populations prospects: 2002 revision population database, http://esa.un.org/unpp/index.asp?panel=2. Williamson, J. and T. Yousef (2002), ‘‘Demographic transitions and economic performance in the Middle East and North Africa’’, in: I. Sirageldin, editor, Population Challenges in the Middle East and North Africa: Towards the 21st Century, London: I.B. Tauris, forthcoming. Zhang, H. (2003), ‘‘Sample selection bias and returns to education in Egyptian Labor Market, 1988 and 1998’’, Paper completed for partial fulﬁllment of Masters in Urban Planning degree program requirements, Humphrey Institute of Public Affairs, Minneapolis, MN.

This page intentionally left blank

Notes on the Authors Ragui Assaad is regional director for West Asia and North Africa at the Population Council. He has over 20 years of research experience on gender, poverty, and labor market and is an expert on statistical and econometric modeling using household survey and census data. Before joining the Population Council, Assaad served as professor of planning and public affairs at the Humphrey Institute of Public Affairs at the University of Minnesota. Assaad has a PhD in City and Regional Planning from Cornell University. He also holds a B.Sc. in physics and an M.Sc. in mechanical engineering from Stanford University. Mohamed Z. Bechri is currently senior economist with the Central Bank in Abu Dhabi. He holds a PhD in Economics from the University of Southern California. He taught economics at the University of Tunisia and worked at the African Department of the International Monetary Fund in Washington. His main ﬁeld of interest is Development Economics, New Institutional Economics and Political Economy. Mohamed Abdelbasset Chemingui is a research scientist at the Techno-economics Division of the Kuwaiti Institute for Scientiﬁc Research (KISR). Before joining KISR, he has worked as associate researcher with the University of Montpellier and as research assistant and economist at the OECD Development Center (1998 and 2000). He has a PhD in economics from the University of Montpellier (France) in 2000. He was awarded a ﬁrst place with a medal prize for his study on income distribution in Tunisia under the Second Global Awards Competition organized by the Global Development Network in Rio de Janeiro in December 2001, and the best policy oriented paper in trade topic during the ERF 12th annual conference held in Cairo 2005. His research interest has been in partial and general equilibrium modeling, macroeconomics, international trade, and regional integration, labor market and poverty analysis, growth and productivity changes, and agricultural and environmental economics. He has worked on projects sponsored by World Bank, ESCWA, European Commission, ETF, IDRC, ERF, GDN, IFPRI, UNIDO, API, and other national agencies in the capacity of Project Leader/Principal Investigator/Consultant. Adam B. Elhiraika holds a PhD in Economics from the University of Glasgow (1991) and a Masters degree in Economics from the University of Kent at Canterbury. He taught Economics at Gezira University (Sudan), Fort Hare University (South Africa) the University of Swaziland (Swaziland), and the United Arab Emirates University. He served as Economist at the Islamic Development Bank (Saudi Arabia) before joining the United Nations Economic Commission for Africa as Economic Affairs Ofﬁcer in 2004. His main

414

areas of specialization are macroeconomics and ﬁnance with special interest in Islamic ﬁnance. He has several publications in internationally refereed journals and books. Moataz Mostafa El-Said is a technical assistance advisor with the Poverty and Social Impact Analysis (PSIA) group at the Fiscal Affairs Department of the International Monetary Fund (IMF), where he works on the distributional implications of the IMF program policies in low-income countries. Prior to joining the IMF, he worked at the Trade and Macroeconomics division and the Development Strategy and Governance Division of the International Food Policy Research Institute (IFPRI). He has a PhD in Economics from the George Washington University. Hasan Ersel received his PhD in Economics from the Ankara University in 1971, where he was a faculty member until 1983. He served at the Capital Markets Board as senior research ofﬁcer and at the Central Bank of Turkey as the head of the research department before becoming its vice governor. He worked for the Yapi Kredi Bank as senior vice president and later as a member of its board of directors. He is on the board of trustees of the Turkish Economic Policy Research Foundation (TEPAV) and a research fellow of the ERF. Presently, he teaches at the Sabanci University. Hadi Salehi Esfahani is a professor of Economics at the University of Illinois at Urbana-Champaign. He has also worked for the World Bank as a visiting staff economist and a consultant. He received a B.Sc. in engineering from Tehran University and a PhD in economics from the University of California at Berkeley. His theoretical and empirical research focuses on the role of politics and governance institutions in the formation of ﬁscal, trade, and regulatory policies and their outcomes. He serves as a research fellow of the ERF and as executive secretary for the Middle East Economic Association. Zeki Fattah holds PhD in Economics from Oxford University. He is a consultant in Middle Eastern economies, and advises member governments in the region in economic development. He is an advisor in the Economic Research Forum for the Arab Countries, Iran, and Turkey in Cairo. He directed the United Nations Economic Development Program and the Economic Analysis Program, Sectoral Issues Program, and Science and Technology Program for the Middle East in the Economic and Social Commission for Western Asia (UNESCWA)) from 1981 to 2003. He lectures on economic development in the Institute of Money and Finance at the American University in Beirut. Anas H. Hamed holds a PhD in Economics from Arizona State University (1992). He served as deputy general manager, Khartoum Stock Exchange (Sudan). He taught Economics at Grambling State University, and the United Arab Emirates University. He served as economist at Abu Dhabi Fund for Development (UAE) before joining Social Security Investment Authority (Sudan) as an assistant general commissioner, Research and Information. His main areas of specialization are International Economics and Finance with special interest in Financial Markets and Institutions. He published in internationally refereed journals. Ahmad R. Jalali-Naini is a member of the Department of Economics at the Institute for Management and Planning Studies in Tehran, Iran, where he served as a chairman during 1998–2004. He has taught graduate macroeconomics and monetary theory/ﬁnance.

