77229 Jobs for Shared Prosperity Time for Action in the Middle East and North Africa Jobs for Shared Prosperity Time for Action in the Middle East and North Africa Roberta Gatti, Matteo Morgandi, Rebekka Grun, Stefanie Brodmann, Diego Angel-Urdinola, Juan Manuel Moreno, Daniela Marotta, Marc Schiffbauer, and Elizabeth Mata Lorenzo © 2013 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW, Washington DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org Some rights reserved 1 2 3 4 16 15 14 13 This work is a product of the staff of The World Bank with external contributions. Note that The World Bank does not necessarily own each component of the content included in the work. The World Bank therefore does not warrant that the use of the content contained in the work will not infringe on the rights of third parties. The risk of claims resulting from such infringement rests solely with you. The fi ndings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions This work is available under the Creative Commons Attribution 3.0 Unported license (CC BY 3.0) http:// creativecommons.org/licenses/by/3.0. Under the Creative Commons Attribution license, you are free to copy, distribute, transmit, and adapt this work, including for commercial purposes, under the following conditions: Attribution—Please cite the work as follows: World Bank. 2013. Jobs for Shared Prosperity: Time for Action in the Middle East and North Africa. Washington, DC: World Bank. License: Creative Commons Attribution CC BY 3.0 Translations—If you create a translation of this work, please add the following disclaimer along with the attribution: This translation was not created by The World Bank and should not be considered an offi cial World Bank translation. The World Bank shall not be liable for any content or error in this translation. All queries on rights and licenses should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank .org. ISBN (paper): 978-0-8213-9719-0 ISBN (electronic): 978-0-8213-9720-6 DOI: 10.1596/978-0-8213-9719-0 Cover design: Bill Pragluski, Critical Stages. Cover photos: Top left, Dana Smillie / World Bank; all others, Arne Hoel / World Bank. Library of Congress Cataloging-in-Publication Data Gatti, Roberta. Jobs for shared prosperity : time for action in the Middle East and North Africa / Roberta Gatti, Matteo Morgandi, Rebekka Grun, Stefanie Brodmann, Diego Angel-Urdinola, Juan Manuel Moreno, Daniela Marotta, Marc Schiffbauer, and Elizabeth Mata Lorenzo. pages cm Includes bibliographical references. ISBN 978-0-8213-9719-0 (alk. paper) — ISBN 978-0-8213-9720-6 (ebk.) 1. Middle East—Economic conditions—21st century. 2. Africa, North—Economic conditions—21st century. 3. Labor market—Middle East—History—21st century. 4. Labor market—Africa, North— History—21st century. 5. Manpower policy—Middle East. 6. Manpower policy—Africa, North. I. Title. HC415.15.G38 2013 331.10956—dc23 2012047957 Contents Foreword . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xix Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xxi Acknowledgments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xxiii Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xxv Main Messages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xxvii Executive Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xxix Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Labor markets in MENA: Inefficient and inequitable . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Who are the most affected by the low-productivity equilibrium in labor markets? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Where are the barriers? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 Is growth a problem? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 The private sector: A dynamic based on privilege . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 A regulatory environment that reinforces the status quo in the labor market . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 Skill gaps and asymmetric information in job search . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 What policy options? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 The political economy of inclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 Conclusion: Time for action in the Middle East and North Africa . . . . . . . . . . . . . . . . . . . 35 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 v vi CONTENTS PART 1 Labor Markets in the Middle East and North Africa: A Low-Productivity Equilibrium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 1. Inefficient and Inequitable Labor Markets: A Low-Productivity Equilibrium . . . . . . . . . . 41 Main fi ndings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 Labor markets in MENA: Key facts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 MENA labor markets: Inefficient and inequitable . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 2. Women, Youth, and the Working Poor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 Main fi ndings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 Women . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 Youth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 The Working Poor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 PART 2 Where Are the Barriers? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 3. A Long-Term View of MENA’s Economies and Labor Markets . . . . . . . . . . . . . . . . . . . . . 89 Main fi ndings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 A long-term view of growth and economic trends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 Contribution of growth to employment generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 Demographic challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96 An unfi nished reform agenda . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104 Annex 3A Macroeconomics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109 Annex 3B Demographics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 4. Privileges, not Competition: The Dynamics of the Private Sector in the Middle East and North Africa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 Main fi ndings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 MENA’s lack of economic dynamism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 Lack of creative destruction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123 Firm dynamics and employment creation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125 Explaining the lack of economic dynamism in MENA . . . . . . . . . . . . . . . . . . . . . . . . . . . 128 The employment price of privilege . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138 5. MENA’s Institutional Framework: Reinforcing the Status Quo . . . . . . . . . . . . . . . . . . . . 141 Main fi ndings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141 Insiders and outsiders in MENA’s labor markets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141 How does segmentation between insiders and outsiders arise? . . . . . . . . . . . . . . . . . . . . . 142 Why does segmentation persist in MENA? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160 What happens if segmentation persists? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160 The unsustainable status quo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162 CONTENTS vii Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162 6. Unemployable and Unemployed: Skills Gaps and a Meritocracy Deficit in MENA . . . . . 167 Main fi ndings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167 Becoming employable in MENA: A matter of educational quality and relevance . . . . . . . 167 Barriers to employability: Why are there skill gaps? . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175 Public sector as the main client of the educational and training system . . . . . . . . . . . . . . 181 A “logic of selection” over a “logic of learning” . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182 The second transition: From employability to employment. . . . . . . . . . . . . . . . . . . . . . . . 186 Importance of informal channels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187 Employers and informal hiring networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189 A meritocracy deficit? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192 Role of public employment services in job matching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194 Role of active labor market programs in facilitating the transition to work . . . . . . . . . . . 195 The employability challenge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 200 PART 3 Policy Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205 7. Aligning Incentives to Invest, Innovate, and Generate Employment . . . . . . . . . . . . . . . . . 207 Main fi ndings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207 Strategies to remove distortions that repress labor demand . . . . . . . . . . . . . . . . . . . . . . . 208 Strategies for competing and investing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210 Facilitating innovation for high-productivity employment . . . . . . . . . . . . . . . . . . . . . . . . 214 Sustaining low-skilled entrepreneurs and promoting the potential of the highly skilled. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223 Realigning the incentives for employment and productivity . . . . . . . . . . . . . . . . . . . . . . . 230 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231 8. Regulatory Framework of the Labor Market: Overcoming Segmentation . . . . . . . . . . . . 235 Main fi ndings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235 Segmentation in MENA’s labor markets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235 How much do labor regulations matter for job creation compared to private sector regulations? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246 Towards more dynamic labor markets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247 9. Educational and Training Systems: Realigning Incentives for Skills That Matter and Making Employability Count . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249 Main fi ndings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249 Overcoming skills gaps, mismatches, and unmeritocratic hiring practices . . . . . . . . . . . 250 Closing the information and knowledge gaps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251 Replacing the logic of selection with a logic of learning . . . . . . . . . . . . . . . . . . . . . . . . . 257 A renewed partnership between educational and training systems and the private sector. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 262 A role for active labor market policies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268 viii CONTENTS Skills as drivers of success in the labor market . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273 Annex . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 276 Spotlight 1. An Overview of Policy Options and Their Sequencing . . . . . . . . . . . . . . . . . . 279 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281 PART 4 The Road Ahead . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283 10. The Political Economy of Inclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285 Main fi ndings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285 A complex legacy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285 Revisiting the political economy of private sector dynamics . . . . . . . . . . . . . . . . . . . . . . . 287 Organized labor: Partners in and challengers of the status quo . . . . . . . . . . . . . . . . . . . . 289 Challenges and opportunities in the post–Arab Spring . . . . . . . . . . . . . . . . . . . . . . . . . . . 293 Enabling a broad reform process: Examples of successful transitions . . . . . . . . . . . . . . . . 294 Including broad-based change in the political agenda . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295 Building stakeholder buy-in based on shared benefits of change . . . . . . . . . . . . . . . . . . . . 296 Reforming for broad-based employment growth in a post–Arab Spring MENA . . . . . . . 298 Leveraging new social forces for a renewed and inclusive social dialogue . . . . . . . . . . . . 298 Improving access to data and information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299 Generating credible interventions with early measurable gains . . . . . . . . . . . . . . . . . . . . . 302 Time for action in the Middle East and North Africa. . . . . . . . . . . . . . . . . . . . . . . . . . . . 305 Annex Using data and dialogue to tackle a politically difficult reform: Social insurance in Jordan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309 Appendix: Main Data Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313 Boxes 1.1 The dissimilarity index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 2.1 Women in MENA want to work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 3.1 Productivity gains and contribution to growth from early reforms in Tunisia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 3.2 Structural change and employment outcomes in the Arab Republic of Egypt, Jordan, and Morocco . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 3.3 Employment miracles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104 3A.1 Factors contributing to higher total factor productivity . . . . . . . . . . . . . . . . . . . . . . . 114 5.1 The conventions collectives in Tunisia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145 6.1 Perceptions of the main constraints to the education-to-work transition in the West Bank and Gaza . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173 6.2 Private sector involvement in skill development systems in the Arab Republic of Egypt, Jordan, and Tunisia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177 6.3 TVET end-of-year examination examples from the Arab Republic of Egypt . . . . . . . 179 6.4 Transition paths in the Tunisian educational and training system . . . . . . . . . . . . . . . 182 6.5 Employers’ hiring practices in Jordan and the Arab Republic of Egypt . . . . . . . . . . . 190 6.6 Hiring criteria of headhunters and private recruitment fi rms . . . . . . . . . . . . . . . . . . . 190 7.1 Agricultural employment and subsidies in the Syrian Arab Republic . . . . . . . . . . . . . 209 CONTENTS ix 7.2 Liberalizing entry into professions: Accounting in Morocco and Tunisia . . . . . . . . . 210 7.3 Jordan’s pharmaceutical sector and trade reforms . . . . . . . . . . . . . . . . . . . . . . . . . . . 213 7.4 Improving fi nancial access for women . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215 7.5 Liberalization, investment, and job creation: The telecom sector in MENA . . . . . . . 216 7.6 From emerging role models to innovation clusters: The experience of fi rms in Lebanon and Morocco . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217 7.7 The International University of Rabat . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219 7.8 Lebanon’s Berytech Technology Park . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220 7.9 Diaspora-led intermediation of talent and entrepreneurship . . . . . . . . . . . . . . . . . . . 221 7.10 Common characteristics of successful research universities . . . . . . . . . . . . . . . . . . . . 222 7.11 Enhancing university autonomy: The case of Shanghai Jiao Tong University . . . . . . 224 7.12 Entrepreneurship training in Tunisia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227 8.1 Wage reform in the Netherlands, 1980s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237 8.2 Jordan NOW: The New Opportunities for Women pilot program . . . . . . . . . . . . . . 243 8.3 Stakeholders’ perceptions of Jordan’s new social insurance law on opening up employment opportunities for women . . . . . . . . . . . . . . . . . . . . . . . . . . . 244 8.4 Expanding child care options in India . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245 9.1 The Arab Regional Agenda on Improving Education Quality and the Regional Network for Education Research Initiative . . . . . . . . . . . . . . . . . . . . . . . . . 253 9.2 Career counseling and guidance: The example of Sweden . . . . . . . . . . . . . . . . . . . . . 256 9.3 Mechanisms for overcoming the equity challenge in tertiary education . . . . . . . . . . 260 9.4 Don Bosco in Egypt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 260 9.5 Tunisian Association for Communication and Technology . . . . . . . . . . . . . . . . . . . . 264 9.6 Introduction of a labor market information system in the United Arab Emirates. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266 9.7 Wage subsidy programs in Jordan and Turkey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 270 9.8 An employment guarantee scheme to build infrastructure in rural Maharashtra . . . 271 9.9 The impact of infrastructure investments on employment . . . . . . . . . . . . . . . . . . . . . 272 9.10 Supporting early childhood development and home care services in South Africa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273 10.1 Social orders and economic management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286 10.2 Structural reforms under broad-based politics: Turkey’s Justice and Development Party in the 2000s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295 10.3 Lessons from Turkey’s reform in the 1980s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297 10.4 Data access in the Russian Federation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 300 10.5 Labor market observatories in Poland . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301 10.6 Improving statistical capacity through regional peer learning: The MECOVI project in Latin America . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303 10.7 Improving learning through parents’ participation: Strengthening school councils in Morocco . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 305 Figures O.1 Composition of the working-age population in MENA and two other regions, 2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 O.2 Rates of female labor force participation, by region, 2008 . . . . . . . . . . . . . . . . . . . . . . 2 O.3 Youth unemployment rates in the Arab Republic of Egypt and Tunisia by education level, ages 15–29, 2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 O.4 Youth unemployment rates by region, ages 15–24, 2008 . . . . . . . . . . . . . . . . . . . . . . . 3 O.5 Work status of employed individuals in selected MENA economies, 2005–10 . . . . . . . 4 x CONTENTS O.6 Expected wage by years of education in the Arab Republic of Egypt, 2006, and Jordan, 2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 O.7 Levels and trends in regional unemployment in Tunisia, 2005–11 . . . . . . . . . . . . . . . . . 5 O.8 Decomposition of the dissimilarity index by circumstance and effort for men ages 21–34 in the Arab Republic of Egypt, Jordan, and Morocco, 2009 and 2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 O.9 Share of workers with high-paying or protected jobs among the working-age population in the Arab Republic of Egypt, Jordan, and urban Morocco, 2009 and 2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 O.10 Preference for public sector employment among youth ages 15–34 in selected MENA economies, 2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 O.11 Percentage of individuals and households with desirable jobs in the Arab Republic of Egypt (2009) and Jordan (2010) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 O.12 A framework for interpreting the labor market disadvantage . . . . . . . . . . . . . . . . . . . . 9 O.13 Youth NEET rates in selected MENA economies by education level, individuals ages 15–24, 2008–10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 O.14 Unemployment rates in selected MENA economies by education level, individuals ages 15–24, 2005–10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 O.15 Female participation in the labor force in five MENA countries, by education level, various years, 2005–10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 O.16 Women’s decision-making on whether they can work and labor force participation, by marital status and age, Morocco, 2010 . . . . . . . . . . . . . . . . . . . . . . . 11 O.17 Male-female wage gap in selected MENA economies, various years, 2006–10 . . . . . . 12 O.18 Informality rate by quintile of per capita consumption in selected MENA countries, 2011 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 O.19 Probability of being in a poor household by education and geographic location among the employed population in Jordan, 2006–10 . . . . . . . . . . . . . . . . . . . 13 O.20 Education level of unskilled rural workers as a percentage of the working population, 2006–10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 O.21 Five-year moving average of real GDP growth rates in GCC and in Non-GCC MENA countries, 1960–2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 O.22 Total annual employment growth in selected regions of the world, 1998–2008 . . . . . 15 O.23 Growth in employed and working-age population in non-GCC MENA countries, 1998–2009 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 O.24 Changes in the composition of GDP in selected world regions, 1980–83 and 2007–10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 O.25 Sectoral contribution to annual employment growth in a typical MENA country and other selected countries, average 2000s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 O.26 Private sector gross fi xed-capital formation as a percentage of GDP in selected countries, average 2004–09 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 O.27 Structure of foreign direct investment in selected MENA countries and the EU, 2000–07 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 O.28 Average fi rm entry density for selected emerging economies, 2004–09 . . . . . . . . . . . . 19 O.29 Relation between fi rm size and age in Brazil (2009) and Jordan (2006) . . . . . . . . . . . 19 O.30 Labor taxes and contributions as a percentage of base salary in world regions, 2011 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 O.31 Labor taxes and contributions as a percentage of base salary in MENA economies, 2011. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 O.32 Average employment shares in the public sector in MENA and selected comparator economies, averages 2000s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 CONTENTS xi O.33 Percentage of labor force contributing to social security in MENA and other regions, 2000–07 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 O.34 Quality of education as measured by TIMSS and PISA, 2011 and 2009 . . . . . . . . . . . 24 O.35 Share of fi rms identifying inadequately educated workforce as a major constraint to growth, by region and selected countries, 2005–11 . . . . . . . . . . . . . . . . 25 O.36 Mismatches in hard and soft skills of newly hired graduates in selected MENA countries, 2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 O.37 Distribution of university graduates by field of study in MENA economies, 2004–10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 O.38 Perceptions of youth in selected MENA economies of the constraints to getting a job, 2009 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 O.39 World scores on proxies of meritocracy in hiring, 2011 . . . . . . . . . . . . . . . . . . . . . . . . 28 O.40 Toward more and better-quality jobs in MENA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 O.41 Linking short-term measures to medium-term objectives. . . . . . . . . . . . . . . . . . . . . . . 34 1.1 Composition of the working-age population in the Middle East and North Africa, Latin America and the Caribbean, and Eastern Europe and Central Asia, 2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 1.2 Labor force participation rates of the working-age population in selected economies in MENA, by gender, 2009–10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 1.3 Unemployment rates in various world regions, by age group and gender, 2006–11. . . . 43 1.4 Unemployment rates in MENA, by age, 1998–2010 . . . . . . . . . . . . . . . . . . . . . . . . . . 44 1.5 Unemployment rates by education in selected economies in MENA, 1998–2010 . . . . 44 1.6 Percentage of the unemployed by education in selected economies in MENA, 2006–10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 1.7 Informality in MENA and other regions, 2001–07 . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 1.8 Informality rates for selected non-GCC economies in MENA, 2000–07 . . . . . . . . . . 47 1.9 Distribution of work status across regions, 1999–2008 . . . . . . . . . . . . . . . . . . . . . . . . 47 1.10 Distribution of employment status by educational attainment in selected economies in MENA, 2008–10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 1.11 Work status of employed males in the Arab Republic of Egypt, Jordan, and Morocco by household per capita wealth and consumption quintile, 2009–10 . . . . . . 48 1.12 Employment composition by sector in selected economies in MENA, 2005–10 . . . . . 49 1.13 Expected wage by years of education in the Arab Republic of Egypt, 2006, and Jordan, 2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 1.14 Preference for and attainment of public sector employment among youths ages 15–34 in selected economies in MENA, 2005–10 . . . . . . . . . . . . . . . . . . . . . . . . 51 1.15 Male-female wage differentials by sector in the Arab Republic of Egypt and Jordan, 2006 and 2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 1.16 Education and wage differentials among migrants and nationals in the United Arab Emirates, 2011 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 1.17 Levels and trends in regional unemployment in Tunisia, 2005–11 . . . . . . . . . . . . . . . . 55 1.18 Decomposition of the dissimilarity index by circumstance and effort for men ages 21–34 in the Arab Republic of Egypt, Jordan, and Morocco, 2009 and 2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 1.19 Decomposition of the dissimilarity index by circumstance and effort for women ages 21–34 in the Arab Republic of Egypt, Jordan, and Morocco, 2009 and 2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 1.20 Share of workers with high-paying or protected jobs among the working-age population in the Arab Republic of Egypt, Jordan, and urban Morocco, 2009 and 2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 xii CONTENTS 1.21 Percentage of individuals and households with “desirable” jobs in the Arab Republic of Egypt and Jordan, 2009 and 2010 . . . . . . . . . . . . . . . . . . . . . . . . . 59 2.1 Female participation in the labor force in three world regions by age, 1990–2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 2.2 Unemployment rates by gender in selected economies in MENA, 2006–10 . . . . . . . . . 66 2.3 Educational specializations in countries in MENA, 2010 . . . . . . . . . . . . . . . . . . . . . . 67 2.4 Male-female wage gap in selected economies in MENA, various years, 2006–10 . . . . 