415

His research interests and works cover economic growth, monetary and exchange rate policies, and business cycles, particularly in the MENA countries. He has also served as the director of Financial Markets in Energy at the International Institute for Energy Studies and the director of Foreign Exchange Studies at the Institute for Monetary and Banking Research, the Central Bank of Iran. He has a PhD in Economics from the University of Kent, a master’s in Public Administration from New York University, and an MA in economics from University of Southern California. He is currently a visiting researcher at the Petroleum Market Analysis Department, OPEC, in Vienna. Magda Kandil is a senior economist at the IMF Institute. Prior to her Fund career, she was professor and department chair at the University of Wisconsin-Milwuakee. Her publications include 66 articles (as author or co-author) in international professional journals, ﬁve chapters in books, six book reviews, nine IMF working papers, and numerous other working papers. In 2000, she was ranked 344th among the top 500 economists in the world based on quality and quantity of published papers during 1994–1998 using BAUWENS. She is a research fellow of the Economic Research Forum for the Arab Countries, Iran, and Turkey. Hanaa Kheir-El-Din is an Egyptian national. She received a B.Com. (Economics) from Cairo University, Egypt and a PhD in Economics from the Massachusetts Institute of Technology (M.I.T), USA. She has been appointed since 1978 as a professor of Economics at the faculty of Economics and Political Science, Cairo University where she held the positions of director of the Center for Economic and Financial Research and Studies (1985–1991) and of the chairperson of the Department of Economics (1990–1996). She has been a member of the Board of the United Nations University (1994–1998). She has acted as a consultant to various national, regional, and international organizations including the Egyptian Ministries of Agriculture, Economy and Foreign Trade, Finance, International Cooperation and Planning, the UN/ESCWA and the World Bank. She is currently a member of the Egyptian Shura Council (Egyptian Senate) and the executive director and director of Research of the Egyptian Center of Economic Studies (ECES). Her areas of research include: trade policy, industrial policy, macro-economic modeling, growth and equity, poverty assessment, and targeting for poverty alleviation. Imed Limam is a Tunisian national holding a PhD degree in Economics from the University of Southern California. He held the position of Economist and then Deputy Director General of the Arab Planning Institute (API) during the period 1992–2005. In September 2005, he joined the Arab Fund for Economic and Social Development as a senior economist. He has a special interest in development economics in general and applied economics, and econometric modeling in particular, areas where he has published regionally and internationally. He was the co-editor of the Journal of Development and Economic Policies since its foundation until 2005. He is also member of numerous international professional associations and advisory committees. Samir Makdisi is a professor of Economics and the director of the Institute of Financial Economics, American University of Beirut and currently member of the board of the Global Development Network; formerly deputy president of the American University of Beirut (1993–1998), Chairman of the board of the Economics Research Forum for the Arab Countries, Iran, and Turkey (1993–2001) and Minister of Economy and Trade,

416

Republic of Lebanon (1992); has been a guest lecturer at major universities in the US, Europe and Asia; the author of several books and numerous scholarly papers in academic and professional journals; holds a PhD in Economics from Columbia University and is the recipient of the American University of Beirut Medal for distinguished service to the university. Tarek Abdelfattah Moursi is a professor at the Department of Economics, Faculty of Economics and Political Science, Cairo University. His primary ﬁelds of interest include economic development, economy-wide planning and poverty targeting with speciﬁc emphasis on the Egyptian economy. Apart from serving with the Egyptian Ministry of Economy and Economic Cooperation and with the American University in Cairo, he has also provided consultation to various national institutions and international organizations including the Egyptian Ministries of Agriculture and of Planning, the (Egyptian) Cabinet Information and Decision Support Centre (IDSC), the Food and Agriculture Organization (FAO) and the Ford Foundation. Sonia Naccache is lecturer of economics at the University of Tunisia. Her research focus is on the Political Economy of Trade and Financial Sector Reforms in developing countries. Jeffrey B. Nugent is a professor of Economics at the University of Southern California. He earlier taught at UCLA, was a research fellow at Yale University, and served as economic affairs ofﬁcer at the United Nations Economoic and Social Commission for Western Asia. He specializes in development economics, and is the author of over 150 journal articles and book chapters and the author or co-author of six books. While he has worked on agriculture, trade, income inequality, foreign investment and macroeconomics, in recent years, much of his work has focused on institutional issues. He is on the editorial boards of several journals, on the boards of directors of several associations and currently serves as president of the Middle East Economic Association. Kıvılcım Metin O¨zcan received her PhD in Economics from Oxford University in 1992. She joined Bilkent University, where she is presently associate professor. She is currently working on sources of growth, on macroeconomic models of inﬂation, budget deficit, and ﬁnancial markets. Her works have been published in journals including Oxford Bulletin of Economics and Statistics, European Journal of Operational Research, Journal of Business and Economics Statistic, Empirical Economics, Applied Financial Economics, Canadian Journal of Development Studies, Multinational Finance Journal, and Substance Use and Misuse. She has also published a book titled The Analysis of Inflation: The Case of Turkey (CMB, 1995). U¨mit O¨zlale received his PhD in Economics from Boston College in 2001. After working in Bilkent University as an assistant professor from 2001 to 2006, he recently joined TOBB Economics and Technology University as an associate professor. In addition, he is afﬁliated with Economic Policy Research Institute and has been serving as a project advisor for the Central Bank of the Republic of Turkey. He is currently working on optimal monetary policy and exchange rate dynamics. He has several publications in journals including Journal of Economic Dynamics and Control, Scottish Journal of Political Economy and Applied Economics.

417

Hashem Pesaran is a professor of Economics at the University of Cambridge and USC, and a professorial fellow of Trinity College, Cambridge. Previously he has been the head of the Economic Research Department of the Central Bank of Iran, the UnderSecretary of the Ministry of Education, Iran, professor of economics at University of California at Los Angeles, and a vice president at the Tudor Investment Corporation. Dr Pesaran is the founding editor of the Journal of Applied Econometrics and has served as a director on the Board of Acorn Investment Trust. He is honorary president of Cambridge Econometrics. He has held visiting positions at Harvard University, UCLA, University of Pennsylvania, and the University of Southern California. He has over 150 journal publications in the various areas of econometrics, empirical ﬁnance and macroeconomics, and the Iranian economy. He is the author of several books and edited volumes, and is a codeveloper of the econometric software package Microfit, published by Oxford University Press. He is a fellow of the Econometric Society, a fellow of British Academy and a recipient of one of the Royal Economic Society Prizes for the best article in the Economic Journal. Christopher A. Pissarides is professor of Economics, London School of Economics. He has written extensively on the economics of labor markets. His book Equilibrium Unemployment Theory is a standard reference in the economics of unemployment. He is a fellow of the British Academy and the Econometric Society and Council member, European Economic Association and the Econometric Society. He is a research fellow of CEPR and IZA, a non-national senior associate, Economic Research Forum for the Arab Countries, Iran and Turkey and a member of the Cyprus Monetary Policy Committee. In 2005 he received, with Dale Mortensen, the IZA Prize in Labor Economics. Djavad Salehi-Isfahani graduated with a PhD in Economics from Harvard University in 1977. He is currently professor of Economics at Virginia Polytechnic Institute and State University in Blacksburg, Virginia. He has taught at the University of Pennsylvania, 1977–1984, and was visiting faculty at the University of Oxford (1991–1992). He is a research fellow and member of the board of trustees of the Economic Research Forum based in Cairo. His research has been in population economics, energy economics, and the economics of the Middle East. His current research is in human resources and the economics of the family in Iran and the Middle East. He has coauthored, with Jacques Cremer, the World Oil Market, 1991, and edited Labor and Human Capital in the Middle East, 2001. His articles have appeared in Economic Journal, Journal of Development Economics, and Economic Development and Cultural Change, among others. C - ag˘rı Sarıkaya, who was born in 1979, took his B.Sc. and M.Sc. degrees from the Economics Department of the Middle East Technical University in 2001 and 2004, respectively. He has been working in the Central Bank of the Republic of Turkey since 2001 and he is currently working as an assistant economist in the Structural Analysis Division of the Research and Monetary Policy Department. His major research interests are the real side of the economy, growth and macroeconometrics. Marie Ange Ve´ganzone`s-Varoudakis is a research fellow at the Centre National de la Recherche Scientiﬁque (CNRS), France, associated with the Centre d’Etudes et de Recherches sur le De´veloppement International (CERDI). She has held assignments with the

418

World Bank, the Organization of Economic Cooperation and Development (OECD), the French Ministry of Economy, and the United Nations. She has published a book on Argentinean economic history, and several articles in academic journals on economic policy, economic growth, governance, foreign direct investment, exchange rate misalignment, productivity and technical efﬁciency, with focus on countries in the Middle East, Latin America, and South Asia.

Subject Index absolute scarcity 374 accommodating monetary policy 120 accumulation 342 of factors (labor and capital), 2 of human capital 140 of physical capital 211 acquisition of skills 154–155 Actual output 255 additive outliers 208 adjustment 153 process 260 Africa 145 Agadir Agreement 8 age–employment proﬁle 395 age structure 174 aggregate demand 121, 257 investment expenditure 204 output 370 production function 199 supply 122 agricultural support programs 110 surplus 163 agriculture 4, 338 Algeria 40, 107, 124, 335–339, 341–356 allocation efﬁciency 110 America 16 animal husbandry 6 anticipated monetary policy 121 appreciation 257 Arab aid 212 Common Market 8 League, 14 Maghreb Union; Egypt 8 oil-producing countries 32 Arab–Israeli conﬂicts 2, 119

Argentina 29.11 51, 107 Asian miracle 46 austerity 361 authoritarianism, 20 autonomy 130 average capital share 47 Bahrain 40 balance of payment 240 of payments crises 220 balanced budget 341 budget law 313 bauxite 15 best growth performers 53 black-market premium 38, 296 blocking reform 20 boom–bust cycles 238, 243, 258 brain drain 11 Brazil 107, 113 bread riots 326 Broad Money 106–107, 117 budget deﬁcit 123, 329 surplus 252 budgetary discipline 252 business climate 53 conﬁdence 244 cycles 256–257, 261 capacity constraints 122 utilization 296 capita labor ratio 211 capital 56 account 253 accumulation 223, 248–249, 282, 346 contribution 48 controls 379