68 2.5 Female participation in the labor force in selected countries in MENA by educational attainments, 2006–10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 2.6 Main reason given by females ages 15–29 in Morocco for not wanting to work, 2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 2.7 Female participation in the labor force, by marital status and number of infants in the household, in the Arab Republic of Egypt, Iraq, Jordan, and Tunisia, 2006–10 . . . .71 2.8 Women’s decision making and labor force participation by marital status and age in Morocco, 2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 2.9 Female participation in the labor force, by employment status of the household head and spouse, in the Arab Republic of Egypt, Iraq, Jordan, urban Morocco, and Tunisia, 2006–10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 2.10 Framework for interpreting the labor market disadvantage . . . . . . . . . . . . . . . . . . . . . 75 2.11 NEET and unemployment rates in selected economies in MENA for individuals ages 15–24 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 2.12 Labor market status of NEET youth in selected economies in MENA, by gender, 2005–10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 2.13 Gap between school and work for young men in the Arab Republic of Egypt, Jordan, Tunisia, and the West Bank and Gaza, 2006–10 . . . . . . . . . . . . . . . . . . . . . . . 77 2.14 Prevalence of refusing a job in selected economies in MENA, by educational level, 2009 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78 2.15 Reason for refusing a job in resource-rich and resource-poor economies in MENA, by education and gender, 2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 2.16 Average self-reported reservation wages for males and females ages 15–29 in Morocco, by educational level, 2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 2.17 Relation of reservation wages to actual wages by gender, males and females ages 15–29 in Morocco, 2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 2.18 Male and female age cohorts by labor force status in the Arab Republic of Egypt, 2006 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 2.19 Percentage of youth ages 15–34 that prefer private sector jobs and gap between youth and adult preferences for private sector employment in selected economies in MENA, 2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 2.20 Share of workers in the Arab Republic of Egypt, Jordan, and Morocco who live in poor households, by urban or rural location, 2009–10 . . . . . . . . . . . . . . . 81 2.21 Education level of unskilled rural workers as a percentage of the working population in selected economies in MENA, 2006–10 . . . . . . . . . . . . . . . . . . . . . . . . 82 2.22 Prevalence of nonwage employment among rural unskilled workers in selected economies in MENA, 2006–10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 2.23 Percentage of those entering entrepreneurship in selected economies in MENA because of no suitable job in the wage labor market, by educational level, 2011 . . . . . 83 3.1 Five-year moving average of real GDP growth rates in GCC and non-GCC MENA countries, 1960–2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 3.2 Growth volatility in MENA and other regions, 1970–2008 . . . . . . . . . . . . . . . . . . . . 91 B3.1.1 Contribution of total factor productivity to growth in Tunisia, 1961–2008 . . . . . . . . 93 CONTENTS xiii 3.3 Decomposition of growth in GDP adjusted for human capital in selected countries in MENA, 1990s and 2000s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94 3.4 Changes in the composition of GDP in selected world regions, 1980–83 and 2007–10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94 3.5 Changes in the composition of GDP, excluding mining, in GCC Countries, 1980–83 to 2007–10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 3.6 Unemployment rates in selected world regions, 1980s–2000s . . . . . . . . . . . . . . . . . . . 96 3.7 Employment creation in MENA by sector, 1998–2008 . . . . . . . . . . . . . . . . . . . . . . . . 96 3.8 Sectoral contribution to annual employment growth in a typical country in MENA and in Brazil, Indonesia, and Malaysia, average 2000s . . . . . . . . . . . . . . . . . . 97 B3.2.1 Structural change in Morocco, 1999–2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 B3.2.2 Structural change in the Arab Republic of Egypt, 2000–07 . . . . . . . . . . . . . . . . . . . . 99 B3.2.3 Structural change in Jordan, 2000–08. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 3.9 Total annual employment growth in selected regions of the world, 1998–2008. . . . .100 3.10 Growth in employed and working-age population in non-GCC countries in MENA, 1998–2009 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 3.11 Demographic trends in MENA, 1950–2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 3.12 Difference between GDP growth and GDP per capita growth, Arab Republic of Egypt and Morocco, 1960–2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102 3.13 Labor market developments in the Arab Republic of Egypt, 1970–2010 . . . . . . . . . . 102 3.14 Median age and dependency ratio in selected subregions in MENA, 1960–2050 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103 B3.3.1 Incidence of “employment miracles” by region, 1980–2008 . . . . . . . . . . . . . . . . . . . 104 B3.3.2 Predictors of the incidence of “employment miracles” . . . . . . . . . . . . . . . . . . . . . . . . 105 3.15 Trade restrictions in selected countries in MENA, 2009 . . . . . . . . . . . . . . . . . . . . . . 106 3.16 Structure of foreign direct investment in selected countries in MENA, in China, and in the European Union, 2000–07 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 3.17 Private sector gross fi xed-capital formation as a percentage of GDP in selected countries and regions, average 2004–09 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108 3A.1 Intensive and extensive margins of exports, excluding oil products, in various world regions, 1997–2007 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112 3A.2 Main trends in selected macroeconomic indicators for MENA since the 1990s . . . . 113 4.1 Evolution of the productivity content of exports in selected regions and countries in Asia and MENA, 1980–2009. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122 4.2 Knowledge economy innovation score and GDP per capita for selected world regions and countries in MENA, 2009 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122 4.3 Average entry density for selected emerging economies, 2004–09 . . . . . . . . . . . . . . . 123 4.4 Firm turnover and sector growth in Moroccan manufacturing, 1986–2001 . . . . . . . 124 4.5 Relationship between fi rm size and age in Brazil, 2009, and Jordan, 2006 . . . . . . . . 125 4.6 Relationship between the contribution of net entry and incumbents to productivity growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125 4.7 Firm turnover and productivity growth in 24 industrial and developing countries, various years, 1979–2000 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126 4.8 Distribution of fi rms in Tunisia by size and share of employment, 1997–2009 . . . . . 126 4.9 Labor productivity, wages, and fi rm size in Moroccan manufacturing, 1996–2006 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127 4.10 Employment growth in selected countries in MENA, 2007–10 . . . . . . . . . . . . . . . . . 128 4.11 Energy use per US$1,000 of GDP in selected countries and regions, 2010 . . . . . . . . 129 4.12 Capital intensity in three manufacturing sectors in the Arab Republic of Egypt and Turkey, 2004–08 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130 xiv CONTENTS 4.13 Energy use of fi rms in the Arab Republic of Egypt by fi rm age and ownership, 2004, 2007, and 2008 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130 4.14 Doing Business rankings for economies in MENA and for selected dynamic emerging economies, 2012. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131 4.15 Share of credit-constrained fi rms in selected countries in MENA, various years, 2004–08 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132 4.16 Foreign ownership of banks in selected countries in MENA, 2006 . . . . . . . . . . . . . 133 4.17 Percentage of nonperforming loans in banks in OECD countries and in various world regions, 2007 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133 4.18 Percentage of survey respondents in selected economies in MENA disagreeing that interpretations of regulations are consistent and predictable, 2005–08 . . . . . . . 134 5.1 Minimum wage as a share of value added per worker in selected countries in MENA and in selected comparator countries, 2010 . . . . . . . . . . . . . . . . . . . . . . . . . 143 5.2 Hourly wage distribution and minimum wage in the Arab Republic of Egypt and Jordan, 2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144 5.3 Labor taxes and contributions to social insurance as a percentage of the base salary in world regions and the OECD, 2011. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146 5.4 Labor taxes and contributions to social insurance as a percentage of the base salary in selected economies in MENA, 2011 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147 5.5 Wages and job protection by employment sector in Tunisia, 2011 . . . . . . . . . . . . . . . 147 5.6 Cost of redundancy in selected regions in the world and in selected economies in MENA, 2011 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149 5.7 Types of work contracts in all sectors in the Arab Republic of Egypt, Jordan, Tunisia, and the West Bank and Gaza for the last available year, 2006–10 . . . . . . . . 151 5.8 Duration of unemployment for male and female fi rst-time job seekers in selected economies in MENA, 2006–11 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152 5.9 Percentage of the labor force contributing to social security in MENA and other regions of the world, 2000–07. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153 5.10 Months of contributory and noncontributory spells of social insurance for men and women in Chile (2008) and Jordan (2010) . . . . . . . . . . . . . . . . . . . . . . 154 5.11 Life-cycle earning profiles according to sector of employment and the generality of sector-specific pension benefits in Morocco . . . . . . . . . . . . . . . . . . . . . 155 5.12 Age-specific composition of the labor force across pension system contributors, employed noncontributors, and the unemployed in Jordan, 2010 . . . . . . . . . . . . . . . 157 5.13 Percentage of those who have “voiced” an opinion to a public official in the past month in selected world economies, 2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161 6.1 Educational attainment in secondary and tertiary education in MENA, 1991–2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169 6.2 Gross enrollment in secondary and tertiary education by gender in selected world regions, 2009 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170 6.3 Gross enrollment in tertiary education in selected economies in MENA by GDP per capita, 2005–09 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170 6.4 Selected economies in MENA, the region as a whole, and the OECD benchmarked against the international median in mathematics, 2007 and 2011. . . . 171 6.5 Quality of math education in selected countries in MENA as measured by TIMSS and PISA, 2011 and 2009 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172 6.6 Dropout rates by highest level of education attained in the Arab Republic of Egypt, 2009 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172 B6.1.1 Perceptions of the main constraints in transitioning from education to work in the West Bank and Gaza . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173 CONTENTS xv 6.7 Share of fi rms identifying inadequately educated workforce as a major constraint to growth, by region and selected countries, 2005–11 . . . . . . . . . . . . . . . 174 6.8 Most problematic factors for doing business in the Arab world . . . . . . . . . . . . . . . . . 175 6.9 Mismatch in hard and soft skills of newly hired graduates in selected countries in MENA, 2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175 6.10 Proportion of graduates employed below their level of qualification, by field of specialization, in Tunisia, 2005 and 2007 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176 6.11 Distribution of university graduates by field of study in selected economies in MENA, 2004–10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176 6.12 Incidence of job training by selected economies’ GDP . . . . . . . . . . . . . . . . . . . . . . . . 180 B6.4.1 Transition paths in the Tunisian educational and training system . . . . . . . . . . . . . . . 183 6.13 Enrollments in technical and vocational training in MENA and worldwide, 1999, 2000, and 2009 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185 6.14 Percentage of the employed who found their job through friends or relatives in the private sector in selected countries in MENA, 2005–10 . . . . . . . . . . . . . . . . . 187 6.15 Percentage of the unemployed using informal networks in their job search in selected economies in MENA, 2005–10. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188 6.16 Percentage of the unemployed using friends or relatives in their job search in selected economies in MENA, by level of education, 2005–10 . . . . . . . . . . . . . . . 188 6.17 Ways university graduates found public and private sector jobs in Tunisia, 2005 . . . 188 6.18 Importance of personal networks in fi nding a job in Jordan, by fi rm size, 2010 . . . . 189 B6.6.1 Signals important for hiring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191 6.19 Perceptions of youth about constraints to getting a job in selected economies in MENA, 2009 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192 6.20 World scores on proxies of meritocracy in hiring, 2011 . . . . . . . . . . . . . . . . . . . . . . . 