420

Subject Index

efﬁciency 43 expenditures 361 ﬂows 250, 256–257 formation 239 goods imports 287 inefﬁciency 44 inﬂows 238 intensity 211 market development 129 market reform 198 markets 378 movements 256 outﬂows 252 per-worker growth 208 shares 47, 49 stock 142 stock data 46 capital-augmented labor 248 capital-intensive 22, 223 cash crops 269 catalytic role 221 causality 258 Central and South America 18 bank 105, 116, 257, 378 planning 311, 339 centralization 7 centrally planned economy 337 cheap labour 360 child mortality rates 388 services 173 welfare 182 children participation in the labor market 207 Chile 51, 107, 118, 314, 322 chow test 217 civil service 394 wars 18, 56 claims on the private sector 111 Cobb–Douglas production function 46 collectivism 23, 321 collectivization 311 commercial banking 117 commitment 406 commodity-oriented economies 273 communication infrastructure 379 Comoros 5

company law 380 comparative advantage 129, 354 compensation schemes 333 tax 340 competition 20, 127, 345 competitive environment 116 ﬁnancial systems 128 forces 110 markets 200 wages 366 competitiveness indices 9 compulsory schooling program 207 computable general equilibrium (CGE) 4 conditional income convergence 37 conditional mean 167 consolidated budget 367 constant returns to scale 201, 342 consumer goods 339 price index 220 consumption 259 smoothing 189 contractions in the economic activity 254 contribution of physical capital to growth 22 control measures 221 variables 115 convergence 140 dynamics 146 cooperativization 321 copper 15 corporate governance 130 corruption in government 52 cost of intermediation ratio 112 costs of operation 117 Council of Arab Economic Unity 8 countercyclical 239, 259 credible 299 credit ratio 323 credit to the private sector 114 cross-country growth regression 41 heterogeneity 6 regressions 37, 54, 142 crowding-out 215 crowding-out 364 cultural heritage 117 currency crisis 250–251

Subject Index

current account 251, 256, 329 expenditure 361 cyclical behavior 259 cyclical shocks 153 debt crisis 153 debt-service ratio 54, 349 declining birth rates 15 demand curve 120 demographic changes 2, 269 explosion 333 gift 3, 174 transition 174 dependency ratio 174 deposits held abroad 109 deposit rate 114 depreciation 339 deregulation 354 devaluation policy 121 developing countries 145 development challenge 32 plans 110, 284 diminishing marginal returns to capital 202 returns to capital 224 direct ﬁnancing 106 dirham 379 Disability Adjusted Life Expectancy 172 disequilibrium exchange rates 296 disinﬂation 242 disinﬂationary monetary policy 125 distortions 38, 151, 164, 340 diversiﬁed economy 360 division of labor at home 21 Djibouti 5 domestic debt 131, 242 industry 290 savings 210, 365 dual gap models 283 Dutch Disease 165, 298, 370 dynamic adjustment 153 efﬁciency 154 misallocation 164 East Asia 140, 275 countries 47

421

economic activities 104 base 336 conditions 119 costs 255 crisis 340 diversiﬁcation 31, 35, 355, 361 efﬁciency 198, 248 freedom 349 growth 39, 105, 198, 223, 243, 250, 269, 271, 273, 275, 277, 279, 281, 283, 285, 287, 289, 291, 293, 295, 297, 299, 301, 303, 305, 345 management 242 policies 57, 109, 269, 271, 273, 275, 277, 279, 281, 283, 285, 287, 289, 291, 293, 295, 297, 299, 301, 303, 305, 336–337 reconstruction 317, 322 reforms 147 slowdown 212 Economic Research Forum 51 economics of technology 336 education 155, 169 education and growth 51 education-augmented labor 280 educational attainment 149 gap 140 effect of ﬁnancial development 20 effective exchange rate 323 effectiveness of monetary policy 20 efﬁciency channel 350 gains 346 of capital 104 efﬁcient allocation 104, 188 growth 211 Egypt 34, 40, 53, 56, 107, 197–200, 202, 204, 207–208, 210–212, 219–220, 222–224, 385–395, 397–402, 404, 407–409 emerging market economies 238, 251 employment 139, 144 guarantee 393 policies 189 ratios 396 relationship 151 endogenous growth models 277 endowment of cultivable land 2

422

Subject Index

energy price 119 engines of growth 138 enrollment rates 183 enterprise sector 341 equilibrium exchange rates 284 equitable income distribution 239 equity market 25 Eritrea 5 ethnic, religious and language diversity 2 Euro Area 113 Europe, America, Asia and Oceania, MENA 16 European Union 7 excess liquidity 107 exchange controls 353 rate 107, 251, 259 rate regime 379 rate uniﬁcation 291 expansionary monetary policy 123 expatriate workers 366 expenditure-reducing 290 expenditure-switching 290 export and import ratios 219 diversiﬁcation 56 earnings 347 promotion 24 push strategy 356 export-led growth 365 exports of primary products 40 external balance 291 borrowing 290 debt 337, 345 environment 32 ﬁnancing 210, 331, 348, 356 payment crisis 341 resources 244 shocks 19, 35, 37–38, 43, 56, 240, 257 factor accumulation 46, 272, 347 elasticities 280 productivity 46 factor-intensive 268 factor-neutral technical progress 200 factors of production 248–249, 342 family planning 319

female human capital, 25 labor force participation 4, 10, 188 school enrollment 179 fertility 172 declines 21, 388 rates 205 ﬁfth plan 285 ﬁnancial constraints 355 ﬁnancial crises 153, 239, 242, 247 deepening 373 development 20, 104, 119, 371 development indicators 113 globalization 108 institutions 105, 111 instruments 105 intermediaries 104–105, 116, 119 intermediation 187 liberalization 253 markets 105, 253, 353 savings 104 services 9 stability 257 structure 253 system 104 First Gulf War 362 oil boom 23 plan period 240 ﬁscal burden 349 debts 252 deﬁcit/GDP 372 discipline 238, 250 dynamics 252 policies 20, 252 stability 253 sustainability 367 variables 268 Five-Year Comprehensive Plan 198 ﬁxed capital investment 275 ﬂuctuations 120, 198, 336 food processing 308 subsidies 327 forecast errors 350 forecasts 216 foreign aid 209, 215 assets 379

Subject Index

borrowing 240 currency reserves 290 demand 339 direct investments 4 exchange 349 ﬁnancial assistance 219 ﬂows 119 labour 376 labour force 12 reserves 328 trade 363 formal education 185 fragmented democracies 7 free riders 312 trade agreement (FTA) 7 zones 365, 380 free-trade area 330 free-trade zones 365 FTAs with the United States, 7

GDP by sector 363 growth rate decomposition 48 growth rates 243, 362 per capita growth 54, 338 gender contract 173 discrimination 164, 189 Empowerment Measure 171–172 gap 184 income inequality 376 norms 164, 180 gender-appropriate 403 geographic size, 2 global ﬁnancial system 108 governance 15,52, 297 government bonds 252 consumption 54, 219 employment 51 expenditure 367 investment 360 policies 20 purchases 121 revenues 3 securities 113 size 56

423

spending 119–120 stability 52 government-saving 277 grand ratios 272 gross domestic ﬁxed 215, 300 domestic investment 36, 364 growth 61–73, 75–80, 85, 91, 104, 154, 299 accounting 143, 199, 247–248, 337 and productivity 52 capacity 356 differential 46 dynamics 247, 261 experience 2, 22 of its trading partners 38 of the money supply 120 pattern 33 performance 19, 31–32, 35, 56, 113 periods 299 potential 222 sustainability 38 theory 138–140 variability 54 volatility 17 growth-enhancing activities 137, 142 guarantee employment in the government sector 25 Gulf Cooperation Council 8 Gulf States 9 Hansen test 216 Health indicators 182 heavy dependence on oil 32 industry 338 heterogeneity 110 Hicks-neutral technical progress 200 hierarchical union 170 high fertility 164 growth rates 261 inﬂation 115 inﬂation countries 115 population growth 32 rates of accumulation of physical and human capital, 23

424

Subject Index

high-growth 243 high-saver countries 186 high-yielding T-bills 251 household economy 403 human capital 20, 37–38, 40, 50, 56, 137–140, 154, 181, 189, 336, 363 development index 40 hydrocarbons 338 illiteracy ratio 50 imperfectly competitive environment 108 import compression 295 substitution 45, 284 tariffs 290 inappropriate use of macroeconomic policies 24 incentive programs 382 structure 352 income gap 34 inequality 240 per capita in the beginning of each period 54 incremental capital/output ratio 339 India 107 indirect ﬁnancing 106 Indonesia 18 industrial relocation 330 zones 365 industrialization 314 industrialization-based development 338 industrialized countries 47 inefﬁcient management 24 infant mortality 182, 388 inﬂation 38, 43, 49, 107, 119, 238, 243, 261, 326, 329 inﬂation rate 115, 246 targeting 238 inﬂation-induced distortions 221 inﬂationary effect 124 inﬂexibility 185 inﬂexible labor market 186 infrastructure 355, 364 initial conditions 37 enrollment rate in primary school 49 income 38, 49 level of income 42

public offering price 204 public offerings (IPOs) 378 institution enrollments 391 institutional capability 348 development 129 environment 52 factors 3, 348 framework 152 indexes 8 matrix 297 norms 402 reform 4, 381 rigidities 124, 150 stratiﬁcation 171 structures 126, 138–139, 141, 149 institutions 49, 57, 117 instrumental variables 40 integration 2, 32, 54 intellectual rights 380 inter-linkages within the MENA 8 interest 107, 111 groups 20 margin 115 rates 187, 252, 322 intermediate goods 259 intermediate-term deposit 289 intermediation cost 111–112 potential 107 internal and external shocks 31 shocks 57 international competitiveness 31, 56 ﬁnancial system 108 Monetary Fund 238, 308 reserves 119, 123 sanctions. 16 interventions 161 invasion of Iraq 2 investment 32, 35, 41–42, 144, 363 climate 24, 379–380 effects 350 growth 210 in education 21 in infrastructure 40 proﬁle 52 ratios 37–38, 44