193 6.21 Employer-reported skill shortages versus reliance on professional recruitment in selected countries in MENA and in countries outside the region, 2006–11 . . . . . 193 6.22 Distribution of active labor market programs by program type, 2008 and 2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196 6.23 Active labor market program targeting as a percentage of all programs, 2009 . . . . . 197 B7.2.1 OECD restrictiveness index for accounting services in selected countries, 2008 . . . . 211 B7.5.1 Mobile and broadband growth in MENA, 2003–11 . . . . . . . . . . . . . . . . . . . . . . . . . 216 B7.10.1 Characteristics of a world-class university . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223 7.1 Share of employed individuals whose household head is an entrepreneur, by work status, selected economies in MENA, 2006–10 . . . . . . . . . . . . . . . . . . . . . . . . 225 7.2 Main constraints to starting a business in MENA, by level of education, 2011 . . . . . 226 7.3 Transition matrix of self-employed and employers ages 25–46 in the Arab Republic of Egypt and Jordan, by work status, 1998–2006 and 1999–2010 . . . . . . . 228 9.1 Linking the barriers to the transition from education to work to long-term policies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 252 B9.9.1 Estimated jobs creation through infrastructure investment per billion U.S. dollars of investment in MENA, 2009 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272 B9.9.2 Cost of creating a direct job in road and bridge construction relative to other sectors in the Arab Republic of Egypt, 2009 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272 10.1 Civic freedom indicators for world regions and OECD countries, various years . . . 291 10.2 Number of publications on labor markets with data from six world regions in the EconLit Directory, 2010. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292 10.3 Growth in GDP forecast for non-GCC countries in MENA, 2000–13 . . . . . . . . . . . 294 10.4 Lessons from successful transitions: Steps in enabling a broad reform process . . . . . 298 10.5 Short-term policies and reform objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299 xvi CONTENTS Tables O.1 Employment transition matrix in the Arab Republic of Egypt, 1998 and 2006 . . . . . . . 4 O.2 Heterogeneity in MENA economies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 1.1 Composition of urban employment by industry in the Arab Republic of Egypt, Iraq, Jordan, and Tunisia, 2006–10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 1.2 Employment transition matrix in the Arab Republic of Egypt, 1998 and 2006 . . . . . . 53 1.3 Employment transition matrix in Jordan, 1999 and 2010 . . . . . . . . . . . . . . . . . . . . . . 54 1.4 Monthly employment transitions of the working-age population in Lebanon, 2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 2.1 Percentage increase in female labor force participation in selected economies in MENA by age group, 1990–2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 2.2 Basic unemployment profi le of women ages 15–64 for the Arab Republic of Egypt, Jordan, Morocco, and Tunisia, 2009–10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 2.3 Job search methods used by young people in the Arab Republic of Egypt, 2009 . . . . . 70 2.4 Reasons for being out of the labor force among working-age individuals in the Arab Republic of Egypt, Iraq, urban Morocco, and Tunisia, 2006–10 . . . . . . . 70 2.5 Summary of legal restrictions on women in selected economies in MENA, 2012 . . . . 74 2.6 Ratio of unemployment rates of tertiary educated and secondary educated in several economies in MENA, 2005–10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78 3.1 Heterogeneity in economies in MENA. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 3.2 Decomposition of growth in MENA, 1960s–90s. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 3A.1 Decomposition of average GDP growth rates in selected countries in MENA, 1960–2000, by decade . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109 3A.2 Economic growth and its volatility, 1982–2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110 3A.3 Policy indicators affecting trade in selected countries in MENA . . . . . . . . . . . . . . . 111 B3A.1.1 Correlates of total factor productivity growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114 3B.1 Population increase, net migration, and rate of population change in selected economies in MENA, mid-1980s–mid-2000s . . . . . . . . . . . . . . . . . . . . . . . 115 3B.2 Demographic variables defi ned . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116 3B.3 Reproduction rates and average annual population growth in selected economies in MENA, mid-1980s–mid-2000s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117 4.1 Mobility among size categories for fi rms in Morocco and Tunisia, 1996–2010 . . . . . 124 4.2 Averages and dispersion of fi rms’ waiting days for regulatory services in selected developing countries, various years, 2004–08 . . . . . . . . . . . . . . . . . . . . . . . 135 4.3 Relationship between delays in policy implementation and demands for bribes in MENA, various years, 2004–08 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135 4.4 Effect of policy implementation uncertainty on competition and fi rm growth in Jordan and the Arab Republic of Egypt, 2006–07. . . . . . . . . . . . . . . . . . . . . . . . . 136 5.1 Share of employment by sector in selected economies in MENA for the latest available year, 2005–10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148 5.2 Standard open-ended contract packages in the public and the private sectors in selected countries in MENA, 2011 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149 5.3 Job turnover in the past year as a share of those currently working in the private sector in selected economies in MENA, 2006–10 . . . . . . . . . . . . . . . . . . 151 5.4 Pension design parameters of mandatory formal sector schemes in MENA, 2009–11 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158 5.5 Unemployment benefit programs in MENA, 2011 . . . . . . . . . . . . . . . . . . . . . . . . . . . 159 6.1 Public employment services in selected countries in MENA, 2009 . . . . . . . . . . . . . . 194 CONTENTS xvii 6.2 Performance of public employment services in selected countries in MENA, 2009 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195 7.1 Credit registry infrastructure in MENA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214 7.2 Programs promoting entrepreneurship in MENA . . . . . . . . . . . . . . . . . . . . . . . . . . . 227 7.3 Programs for small-scale entrepreneurs in various developing countries . . . . . . . . . . 229 9A.1 Selected short- and medium- to long-term educational policies and measures for MENA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274 S1.1 Policy options by context . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279 S1.2 Overview of constraints by country and policy area . . . . . . . . . . . . . . . . . . . . . . . . . 280 10.1 Historical structure of unions in selected countries in MENA . . . . . . . . . . . . . . . . . 290 10.2 Access to and quality of survey data in MENA . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292 10.3 Extraordinary policy measures implemented during the Arab Spring in MENA, 2011–12 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294 B10.5.1 Main differences between the two types of labor market observatory in Poland . . . . 301 Foreword A job has always meant more than a sal- educational system to government policy ary. As a basic form of social engage- and the current governance of the private ment, it can be a critical source of sector. The main message is that the cur- self-fulfillment and self-worth. By that same rent regulatory environment, by protect- token, unemployment and underemploy- ing certain markets and a few privileged ment can have a significant psychological insiders, stands in the way of a dynamic as well as economic impact. They can be a private sector and the more and better jobs source of deep frustration and humiliation. it could provide. The rules need to change The Arab Spring, and its call for jobs and to ease the entry of new fi rms, supported dignity, made this connection explicit. The by better access to credit. This adjustment revolutions were also a stark lesson in how would increase competition and serve as a pervasive unemployment and underemploy- vital catalyst for the innovation and invest- ment can fuel instability. Along with the ment that ultimately lead to more demand demand for more political inclusion, young for labor. The current incentives that affect people in particular took to the streets out decisions about where to work and when to of frustration with the lack of opportunities hire also need to be addressed. This effort to put their skills and talents to productive would include phasing out the fuel subsi- use. Identifying and lowering the social and dies that encourage investment in machines economic barriers that have idled large seg- over the hiring of workers and diminishing ments of the population continue to be one of the pull of the public sector to allow labor the most significant policy challenges in the to flow to where it would be most produc- Middle East and North Africa region. tive. Easing employment regulations, cou- This report focuses on jobs as the key pled with a strong and well-targeted social to understanding the many layers of exclu- safety net to protect the unemployed, would sion that have produced the world’s high- facilitate both hiring and mobility. An edu- est youth unemployment rate and left three cational system more in tune with the needs out of every four working-age women out- of the private sector would also be another side the labor force. Each chapter unpacks vital source of growth and jobs, equipping one of the multiple factors that affect labor students with the skills and knowledge markets in the region, ranging from the required by innovative businesses. xix xx FOREWORD These conclusions form the basis of a com- credibility of the reform process with short- prehensive road map for change. The report term interventions, such as youth-targeted identifies the specific barriers in the way of employment programs and investments in inclusive growth and private sector–led job essential infrastructure that address immedi- creation and offers a series of policy options ate needs and produce visible results. for overcoming them. It does not stop there, The Arab Spring revealed the true extent however. In recognition of the urgency of the of the many challenges the region faces, but situation, and of the difficulty in achieving it also offered immense opportunities for the necessary consensus to effect change, breaking with the past and adopting a devel- the report proposes a series of steps to ini- opment model that benefits all its citizens. tiate the process. It recommends embracing This report offers a formula for seizing those the new spirit of openness to engage a wide opportunities. cross-section of society. The goal would be to establish a common understanding of the Inger Andersen nature of the problem and a shared commit- Vice President ment to a program for solving it. The report Middle East and North Africa Region further suggests ways of bolstering the The World Bank Preface I n the aftermath of the Arab Spring, when identifying the distortions and incentives thousands of young women and men that the many actors—fi rms, governments, fought for the opportunity to realize workers, students, education, and training their aspirations and potential, the question systems—currently face, and which ulti- of jobs continues to be crucial in the Middle mately determine the equilibrium in labor East and North Africa region. This report markets. uses jobs as a lens to weave together the Second, the report proposes a medium- complex dynamics of employment creation, term roadmap of policy options that could skills supply, and the institutional environ- promote the robust and inclusive growth ment of labor markets. Consistent with the needed to tackle the structural employment framework of the forthcoming World Devel- challenge for the region. opment Report 2013: Jobs (of which this Third, the report aims to inform and open report is the regional companion), this work up a platform for debate on jobs among a goes beyond the traditional links between broad set of stakeholders, with the ultimate jobs, productivity, and living standards to goal of contributing to reach a shared view include an understanding of how jobs matter of the employment challenges and the reform for individual dignity and expectations—an path ahead. aspect that was clearly central to the Arab This work builds on a large body of Spring. Just as important, this report com- existing literature, among which are a plements the economic perspective with an number of World Bank publications across analysis of the political economy equilib- various disciplines, including, but not rium, with a view to identifying mechanisms limited to, the regional education flag- that would trigger a reform process. ship, The Road Not Travelled: Education As such, the report has three objectives: Reform in the Middle East and North First, it seeks to provide an in-depth Africa (World Bank 2008); the regional characterization of the dynamics of labor private sector flagship, From Privilege markets in the Middle East and North to Competition: Unlocking Private-Led Africa and to analyze the barriers to the Growth in the Middle East and North creation of more and better jobs. It does Africa (World Bank 2009); the regional so by taking a cross-sectoral approach and fi nancial sector fl agship, Financial Access xxi xxii PREFACE and Stability: A Road Map for the Mid- about institutional structures eventually shaped dle East and North Africa (World Bank the geographic focus of the report. 2011a); the regional companion report of the World Development Report 2012 on gender, Opening Doors: Gender Equal- ity in the Middle East and North Africa References (World Bank 2013); the regional reports Angel-Urdinola, Diego, Arvo Kuddo, and Amino Striving for Better Jobs: The Challenge Semlali. 2013. Building Effective Employment of Informality in the Middle East and Services for Unemployed Youth in the Middle North Africa (Gatti et al. 2012) and “Pub- East and North Africa . Washington, DC: lic Employment Services in the Middle World Bank. Gatti, Roberta, Diego Angel-Urdinola, Joana East and North Africa” (Angel-Urdinola, Silva, and Andras Bodor. 2012. Striving for Kuddo, and Semlali 2013); and various Better Jobs: The Challenge of Informality in Middle East and North Africa Economic the Middle East and North Africa. Directions Developments and Prospects Reports (such in Development Series. Washington, DC: as World Bank 2011b). World Bank. Jobs in the Middle East and North Africa World Bank. 2008. The Road Not Travelled: are a complex matter. This report captures Education Reform in the Middle East and many, though not all, of the contributing fac- North Africa. Washington, DC: World Bank. tors. For example, while the report addresses ———. 2009. From Privilege to Competition: energy and agricultural subsidies, as well as Unlocking Private-Led Growth in the Middle the link between jobs and infrastructure, East and North Africa . Washington, DC: World Bank. others issues of relevance for job creation— ———. 2011a. Financial Access and Stability: such as the role played by urbanization in the A Road Map for the Middle East and North agglomeration process, or the role of ongoing Africa. Washington, DC: World Bank. conflict—are left for future research. ———. 2011b. Economic Development Prospects: Finally, this report covers all the countries Investing for Growth and Jobs. Washington, of the Middle East and North Africa region in DC: World Bank. line with the World Bank definition. However, ———. 2013. Opening Doors: Gender Equality availability of recent and good-quality data on in the Middle E ast and Nor th Afric a . firms and households and access to information Washington, DC: World Bank. Acknowledgments T his report is the product of a collab- provided by Ghassan Alkohja, Mohamad orative effort of a core team led by Alloush, Yoonyoung Cho, Zaina Dawani, Roberta Gatti and comprising Diego Angela Elzir, Lire Ersado, Callie Ham, Amira Angel-Urdinola, Stefanie Brodmann, Rebekka Kasem, Nandini Krishnan, Maurice Saade, Grun, Elizabeth Mata Lorenzo, Daniela Ernest Sergenti, Rene Leon Solano, Simon Marotta, Juan Manuel Moreno, Matteo Thacker, Tara Vishwanath, and Jeffrey Waite. Morgandi, and Marc Tobias Schiffbauer, The report benefited greatly from the feed- with important contributions from Andras back received during an extended round of Bodor, Anne Hilger, Yevgeny Kuznetsov, Bob consultations. In particular, we are grateful Rijkers, Amina Semlali, Sebastian Trenner, to representatives of academia, civil society, Paolo Verme, and Thomas Walker. In addi- governments, and international organizations tion, Stefanie Brodmann, Rebekka Grun, and who participated in the consultations held in Matteo Morgandi supported the coordination Algiers, Beirut, Cairo, Marseille, Rabat, and of the project at different stages. Kelly Cassa- Washington, DC. We thank the Economic day and Amy Gautam edited the report, with Research Forum in Cairo, the European further efforts of Mark Ingebretsen and Paola Union delegation in Tunis, the Economic and Scalabrin from the World Bank’s Office of the Social Commission for Western Asia in Bei- Publisher. rut, the International Labour Office regional Background papers for the report were Beirut office, the International University of written by Ahmed El-Ashmawi (Arab Repub- Rabat, and the World Bank Executive Direc- lic of Egypt), Saad Belghazi (Morocco, tor Office for Maghreb countries, as well as Tunisia), Mongi Boughzala (Tunisia), Odile World Bank country offices in Beirut, Mar- Mornet-Cariou (Tunisia), Carole Chartouni seille, Rabat, and Tunis for kindly hosting (United Arab Emirates), Carl J. Dahlman the many consultations for this report. We (regional), Abdelkader Djeflat (regional), Rana thank Nabil Ben-Nacef (European Union Hendy (Egypt), Mouloud Mohamed-Meziani delegation, Tunis), Dale Lautenbach, Haneen (Algeria), Nader Mryyan (Jordan), Saib Sayed, and Anwar Soulami for their skillful Musette (Algeria), Tania Rajadel (Tunisia), Ali moderation of the sessions. Sanaa (Tunisia), Irene Selwaness (Egypt), and We are thankful to our peer reviewers— John Sloan (Jordan). Other useful inputs were Arup Banerji, Omar Razzaz, and Michal xxiii xxiv ACKNOWLEDGMENTS Rutkowski—for their valuable guidance; David Robalino, Jan Rutkowski, Andrew our advisory board—Ragui Assaad, Gordon Stone, and Emiliana Vegas for detailed com- Betcherman, Ishac Diwan, and Ahmed Galal; ments on earlier drafts of the report’s chap- to Gary Field and our colleagues—Hana ters. During the preparation of this report, the Polackova Brixi, Ndiame Diop, Mourad team benefited from a productive exchange Ezzine, Guenter Heidenhof, Jesko Hentschel, with the team for World Development and Nick Manning—for their insightful feed- Report 2013: Jobs. back. We thank Omar Arias, Luis Crouch, This work was conducted under the stra- Mary Hallward-Driemeier, Elena Ianchovi- tegic guidance of Inger Andersen and was china, Phillip Keefer, Jennifer Keller, Gloria overseen by Caroline Freund and Steen La Cava, William Maloney, Mamta Murthi, Jorgensen. Contributors While this report is the product of a collaborative effort, primary authors for the sections are as follows: Preface and Overview Roberta Gatti Chapter 1 Diego Angel-Urdinola, Matteo Morgandi, and Elizabeth Mata Lorenzo, with inputs from Anne Hilger Chapter 2 Diego Angel-Urdinola, Elizabeth Mata Lorenzo, and Matteo Morgandi, with inputs from Anne Hilger and Thomas Walker Chapter 3 Daniela Marotta and Paolo Verme Chapter 4 Marc Schiffbauer with Bob Rijkers Chapter 5 Rebekka Grun, with inputs from Andras Bodor Chapter 6 Stefanie Brodmann and Juan Manuel Moreno Chapter 7 Matteo Morgandi and Bob Rijkers Chapter 8 Rebekka Grun, with inputs from Andras Bodor and Amina Semlali Chapter 9 Juan Manuel Moreno and Stefanie Brodmann Chapter 10 Roberta Gatti, Rebekka Grun, and Matteo Morgandi Abbreviations AKP Turkish Justice and Development Party CAPMAS Central Agency for Public Mobilization Statistics CEO chief executive officer CSO civil society organization CWA collective wage agreement D-index dissimilarity index ECA Europe and Central Asia ELMPS Egypt Labor Market Panel Survey EPZ export processing zone ER employment rate ES Enterprise Surveys EU European Union FDI foreign direct investment GCC Gulf Cooperation Council GDP gross domestic product HBS Household Budget Survey ICA investment climate assessment ILO-KLM International Labour Organization–Key Indicators of the Labour Market IRR internal rate of return ISCO International Standard Classification of Occupations LAC Latin America and the Caribbean LAO limited access order LFP labor force participation LFPR labor force participation rate LFS Labor Force Survey LI labor inspection LIC low-income countries LLL lifelong learning LMIS labor market information systems LMO labor market observatory xxv xxvi ABBRE VIATIONS LMPS Labor Market Panel Survey LSMS Living Standards Measurements Survey M&E monitoring and evaluation MECOVI Program for the Improvement of Surveys and the Measurement of Living Conditions in Latin America and the Caribbean MENA Middle East and North Africa MFN most-favored nation NEET neither in education, employment, or training NGO nongovernmental organization OAO open access order ODA official development assistance OECD Organisation for Economic Co-operation and Development OOLF out of the labor force PAYG pay as you go PES public employment service PISA Programme for International Student Assessment REER real effective exchange rate SME small and medium enterprises SOEC single open-ended contract SSC Social Security Corporation SYPE Survey of Young People in Egypt TFP total factor productivity TIMSS Trends in International Mathematics and Science Study TVET technical and vocational education and training UA unemployment assistance UAE United Arab Emirates UI unemployment insurance UR unemployment rate USAID U.S. Agency for International Development VET vocational education and training WAP working-age population Main Messages T he Middle East and North Africa has a large reservoir of untapped human resources, with the world’s highest unemployment rate among youth and the lowest participation of females in the labor force. Desirable jobs, defi ned as high-paying or formal jobs, are few, and private employment is overwhelmingly of low added value. Overall, the region’s labor markets can be characterized as being inefficient, inequitable, and locked in a low-productivity equilibrium. High wage differentials and low mobility of workers to better-quality jobs underscore the fact that human capital is not allocated to its most productive use. The fact that access to desirable jobs depends more on circumstances beyond individual control than on merit leads to an inequitable distribution of employment opportunities. Finally, while some of the most coveted jobs are in the public sector and provide high indi- vidual returns, these are not necessarily associated with the highest productivity for society. MESSAGE 1: Change the “rules of the game” to create a dynamic private sector that capitalizes on the full range of the region’s human capital. The slow growth of fi rms and their limited capacity to generate quality employ- ment are directly linked to the rules governing the business environment. Uneven enforcement of regulations, access to credit on the basis of privilege and connec- tions, and lucrative markets still protected by multiple legal and regulatory barriers have combined to stifle competition. However, an aggressive simplification of busi- ness procedures and regulations, making them transparent and easily understood and holding the authorities that administer them accountable, could provide a rem- edy. Access to credit could be expanded by increasing bank competition, embracing regulatory reforms, and building credit information systems, which would allow banks to lend to more fi rms and small borrowers. A comprehensive reduction of barriers restricting entry and exit into protected markets would create the incen- tives to invest and innovate and thereby increase the demand for labor. xxvii xxviii MAIN MESSAGES MESSAGE 2: Let skills flow into productive private sector jobs by realigning employment conditions in both the private and the public sector, and by rethinking labor regulation. Lower the barriers holding back women who want to work. Strict regulations and a public sector that continues to offer better employment con- ditions severely limit the flow of skilled labor to the most productive sectors of the economy. While these restrictions protect a minority of existing workers, they have contributed to high rates of informal employment and joblessness, especially among youth and women. These distortions are further exacerbated by energy subsidies that make investment in machinery relatively cheaper than hiring workers. The introduc- tion of unemployment insurance and a well-targeted social safety net would allow governments to ease labor regulations and gradually phase out very costly energy subsidies. Specific policies would be needed to lower the barriers that women who want to work face so as to guarantee a safe working environment and support for the extra domestic burdens that they shoulder. MESSAGE 3: Make young people employable by closing information gaps, improving quality and relevance of skills, and partnering with the private sector in training. Although young people and families in the Middle East and North Africa invest heavily in education and training, the majority of youth cannot capitalize on such investments and use their skills to best advantage. The quality and relevance of the skills acquired are low, and the importance of merit in gaining a job is limited. Meri- tocracy in access to education and hiring, the availability of multiple pathways in education, and the provision of second-chance options are key elements for develop- ing a productive workforce. More meritocracy signals market needs more clearly to educational and training systems. By doing so, it creates demand for the “right” skills in the “right” areas and reduces mismatches. Finally, the quality of education and its value in the labor markets need to be assessed regularly through standardized tests and the results fed back as a basis for reform and as a way to empower users. MESSAGE 4: Use short-term interventions to respond to immediate needs while building the credibility and consensus for medium-term, game-changing reforms. Establishing the right environment for the implementation of reforms requires a broad consensus on the nature of the employment problem and building inclusive constituencies for change. Some of the critical first steps are improving access to data and information, including the full range of social actors in the reform dialogue, and implementing short-term programs that deliver visible results and thereby support the credibility of the process. Executive Summary J obs are crucial for individual