Subject Index

inward FDI performance index 13 inward-looking economic development 353 inward-oriented growth 241 strategy 239 Iran 1, 3, 107, 166, 172, 267–272, 274–284, 287, 289–290, 292–293, 296–301 Iran–Iraq war 2, 23, 276 Iranian economy 23 Iraq 3, 5 iron 15 irrigation 165 Islam 15 Islamic fundamentalism 2 laws 171 revolution 285 state 7 jobless growth 245 joint stock companies 380 Jordan 40, 53, 107–108, 124, 154 judicial institutions 20 Korea 51, 107 kurtosis 213 Kuwait 40, 107, 165, 368 labor demand 406 endowments, 9 ﬂows across the countries 9 force growth 143 force in agriculture 163 growth 207 law 169 market 21, 137, 140 market institutions 138, 151 market regulations 155, 166 productivity 22, 153, 343 supply 180, 406 labor-augmenting technology 138 labour market distortions 381 regulation 376 laissez-faire 322 land tenure systems 163

landlocked 6 large role of the state 21 Latin America 47, 50, 140, 153 Latin America and the Caribbean 162 League of Arab States 1 learning-based growth 353 Lebanon 107 legal environment 21, 109 systems 117, 128 lending rate 114 less developed countries 284 level of consumption 372 shifts 208 liberalization 110, 117, 250, 252–253 Libya 5 life expectancy 182, 338 linear programming 342 long-run growth 238 long-term loans 350 low productivity 347 productivity growth 339 low-growth equilibrium 286 low-quality education 154 macroeconomic dynamics 238 Environment Index 9 fundamentals 252 imbalances 340 instability 240, 249, 337 performance 37–38 policies 38, 250 shocks 152 stability 341, 367 stabilization 340 Maghreb 11 Malaysia 18, 107 managerial autonomy 338 training courses 22 manpower allocation scheme 394 manufactures exports 54 manufacturing 147 growth 166 value added 341 marginal costs 260–261 market competition 150 economy 105, 403

425

426

Subject Index

ﬂexibility 152–153 institutions 32, 336 mechanism 128 orientation 242 rigidities 142 segmentation 381 structure 153 market-friendly 285 Mauritius 322 measure of institutions 49 measurement errors 202 Mexico 107, 113 micro decisions 171 levels 347 studies 189 units 160 Middle East 1, 61–62, 90 middle-income countries 108, 117 migrants’ ﬂows 107 migration 153, 269 ﬂows 11 military 7 mineral deposits 6 minimum wages 152 mining 6 misallocation of labor 152 of resources 150 misspeciﬁcation 216 mixed economy 239 mobilization of capital 104 modern economic growth 161, 173, 268 education 24 monarchies 7 monetary authorities 117 expansion 285 growth 115, 120, 123 policy 105, 116, 120, 238, 257, 350 monetization 242, 252 money demand 120 monopolies 308, 341 moral hazard 110 Morocco 40, 53, 56, 107, 170, 173, 313 mortality rates, 3 multiple exchange rate system 286 multiplier effects 279

National Charter 329 National Development Plan 319 national savings 186 nationalization 188, 285 natural economy 268 gas 2, 6 resource abundance 2, 15, 37, 39, 49, 381 negative real interest rate 340 neoclassical economics 160 growth model 351 model 200 production function 142 neutrality of monetary policy 122 new economic policy 214 nominal shocks 278 non-agricultural labor 147 non-bank ﬁnancial intermediaries 129 non-capitalist 286 non-hydrocarbon sectors 342 non-inﬂationary 299 non-neutral effects 124, 126 non-oil and diversiﬁed economies 31–32, 53, 346 Arab countries 32, 247 GDP 269 producing countries 32 revenue 382 non-performing loans 127, 332 non-strategic sectors 295 non-wage beneﬁts 151 family income 178 nontariff barriers 219 normality 213 North Africa 1, 61–62, 90 nuclear family 181 number of schooling years 141 OECD economies 153 ofﬁcial capital inﬂows 15 oil 2 and gas exports 3, 15, 40, 375 booms 9, 173 crisis 153 markets 361 price index 374 price revolution 163

Subject Index

price shock 123 stabilization funds 165 windfall 285 oil-exporting countries 31, 32, 53, 122, 164 oil-importing countries 3, 122 old-age security 182 Oman 34, 53 on-the-job training 22 open economy 287 open-endedness 215 openness 7, 37, 39, 44 operational efﬁciency 104, 111–112 optimal growth 297 Organization of Islamic States, 14 of Petroleum Exporting Countries 375 organizational problems 116 out-migration 12 outliers 208 output gap 23, 250, 254–255, 258–259, 261 growth 126 per worker 352 volatility 37, 40, 278 outward-oriented 241 over-the-counter (OTC) market 378 overbureaucratization 286 overstafﬁng 147 overvalued exchange rate 39, 287, 340 Palestinian territories 5 patriarchal 21, 170 peaks 205 per capita consumption 340–341 per capita GDP growth rates 36 per capita income 115, 337 periodization 205 perpetual inventory method 280 persistent inﬂation 252 Phillipines 107 physical capital 209, 336 planning 239 policy makers 116, 119 mismanagement 3 objective 119 political conﬂicts 25 economy 4, 61, 65–66, 68–69, 91, 95–96 factors 114, 117 instability 35

relations, 14 stability 49 upheavals 285 polygamy 318 population 5, 6 control 23, 319 growth 222, 240, 247 populist policies 311 post-adjustment years 330 post-war period 292 potash 15 potential output 255, 260 PPI inﬂation expectations 244 price control policies 124 deregulation 24, 356 distortions 340 dynamics 254 negotiations 120 signals 290 stability 125, 238 price-ﬁxing 369 primary school enrollment ratio 38 private consumption 245, 260 investment 187, 240, 243, 260 property 321 rate of return 142 sector 7, 187, 400 privatization 8, 127–128, 341 of education 170 procyclical 153, 239, 259 producer groups 323 production 33 capacity 287 efﬁciency 369 productive human capital 189 productivity 342 of capital stock 208 per potential laborer 352 proinvestment laws 210 property rights 4, 128, 285, 348 protection 44, 325 proven oil reserves 2 public administration 163 deﬁcits 251 enterprises 8, 147