well-being. gender, location at birth, and parental educa- They provide a livelihood and, equally tion rather than individual effort. Some of the important, a sense of dignity. They are most coveted jobs, such as those in the public also crucial for collective well-being and sector, provide high individual returns but do economic growth. However, the rules and not necessarily provide the highest social pro- incentives that govern labor markets in Mid- ductivity, exacerbating these inefficiencies and dle East and North Africa (MENA) coun- inequities. Gallup’s recent polls show that the tries have led to inefficient and inequitable overwhelming majority of youth would rather outcomes, both individually and collectively. work in the public sector. The incentives that Several underlying distortions prevent a workers face therefore push MENA labor more productive use of human capital and markets into a low-productivity equilibrium: have led to a widespread sense of unfairness to increase the chances of securing a public and exclusion, of which the Arab Spring was sector job, young people choose higher educa- a powerful expression. tion degrees that are not relevant to the pri- The MENA region has vast untapped vate sector. Those who can afford to, spend human resources, along with the world’s time queuing for those public sector jobs with highest unemployment rate among youth and the expectation of relatively high pay for low the lowest female participation in the labor productivity. In light of the fiscal constraints force. At the same time, desirable jobs— that countries are facing and the rising share defined as high-paying jobs or those with of highly educated individuals, this model of social insurance coverage—are few: private labor market success is unsustainable. employment is overwhelmingly of low value Many barriers contribute to this equilib- added, and the public sector still provides the rium. The process of creative destruction, majority of formal jobs. High wage differen- which led to technological upgrading in tials, low mobility, and stark and persistent fast-growing East Asian and Eastern Euro- geographic differences in employment out- pean economies, is weak in MENA’s private comes underscore the inefficiency with which sector. A few well-established fi rms operate labor markets in MENA allocate human under a regime of privilege and modest com- capital. Moreover, access to desirable jobs petition, while young businesses and small predominantly reflects circumstances such as and medium enterprises struggle to grow or xxix xxx EXECUTIVE SUMMARY gain access to markets. Moreover, energy employment challenge in the region are subsidies distort relative input prices, thereby linked by the common thread of breaking promoting capital-intensive rather than privilege to foster inclusive economic and labor-intensive production. Discretionary employment growth. enforcement of regulation and constrained In the private sector, creating a level play- access to credit undercut competition and ing field on which firms can compete and dynamism in the private sector. As a result of grow and enacting reforms that improve these factors, job creation has not been suffi- access to credit and remove the current dis- cient to keep up—in numbers or in quality— tortions in relative input prices are critical. with the demographic pressure from new More moderate employment-protection leg- labor market entrants nor with the expecta- islation, accompanied by a system of income tions of increasingly educated youth. support for displaced workers, could promote Existing labor market regulations rein- more dynamic labor markets. A realign- force the status quo. In some countries, ment of public sector employment will likely wage floors and rigid employment protec- reduce public-private dualism. Finally, pen- tion generate incentives for fi rms to operate sion reform will be needed to address the cur- informally, which limit their opportunities rent fiscal unsustainability of these systems to grow. Moreover, the employment package and make the extension of coverage afford- offered by the public sector creates impor- able. Producing the skill sets necessary to sus- tant distortions: with generally good salaries, tain this new dynamic economy will require benefits, and job security, it outcompetes the better governance, including more system- private sector in attracting qualified workers. atic measurement of results and increased With no feedback mechanisms from firms involvement of the relevant stakeholders. and a legacy of state-led industrialization, Most of these policy recommendations the main focus of educational systems con- are well known and broadly shared. If this is tinues to be the production of future employ- the case, why haven’t these reforms come to ees for the public sector. While MENA has pass? What could fi nally trigger the reform the highest percentage of fi rms in the world process? complaining of inadequate skills, the incen- MENA countries face a complex legacy. tives for educational and training systems to Before the Arab Spring, governments tended cultivate relevant skills and competencies are to rely on a system of rent sharing among muted. Moreover, a rigid tracking system and a relative minority, as well as on univer- restrictive university admission policies pro- sal subsidies of basic consumption goods. mote a logic of selection rather than of qual- There were many aspects to this equilibrium: ity of education. This result is reflected in (1) public employment responded to the MENA’s disappointing performance in inter- demand for guaranteed secure jobs by the national assessments such as the Programme middle and upper classes; (2) access to credit for International Student Assessment (PISA) was concentrated in an extremely small set of and the Trends in International Mathematics well-connected fi rms; (3) restrictions on the and Science Study (TIMSS) when compared right of association strongly limited the role to countries with a similar gross domestic of civil society organizations; and (4) lack of product. When formal skills have less value access to data prevented an informed social in hiring than other factors such as personal dialogue. networks, the distribution of jobs becomes The Arab Spring’s powerful demand for even less equitable. democracy and voice carries unprecedented The Arab Spring and increasing global opportunities for disrupting this system of competition have shown that this fragile privilege and for moving toward a new and equilibrium cannot be sustained any longer— more inclusive model of development. How- socially or economically. The medium-run ever, it also carries challenges: governments policy options for addressing the structural are under tremendous pressure to deliver EXECUTIVE SUMMARY xxxi results rapidly, which might push them on jobs include well-evaluated programs toward populist, fiscally costly measures that to improve youth employability, as well as continue to reinforce current divides. Long- labor-intensive public works. Implement- time outsiders might see the current political ing reforms to improve access to credit transformation as the opportunity to fi nally would reduce one of the biggest constraints become the new insiders, thus generating a to growth and job creation for small and mere alternation of elites. Moreover, there is medium fi rms and also endogenously pro- evidence that it is difficult for new govern- mote a broader private sector constituency ments to embrace ambitious reform agendas, for reform. However, without transparent something that could be particularly daunt- information and a truly inclusive social dia- ing in the midst of the current drop in eco- logue, it will be hard for the region to live up nomic growth and fiscal revenues. to its great potential and to come to a shared A number of processes, however, could view on the merits of reforms. As it has trigger and support the political economy occurred historically in other regions, this of reform. By investing early in visible and process will require investing in data open- measurable gains, governments in MENA ness and leveraging the new social forces, could consolidate their credibility and pre- including a bigger and more representative pare the ground for game-changing reforms role for trade unions, employers’ associa- to come. Short-term interventions that can tions, and civil society. have important, albeit transitory, impacts And, if not now, when? Overview Introduction overwhelmingly affects youth and women. In some countries, such as the Arab Republic J obs are crucial for individual well-being. of Egypt and Tunisia (figure O.3), the highly They provide a livelihood and, equally educated are more likely to be unemployed. important, a sense of dignity. They are In most countries however, the majority of also crucial for collective well-being and the unemployed are still medium- or low- economic growth. However, the rules and the skilled individuals. Among those who are incentives that govern labor markets in coun- employed, low-quality jobs—those char- tries of the Middle East and North Africa acterized by low pay and productivity, and (MENA) region have led to inefficient and lack of access to social security—tend to be inequitable outcomes both individually and the majority. The formal private sector— collectively. In particular, several underlying quite likely the most productive segment of distortions prevent a more productive use of the economy—is small, and in no country human capital and have led to a widespread in MENA for which data are available does sense of unfairness and exclusion, of which the formal private sector employ more than the “Arab Spring” was a powerful expression. 20 percent of workers. At the same time, pub- MENA has a large share of untapped lic sector employment continues to be large in human resources. Unemployment and inac- the Persian Gulf countries and in countries tivity are more prevalent in MENA than in such as Egypt, Iraq, Jordan, and, to a lesser other middle-income regions like Eastern extent, Tunisia, making up between 60 and Europe and Central Asia and Latin America 80 percent of total formal employment (see and the Caribbean (figure O.1), as a high pro- figure O.5). portion of the working-age population, par- How can one interpret the realities of ticularly women, is inactive or experiences labor markets in MENA? What are the bar- high unemployment. riers to creating more and better-quality Three out of four working-age women do jobs? What policy options do governments not participate in the labor force and con- have for addressing these challenges? And, stitute 80–90 percent of MENA’s inactive finally, what political-economy processes population (see figure O.2). Unemployment is might facilitate and enable reform? This persistently higher than in other regions and report tries to answer these questions by 1 2 JOBS FOR SHARED PROSPERITY FIGURE O.1 Composition of the working-age population in MENA and two other regions, 2010 percent a. Selected MENA countries b. Selected LAC countries c. Selected ECA countries 19 31 27 38 40 48 27 5 37 6 6 16 Formal workers Informal workers Unemployed Inactive Source: Based on the International Labour Organization–Key Indicators of the Labour Market (ILO-KILM) database. Note: ECA = Europe and Central Asia; LAC = Latin America and the Caribbean; MENA = Middle East and North Africa. FIGURE O.2 Rates of female labor force participation, by region, job creation and quality. It then discusses 2008 how these factors have collectively produced the employment challenges in the region and 80 what policy options would be effective in overcoming them. 60 Labor markets in MENA: Inefficient and inequitable Percent 40 Since a large share of the population either does not participate in any economic activ- 20 ity or gains little out of it because of the lim- ited availability of high-value-added jobs, the labor markets in the region function at 0 a low-productivity equilibrium. In addition, ia e C cific an e rth a ca EU a t ld as si ric op the allocation of talent is distorted toward As or fri be hA eE nd Af ur W Pa nA st ib dl a tE ut Ea d the public sector, which has limited pro- ar a CD id So an as r No a M OE he h sia th Sa ductivity in the aggregate but offers highly ut A b- d So st an Su a desirable conditions from the individual nd he ica la ut er ra So standpoint. This equilibrium is both ineffi- Am nt Ce tin Region cient and inequitable. La Source: ILO–KILM database. Inefficient allocation of human capital Note: EU = European Union; OECD = Organisation for Economic Co-operation and Development. A number of symptoms indicate that looking at the development process through labor markets in MENA do not allocate a jobs lens, with a focus on the region’s human capital to obtain its highest return: underlying labor demand, the incentives for (1) important wage differentials persist investing in and generating relevant skills, across sectors and individuals that are not and the regulatory environment as it affects explained by differences in human capital; OVERVIEW 3 (2) workers’ mobility from low-productivity FIGURE O.3 Youth unemployment rates in the Arab Republic of to high-productivity employment is very Egypt and Tunisia by education level, ages 15–29, 2010 limited; and (3) geographical differences in labor market outcomes persist. 50 45 Wage differentials for similar levels 40 of education 35 In many MENA countries, returns to edu- cation (especially primary and secondary 30 Percent education) are high in the public and formal 25 private sector and very low in the informal 20 sector. Figure O.6 plots the expected wage 15 of an average worker by years of education 10 and employment in the public, formal pri- 5 vate, or informal private sector in Egypt 0 and in Jordan. A typical worker in the for- Primary or below Secondary Tertiary mal private sector, with no experience and Education level 12–16 years of education (equivalent to Egypt, Arab Rep. Tunisia completing secondary and tertiary educa- tion, respectively), would earn a salary twice Source: Based on the Tunisia Labor Force Survey (LFS) 2010 and the Egypt LFS 2010. to three times that of a worker with similar education in the private informal sector.1 FIGURE O.4 Youth unemployment rates by region, Low mobility across high-pay ages 15–24, 2008 and low-pay sectors Low mobility across different kinds of 30 employment, especially toward better- quality jobs, indicates that labor markets do 25 not arbitrage the allocation of human capi- 20 Percent tal effectively across job opportunities. For 15 instance, when data from the Egypt Labor 10 Market Panel Surveys (ELMPSs) are used to track changes in employment status for the 5 same individuals from 1998 to 2006, results 0 indicate high persistence in public sector a ia sia sia an ca EU ic No East ld ric As cif fri be or lA hA d Af Pa nA an st W e employment and in formal wage work (see t A arib ra ut dl rth Ea d nt ra CD id So an eC Ce ha M OE table O.1). sia th Sa d an b- d an Although mobility is higher among infor- Su pe as ica he ro Eu ut mal wage workers, their chances of moving er So Am to better jobs over the eight-year period cov- tin La ered by the surveys are still limited. Similarly, self-reported mobility data from Jordan and Source: ILO–KILM database. Lebanon indicate that job mobility across Note: EU = European Union; OECD = Organisation for Economic Co-operation and Development. informal and formal employment is lower in these countries than observed in comparator jobs are in limited supply. In addition, strict countries such as Mexico (Gatti et al. 2012). labor regulation prevents employers from Overall, the evidence points to limited dismissing workers in response to business labor reallocation, especially toward more needs. As a result, few vacancies are likely to productive jobs. Two factors contribute to this open at any point in time, giving incentives “mobility deficit.” First, high-value-added to those who obtain jobs early in life to stay 4 JOBS FOR SHARED PROSPERITY in them for long periods and to those who do FIGURE O.6 Expected wage by years of not to queue for an opening. education in the Arab Republic of Egypt, 2006, and Jordan, 2010 Stark and persistent geographical differences in unemployment rates a. The Arab Republic of Egypt, 2006 4.0 Stark and persistent differences in unem- 3.5 ployment rates across geographical areas Log hourly wage 3.0 indicate that geographical mobility has a 2.5 2.0 1.5 FIGURE O.5 Work status of employed individuals in selected 1.0 MENA economies, 2005–10 0.5 0 0 2 4 6 8 10 12 14 16 18 20 100 Years of education 80 b. Jordan, 2010 4.0 60 Percent 3.5 Log hourly wage 3.0 40 2.5 20 2.0 1.5 0 1.0 p. a . n Re cc o az isi a ep a Ira q ro G n bR rd 0.5 en o nd Tu a Jo m M ka t, Ar 0 Ye n yp 0 2 4 6 8 10 12 14 16 18 20 t Ba Eg es Years of education W Self-employed and unpaid Private informal wage Public sector Private formal Private informal Private formal Public Source: Based on the Jordan LMPS 2010 and the Egypt Labor Market Panel Source: Based on the Jordan Labor Market Panel Survey (LMPS) 2010; the Egypt LES 2010; the Tunisia Surveys (ELMPS) 2006; E(w|Yrs of edu) = exp(C + β1 Yrs of Edu. + β2 Experience LFS 2010; the Iraq Household Socio-Economic Survey 2006, 2007; the West Bank and Gaza LFS 2009; + β3 Experience squared). the United Arab Emirates LFS 2009; the Yemen Household Budget Survey (HBS) 2005–06; and the Note: Estimations based on Mincer model. Sample includes wage earners Morocco Household and Youth Survey (HYS) 2009. in urban areas working between 30 and 60 hours per week. TABLE O.1 Employment transition matrix in the Arab Republic of Egypt, 1998 and 2006 Formal Informal wage wage Self- Inactive, Unemployed, Public, earner, earner, Employer, employed, Unpaid, Total, All workers, 1998 2006 (%) 2006 (%) 2006 (%) 2006 (%) 2006 (%) 2006 (%) 2006 (%) 2006 (%) 2006 (%) N Workers with university education Public 11 1 87 1 1 0 0 0 100 710 Formal wage earner 5 0 20 65 6 2 2 0 100 86 Informal wage earner 11 4 21 23 26 9 7 0 100 57 Employer 7 2 10 0 2 67 11 2 100 61 Self employed 3 0 0 3 3 41 48 0 100 29 Workers with at most secondary education Public 12 0 79 2 2 2 1 0 100 971 Formal wage earner 6 1 7 52 17 9 9 0 100 172 Informal wage earner 6 2 9 9 42 20 11 1 100 598 Employer 7 0 4 2 5 69 12 1 100 297 Self-employed 11 1 5 2 10 27 40 5 100 320 Source: Based on the Egypt ELMPS, various years. Note: N = number. Sample data are for individuals between 31 and 64 years of age. OVERVIEW 5 limited role in evening out differences in the highest rates of increasing unemploy- labor market outcomes within a country— ment between 2005 and 2011, suggesting another symptom that labor markets are that the workforce in these regions faces segmented and operate in isolation. 2 This mobility constraints (figure O.7b). In par- phenomenon is exacerbated when better allel, labor force data for Tunisia indicate jobs are concentrated in particular regions that among the lower skilled and female (for instance, public sector jobs are often unemployed, the distance to available jobs concentrated in the main metropolitan is among the top reasons for refusing job areas). Tunisia, for example, has important offers. variations in labor market outcomes across regions; some of the interior regions border- Inequitable distribution of job ing Algeria face particularly high unemploy- opportunities ment rates (figure O.7a). Some of the regions with highest unemployment rates in 2005 Desirable jobs in MENA are jobs that offer (such as Gafsa and Tataouine) also displayed either protection and employment stability FIGURE O.7 Levels and trends in regional unemployment in Tunisia, 2005–11 percentage points a. Unemployment 2005 b. Change in Unemployment IBRD 39569 Mediterranean Mediterranean SEPTEMBER 2012 Sea BIZERTE Sea BIZERTE ARIANA ARIANA NABEUL NABEUL TUNIS MANOUBA TUNIS JENDOUBA JENDOUBA BEJA BEJA ZAGHOUAN ZAGHOUAN SILIANA Gulf of Hammamet SILIANA Gulf of Hammamet LE KEF SOUSSE LE KEF SOUSSE KAIROUAN MONASTIR KAIROUAN MONASTIR KASSERINE KASSERINE MAHDIA MAHDIA SIDI BOUZID SIDI BOUZID SFAX SFAX Kerkenna Kerkenna Islands Islands GAFSA GAFSA Gulf of Gulf of Gabes Gabes TOZEUR TOZEUR Djerba Djerba Chott el Jerid GABES Island Chott el Jerid GABES Island MEDENINE MEDENINE KEBILI KEBILI ALGERIA ALGERIA TATAOUINE TATAOUINE L I B YA L I B YA 15.6–22.2 12.1–24.5 12.2–15.6 4.4–12.1 This map was produced by the 10.2–12.2 Map Design Unit of The World Bank. The boundaries, colors, denominations 0.1–4.4 and any other information shown on 8.1–10.2 this map do not imply, on the part of The World Bank Group, any judgment -5.4–0.1 0 25 50 75 KILOMETERS on the legal status of any territory, or any endorsement or acceptance of such boundaries. 0 25 50 MILES Source: Based on the Tunisia LFS, various years. 6 JOBS FOR SHARED PROSPERITY or a high income. The evidence suggests circumstances outside the individual’s con- that, in many countries, the process through trol. Inequality in the outcomes “formal which individuals obtain these jobs often employment” and “public sector employ- does not reflect effort or merit (as proxied, ment” (which are correlated) is rather high by education and experience) but instead for all countries (the D-index oscillates reflects circumstances over which the indi- between 0.2 and 0.7) but is consistently low- vidual has little control, such as gender, est in Jordan and highest in Morocco (where location, family connections, and parents’ these jobs are scarcer). education (figure O.8). Building on the large literature on Disproportionate effect of the low- inequality of opportunity, the dissimilarity productivity equilibrium index (D-index) offers a synthetic measure on women and youth of inequality in the distribution of labor Labor markets in MENA seem to offer sev- market outcomes among groups with differ- eral paths to what could be understood as ent circumstances (see Barros et al. 2008). “success” in labor markets. Workers can When used to explain outcomes among achieve success through a well-protected youth in Egypt, Jordan, and Morocco, this job (especially in the public sector), through index indicates that gender plays an over- h ig h-pay i ng employ ment i n t he pr i- whelming role in determining the extent of vate sector, or through high-paying self- inequality in employment outcomes across employment. 3 However, the inefficiencies individuals. This result is unsurprising, laid out earlier affect the chances of achiev- given women’s low participation in the labor ing a “desirable job” in different ways for force and the high rates of unemployment different individuals. Figure O.9 maps the among women who do participate. Among extent to which working-age members of young men, inequality of opportunity in different socioeconomic groups hold a high- all cases is explained mainly by exogenous paying or protected job. Groups are con- structed along four main socioeconomic dimensions: gender, location (urban or FIGURE O.8 Decomposition of the dissimilarity index by rural), age (young or prime age), and skills circumstance and effort for men ages 21–34 in the Arab Republic of Egypt, Jordan, and Morocco, 2009 and 2010 (high or low). Figure O.9 suggests that the allocation of these desirable but scarce jobs (benefit- 0.7 ing 18 percent of the workforce in Morocco 0.6 and 24 percent in Egypt) is not uniform 0.5 0.4 across the population. In spite of impor- Index 0.3 tant variations across countries, in general 0.2 young people, women, and rural, low- 0.1 skilled workers are markedly less likely to 0 have such a job, while the opposite is true in Jordan Egypt, Arab Rep. Morocco Jordan Egypt, Arab Rep. Morocco Jordan Egypt, Arab Rep. Morocco Jordan Egypt, Arab Rep. Morocco the case of prime-age men. A n equally important point is that some of the most coveted jobs, such as those in the public sector, provide high Being Formal Public Permanent returns individually but are not necessar- employed employment sector job ily associated with the highest productiv- employment ity socially. Data on individual preferences % effort % circumstances from Gallup show that in some countries the overwhelming majority of the popula- Source: Based on Jordan LMPS 2010, Survey of Young People in Egypt (SYPE) 2009, and Morocco tion would rather work in the public sec- HYS 2010. tor and that this attitude persists among OVERVIEW 7 FIGURE O.9 Share of workers with high-paying or protected jobs among the working-age population in the Arab Republic of Egypt, Jordan, and urban Morocco, 2009 and 2010 a. Arab Republic of Egypt, 2009 b. Jordan, 2010 25 25 20 Rur-F-Primary 20 Rur-M-Primary % of population % of population 15 15 Urb-M-Primary-Y Rur-M-Skill Urb-F-Primary-Y Urb-F-Skill-Y Urb-M-Skill-Y 10 10 Urb-F-Primary-A Urb-M-Skill-A Rur-F-Skill Urb-F-Skill-Y Urb-M-Skill-Y Urb-F-Skill-A Urb-M-Primary-A Urb-F-Primary-A Urb-M-Skill-A Rur-F-Primary Rur-M-Primary 5 Urb-F-Primary-Y Urb-F-Skill-A 5 Rur-F-Skill Rur-M-Skill Urb-M-Primary-Y Urb-M-Primary-A 0 0 0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100 % in well-paid or protected job % in well-paid or protected job c. Urban Morocco, 2010 25 Urb-F-Primary-A 20 Urb-F-Primary-Y Urb-M-Primary-A % of population 15 Urb-M-Primary-Y Urb-M-Skill-Y 10 Urb-F-Skill-Y Urb-M-Skill-A 5 Urb-F-Skill-A 0 0 10 20 30 40 50 60 70 80 90 100 % in well-paid or protected job Less than 50% in well-paid or protected job 50% or more in well-paid or protected job Sources: Based on Egypt SYPE 2009, Jordan LMPS 2010, and Morocco HYS 2010. Note: The x-axis represents the share of workers in group J (say, urban-male-skilled-adult) holding desirable jobs, as a percentage of the working-age population; the y-axis repre- sents the share of that group in the working-age population. Group characteristics are abbreviated as follows: Urb: urban; Rur: rural; F: female; M: male; Primary: with primary educa- tion or below; Skill: with secondary education or above; A: adult (35–64); Y: youth (15–34). For instance, 79 percent of urban men aged 35–64 with secondary education or above were in well-paid or protected jobs in 2009 Egypt, and they represented 6 percent of the working-age population. the young generation, in spite of changes in relevant to the private sector. Those who hiring rates in the public sector over time can afford it spend time “queuing” with in nonoil countries (see figure O.10). In this the expectation of relatively high pay for way, workers’ individual incentives push low intrinsic productivity. In light of the MENA labor markets to a low-productivity fiscal constraints that countries are fac- equilibrium: to increase the chances of ing and the rising share of highly educated securing a public sector job, young people individuals in the labor market, this model choose advanced degrees that are not as of “success” is unsustainable. 