427

428

Subject Index

expenditures 167, 294 Institutions Index 9 investment 240, 294, 339, 379 sector 114, 137, 400 sector employment 51, 148–149, 167 sector in non-agricultural employment 149 sector reforms 131 Qatar 40 quality of bureaucracy 52 of education 51 of institutions 50, 52, 349 quantitative trade restrictions 291 quasi-ﬁscal 289 quasi money 114 R&D 22, 140 random-effect panel regression 54 rapid economic growth 336 rate of accumulation 104 of growth of human capital 143 of growth of the capital stock 142 of return to schooling 155 rates of return 400 of returns on investment 32 rational agents 128 raw materials 338 re-exports 361 reaction function 122, 125 real depreciation 290 exchange rate 257 interest rates 253 output growth 120–121 wages 260 recursive residuals 218 red tape 53, 220 redistribution with growth 286 reform 216, 287, 312 Regional Cooperation for Development 8 disparities 338 trade 269 regulation 348 and supervision 130 regulatory environment 105 quality 52

related party lending 110 relative income (PPP) of MENA region countries with respect to US 35 wage system 154 remittances 10, 37, 56, 240 rent-seeking 52, 142, 268 rentier system 299 reserve money 117, 118 requirement ratio 117 residual 124 growth 200 resource allocation function 104 endowment 273 gap 210 mobilization function 104 rents 20 utilization 261 restrictions on capital movement 360 restrictive trade regimes 39 returns on investment 336 rigidities 126 risk premium 251 risk-free asset 113 role of households 21 of the state 54, 166, 348 Saudi Arabia 3, 40, 107 saving effects 268 savings 32, 186–187 rate 114–115, 142 savings-deﬁcit 106 savings-surplus 106 Second Gulf War 373 Plan period 240 sectoral distribution 33, 139, 148 growth rates 275 self-employment 177 services sector 274, 361 share of Government consumption in GDP 42 share of manufactures in exports 42 shock persistence 293 short-run supply curve 120 short-term capital ﬂows 23, 253, 255, 258, 261 Singapore 51, 378 size of government 148, 168

Subject Index

skewness 213 skilled labor 11 social capital 297 heterogeneity, 20 indicators 338 infrastructure 360 modernization 313 norms 21, 25 organization 160 outcomes 171 rate of return 154 returns 161 subsidy 286 socialism 24, 160, 198, 308, 338 sociological 114 sociopolitical instability 56 soft-budget constraints 23 Solow accounting exercise 336 model 351 residuals 281, 345 Somalia 5 sources of demand 121 of growth 32, 200 South Asia 140 South-East Asia 153 special processing zones 312, 322 tax regimes 340 speculative arbitrage 255 stabilization 23, 120, 238, 240, 251, 311 stable economic growth 337 stand-by agreement 242 standard deviation of inﬂation 375 deviation to the absolute mean of the growth rates 16 state 32, 57, 171 enterprises 4, 240, 308, 337 Planning Organization 245 regulation 170 state-owned banks 127, 131 static efﬁciency 147 misallocation 164 stock exchanges 378 strategic sectors 353 structural adjustment 198, 238, 338, 340, 345 barriers 347

breaks 122, 372 change 238, 240, 272 reforms 23, 249, 340–341 shift 123 transformation 272 Sub-Saharan Africa (SSA) 32, 50, 140, 147, 336 subsidies 20, 221 subsistence farming 308 Suez Canal 210 supply response 340–341 supply-side constraints 107, 126 policies 356 surplus extraction 163 sustainability 367 sustainable growth 33, 336, 239 growth path 254, 258, 261 Syria 40, 53, 113 Taiwan 51 targeted investments 356 tariff rates 7 technical controls 330 progress 200, 211, 223, 246 technological changes 347 diffusion 39 dynamism 224 efﬁciency 200 frontier 140 shocks 214 spillover 219 transfer 336 Technology Index 9 telecommunications 332 terms of trade 36, 43–44, 339 of trade variability 36 Thailand 107 time series 202 time-varying parameter 218 total employment 244, 396 Total Factor Productivity Growth 19, 22, 31, 38, 48, 143, 249 total fertility rate 176 tourism revenues 212 trade 4 barriers 45

429

430

Subject Index

credit 108 liberalization 24, 356 openness 56, 198, 219 policies 363 protection indicators 45 ratio 39 reforms 340 shares 54 shocks 165 trading boards 331 partners 371 traditional ﬁnance 108 transaction costs 297 transitory change 208 transport infrastructure 355 trend growth 293 reverting 293 troughs 205 Tunisia 23, 34, 40, 53, 56, 107, 124, 173, 307–310, 312–316, 319–322, 325–326, 328–333, 337 Turkey 1, 23, 40, 53, 56, 107, 173, 238, 240–242, 245–250, 252, 255, 259–261 turnover rate 406 two-track 312 UAE 7, 40, 359–381 unanticipated changes in monetary growth 120 shocks 205 uncertainty 40, 109, 296 under-ﬁve Mortality 388 underutilized workers 147 unemployment 32, 245, 337 rates 149–150, 252 unit of social organization 160 universal banking 129

unproductive assets 188 unskilled labor 11 unsustainable growth path 254 output growth 261 upper middle-income economies 170 urban households 161, 163 urbanization 162, 269, 388 variation in trading partner growth 56 Venezuela 107, 368 vintage 208 virtuous circle 319 volatile inﬂation 247, 250 volatility 14, 19, 35, 56 of growth rates, 18, 22, 40 of oil prices 2 vulnerability to external 37, 43 wage determination 139 differentials 407 inﬂexibility 155, 376 setting 151, 153 water resources 4 weak ﬁnancial sector 24 wealth 268 windfall gains 299 window of opportunity 174 women rights 24, 313 world capitalist system 285 Development Indicators 343 oil prices 56 Trade Organization 380 Yemen 5, 154 Yemen war 209

Fighting Corruption in the Public Sector, Volume 284 (Contributions to Economic Analysis) (Contributions to Economic Analysis) (Contributions to Economic Analysis)

Cognitive Economics: New Trends, Volume 280 (Contributions to Economic Analysis) (Contributions to Economic Analysis) (Contributions to Economic Analysis)

Designing the New European Union, Volume 279 (Contributions to Economic Analysis) (Contributions to Economic Analysis) (Contributions to Economic Analysis)

Explaining Growth in the Middle East, Volume 278 (Contributions to Economic Analysis) (Contributions to Economic Analysis)

Fighting Corruption in the Public Sector, Volume 284 (Contributions to Economic Analysis) (Contributions to Economic Analysis) (Contributions to Economic Analysis)

Cognitive Economics: New Trends, Volume 280 (Contributions to Economic Analysis) (Contributions to Economic Analysis) (Contributions to Economic Analysis)

Designing the New European Union, Volume 279 (Contributions to Economic Analysis) (Contributions to Economic Analysis) (Contributions to Economic Analysis)

New Directions in Macromodelling, Volume 269 (Contributions to Economic Analysis) (Contributions to Economic Analysis)

Tourism, Trade and National Welfare, Volume 265 (Contributions to Economic Analysis) (Contributions to Economic Analysis)

Urban Dynamics and Growth: Advances in Urban Economics, Volume 266 (Contributions to Economic Analysis) (Contributions to Economic Analysis)

The Single Server Queue (Contributions to Economic Analysis)

Contributions to non-standard analysis

Introduction to Economic Analysis

Introduction to Economic Analysis

Regionalisation, Growth, and Economic Integration (Contributions to Economics)

Introduction to Economic Growth

Advances in Psychology Volume 52 - Human Mental Workload (Contributions to Economic Analysis)

Introduction to Economic Growth

European Union and the Euro Revolution, Volume 283 (Contributions to Economic Analysis)

Unfunded Pension Systems: Ageing and Migration, Volume 264 (Contributions to Economic Analysis)

Military Expenditure and Economic Growth in the Middle East

Introduction to Modern Economic Growth

Introduction to Modern Economic Growth

Introduction to Modern Economic Growth

Introduction to Modern Economic Growth

Introduction to Modern Economic Growth

Using Economic Indicators to Improve Investment Analysis

Using Economic Indicators to Improve Investment Analysis

Using economic indicators to improve investment analysis

Selected Contributions in Data Analysis and Classification

Economic Analysis in Talmudic Literature

Mondragon: an economic analysis

Development of Economic Analysis

Development of Economic Analysis

Explaining Growth in the Middle East, Volume 278 (Contributions to Economic Analysis) (Contributions to Economic Analysis)

Fighting Corruption in the Public Sector, Volume 284 (Contributions to Economic Analysis) (Contributions to Economic Analysis) (Contributions to Economic Analysis)

Cognitive Economics: New Trends, Volume 280 (Contributions to Economic Analysis) (Contributions to Economic Analysis) (Contributions to Economic Analysis)

Designing the New European Union, Volume 279 (Contributions to Economic Analysis) (Contributions to Economic Analysis) (Contributions to Economic Analysis)

New Directions in Macromodelling, Volume 269 (Contributions to Economic Analysis) (Contributions to Economic Analysis)

Tourism, Trade and National Welfare, Volume 265 (Contributions to Economic Analysis) (Contributions to Economic Analysis)

Urban Dynamics and Growth: Advances in Urban Economics, Volume 266 (Contributions to Economic Analysis) (Contributions to Economic Analysis)

The Single Server Queue (Contributions to Economic Analysis)

Contributions to non-standard analysis

Introduction to Economic Analysis

Introduction to Economic Analysis

Regionalisation, Growth, and Economic Integration (Contributions to Economics)

Introduction to Economic Growth

Advances in Psychology Volume 52 - Human Mental Workload (Contributions to Economic Analysis)

Introduction to Economic Growth

European Union and the Euro Revolution, Volume 283 (Contributions to Economic Analysis)

Unfunded Pension Systems: Ageing and Migration, Volume 264 (Contributions to Economic Analysis)

Military Expenditure and Economic Growth in the Middle East

Introduction to Modern Economic Growth

Introduction to Modern Economic Growth

Introduction to Modern Economic Growth

Introduction to Modern Economic Growth

Introduction to Modern Economic Growth

Using Economic Indicators to Improve Investment Analysis

Using Economic Indicators to Improve Investment Analysis

Using economic indicators to improve investment analysis

Selected Contributions in Data Analysis and Classification

Economic Analysis in Talmudic Literature

Mondragon: an economic analysis

Development of Economic Analysis

Development of Economic Analysis

Recommend Documents