8 JOBS FOR SHARED PROSPERITY FIGURE O.10 Preference for public sector employment among A distribution of job opportunities that youth ages 15–34 in selected MENA economies, 2010 limits the incentives to change 80 Although well-paid or protected jobs are few, 70 a large number of households benefit from 60 them because at least one household mem- 50 ber is in this condition: for example, about 6 Percent 40 of every 10 households in Egypt and Jordan 30 have at least one individual with a desirable 20 job (figure O.11). With so many people ben- 10 efiting indirectly from these jobs, the distri- 0 bution of job opportunities, though unequal, constitutes a political-economy equilibrium ria nk ba a a Ye Q es M ibya Al co ra nd G n Sa Tu lic i A ia yp Dj raq ra uti Jo p. Ar Ba dan en ar it Re aza Em rain Ku p. Sy Ba Le ri bi n A a no wa ud nis Re e m at t b oc ge t, A ibo ira ,R ra I pu r ab h L or b that proves hard to alter. This equilibrium b is a mixed blessing for the region. On the Eg d t ite es one hand, it can be considered a (very inef- W Un ficient) safety net. Even if inequality in labor Source: Based on Gallup World Poll Survey 2010. markets tends to be high, this distribution Note: MENA = Middle East and North Africa. keeps overall income inequality at a moder- ate level in MENA countries. On the other hand, the equilibrium sustains relatively high FIGURE O.11 Percentage of individuals and households with reservation wages (that is, the lowest wage desirable jobs in the Arab Republic of Egypt (2009) and Jordan (2010) for which an individual is willing to work) among both women and youth in a large a. Arab Republic of Egypt b. Jordan section of the population. It explains in part why households can afford to have so many Individuals in LF with Individuals in LF with desirable jobs desirable jobs unemployed youth and so many women out- side the labor force. Finally, this distribution of job opportunities does not provide the best 27.2% incentives for individual talent to fi nd pro- 25.5% ductive outlets. 74.5% 72.8% Who are the most affected by the low-productivity equilibrium in labor markets? Households with at least one member Households with at least one member Youth, women, and rural, low-skilled work- having a desirable job having a desirable job ers are the least likely to have a protected or well-paying job. As such, they are over- whelmingly inactive or unemployed or work informally for low pay. Figure O.12 repre- 39.4% 43.3% sents the varying degrees of disadvantage or 60.6% labor market exclusion that underlie these 56.7% outcomes. For example, some women are inactive because they are not allowed to work (exclusion), while others might decide not to seek work voluntarily (choice). % yes % no The following sections explain the spe- Source: Based on the Jordan LMPS 2010 and the Egypt SYPE 2009. cific extent of disadvantage for each of these Note: LF = labor force. groups. OVERVIEW 9 Youth FIGURE O.12 A framework for interpreting the labor market disadvantage Youth in MENA face a limited number of good employment opportunities com- pared to the adult population. Depending queuing on their socioeconomic circumstances, young people respond to labor market conditions either by weathering long-term Unemployed unemployment, if they can afford it; work- with no job offer ing informally out of need; or becom- ing discouraged and stopping their job Labor search altogether. Many young women discouraged working poor remain inactive because of choice or family willingly market disadvantage pressure. inactive Inactive Informal High joblessness and young people’s forbidden distance from the labor market student entrepreneur Unemployment is conventionally singled out as overwhelming evidence for a youth dis- advantage in the labor market, as it is high- est among the young. However, the share of young people neither in employment, educa- FIGURE O.13 Youth NEET rates in selected MENA economies tion, or training (NEET) better captures the by education level, individuals ages 15–24, 2008–10 extent to which young people are not partici- pating in the labor market and not building Morocco skills to participate in the future. Figures O.13 and O.14 show that NEET Yemen, Rep. rates in MENA countries are pervasive and that they can be many times higher than West Bank and unemployment rates, particularly for those Gaza with little education. For instance, according Tunisia to household survey data, in Iraq unemploy- ment among individuals aged 15–24 with a Jordan primary school education is 18 percent; per- haps more tellingly, though, 58 percent of Iraq individuals in the same reference group are Egypt, Arab Rep. neither in education nor working. Moreover, for a large portion of out-of- 0 10 20 30 40 50 60 70 school youth, inactivity is more prevalent Percent than unemployment. While this fact is less Tertiary Secondary Primary surprising with respect to young women, Source: Based on the Republic of Yemen HBS 2005; the West Bank and Gaza LFS 2008; the Tunisia LFS given historical trends in the region, it is wor- 2010; the Jordan LMPS 2010; the Iraq HSES 2006; the Egypt SYPE 2009; and the Morocco LFS 2009. rying that inactivity is more prevalent than Note: MENA = Middle East and North Africa. Data refer to those out of school and out of work (OSOW). unemployment even among male school leav- ers in many countries, including Egypt, Iraq, Slow transition to work Lebanon, and the Republic of Yemen. Much High youth joblessness can be interpreted of the inactivity among young men could mainly as the result of a slow, incomplete be due to discouragement, which in turn is transition from school to work—in other indicative of their hopelessness and progres- words, an extended period of entry into the sive disengagement from labor markets. labor market for those seeking jobs for the 10 JOBS FOR SHARED PROSPERITY FIGURE O.14 Unemployment rates in selected MENA economies by whatsoever, and even many years after they education level, individuals ages 15–24, 2005–10 complete their studies, the prevalence of unemployment decreases only partially for Morocco this group. For instance, in the West Bank and Gaza, it takes nine years for 75 percent Yemen, Rep. of secondary-educated males (who no lon- West Bank and ger study) to enter employment. The transi- Gaza tion is much shorter for tertiary-educated Tunisia men, at three years. In spite of their relatively Jordan small size and relatively quicker transition to employment, tertiary educated job-seekers Iraq benefit from about half of all active labor Egypt, Arab Rep. market programs across MENA countries, according to a recent survey of public employ- 0 10 20 30 40 50 60 Percent ment services (Angel-Urdinola, Kuddo, and Tertiary Secondary Primary Semlali 2013). Working for the public sector continues to dominate young people’s job preferences. Source: Based on the Republic of Yemen HBS 2005; the West Bank and Gaza LFS 2008; the Tunisia LFS 2010; the Jordan LMPS 2010; the Iraq HSES 2006; the Egypt SYPE 2009; and the Morocco LFS 2009. Public employment has become ingrained in Note: MENA = Middle East and North Africa. Data refer to those who are unemployed. society as the main path toward social mobil- ity, even though it is no longer as viable as before. New cohorts of labor market entrants fi rst time. This slow transition results from continue to expect jobs resembling those two concurrent dynamics. The first is the occupied by previous generations. Analysis of historically sluggish demand for labor in the internal rate of return to a government job the private sector and falling hiring rates in in Morocco over a worker’s lifetime shows the public sector in some countries, which that even accounting for the forgone income together curtail the number of jobs avail- during unemployment, the current generous able relative to the number of labor market conditions for public sector work continue to entrants. The second involves young peo- make queuing for a public job a frustrating ple’s expectations and preferences for seek- but rational strategy for educated job seekers ing a public sector or a highly paid private (Bodor, Robalino, and Rutkowski 2008). sector job exclusively. In most countries in MENA, men with Women higher-level education make the transition from school to work surprisingly more rap- Low rates of labor force participation idly than men with secondary education, On average, three out of four women in with the notable exception of Egypt and MENA are outside the labor force, and they Tunisia. Hence, even if during the years of constitute the vast majority of the inactive transition the more highly educated face population. Although women’s participation higher unemployment, tertiary education in the labor force in the region has increased represents a relative advantage in the labor in the past few decades, that increase has market. According to one interpretation of been slow. At the current rate, and given the this phenomenon, after completing their stud- low starting point, it would take 150 years for ies, those with a tertiary education appear MENA countries to reach the current world to search for jobs more intensively and to be average for the labor force participation of more selective about which jobs they accept women (World Bank 2013). Low levels of and after some years tend to settle into a job. female labor force participation are explained Less-educated job seekers appear to encoun- primarily by the lack of participation among ter much greater difficulty in finding any job women without tertiary education, who OVERVIEW 11 constitute the vast majority of the working- FIGURE O.15 Female participation in the labor force in five MENA age female population (figure O.15). countries, by education level, various years, 2005–10 The typical factors that determine an increase in women’s labor force participa- 100 tion worldwide, such as a decrease in fertil- 80 ity and improvements in education, also play a role in MENA. However, other factors, 60 Percent such as social norms and the role assigned to women within the family, remain impor- 40 tant determinants. For example, evidence 20 from Morocco indicates that women who are allowed to decide alone whether to par- 0 ticipate in the labor force are much more a q n co p. isi a Ira Re oc rd n Tu Jo or likely to participate (figure O.16). The same b ra M t, A data also show that, after marriage, wom- yp Eg en’s decision-making power is substantially reduced. Primary or below Preparatory, general secondary, Tertiary or vocational secondary Disadvantages for working women in MENA Source: Based on the Tunisia LFS 2010, the Jordan LMPS 2010, the Iraq HSES 2006, the Egypt LFS 2010 As in many other countries, in MENA and the Morocco LFS 2009. women who work are often concentrated Note: MENA = Middle East and North Africa. in specific occupations. Data for 2006 from Egypt, for example, show that most women FIGURE O.16 Women’s decision-making on whether they can work in agriculture and education (about work and labor force participation, by marital status and age, 60 percent of all employed women), while Morocco, 2010 most men work in transport, retail, tourism, and manufacturing (which are generally bet- ter-paid industries). In Egypt, women work 60 disproportionately in the public sector, which % participating in the 50 offers employment conditions that are per- 40 labor force ceived as more suitable for them, particularly 30 in respect to job safety. Overall, 54 percent of female workers in Egypt are estimated to 20 work in civil service jobs, whereas less than 10 10 percent work in the private formal sector 0 (see also World Bank 2010). Those women cid t cid t es es de s no de s no id cid e e c e e de de for whom public sector jobs are not an do do e: d: e: d: gl rie gl option end up working in informal jobs that rie Sin ar Sin ar M M pay little. Consistent with these patterns, at school women also tend to stream into the Ages 15–24 Ages 25–34 Ages 35–44 humanities and social sciences, while men study science and engineering, which are Source: Based on Morocco HYS 2010. generally associated with higher wages in private sector jobs (World Bank 2013). A l t h o u g h i n M E N A t h e av e r a g e sector. However, the estimated gender gap employed woman is more educated than varies signifi cantly from country to coun- the average employed man (World Bank try, highlighting the heterogeneity of the 2013), female workers generally earn lower phenomenon in the region (figure O.17). wages than men, especially in the private Gaps are much wider in private sector jobs 12 JOBS FOR SHARED PROSPERITY (40–80 percent in the West Bank and Gaza actually earn more than men on average, and Egypt) and—as might be expected— perhaps because only the best female work- smaller in the public sector. In fact, in the ers self-select for formal public sector jobs. West Bank and Gaza’s public sector, women The working poor FIGURE O.17 Male-female wage gap in selected MENA economies, various years, 2006–10 Who are the working poor? Earlier analysis of earnings, benefits, and working condi- 100 tions has demonstrated that across MENA’s labor markets, informal jobs tend to be 80 characterized by low pay and low produc- tivity. Throughout MENA, informality con- 60 sistently decreases as wealth increases, even Percent 40 though in some countries the phenomenon is widespread enough to affect the wealthier 20 segments of the population as well, such as high-earning entrepreneurs (see Gatti et al. 0 2012) (figure O.18). Informality is a complex –20 phenomenon; yet one segment of the work- All workers Public sector Private sector force displays the lowest chance of access- Egypt, Arab Rep. Jordan West Bank and Gaza ing desirable jobs and the highest chance of being informal and poor unskilled rural Source: Based on the Egypt LMPS 2006, the Jordan LMPS 2010, and the West Bank and Gaza LFS 2008. workers, defined as those with a primary Note: MENA = Middle East and North Africa. The sample is urban workers working between 30 and education or below who live in rural areas.4 60 hours per week. FIGURE O.18 Informality rate by quintile of per capita consumption in selected MENA countries, 2011 100 80 60 Percent 40 20 0 Yemen, Rep. Morocco Syrian Arab Republic Lebanon Quintile 1 Quintile 2 Quintile 3 Quintile 4 Quintile 5 (poorest) (richest) Source: Angel-Urdinola and Tanabe 2012. Note: MENA = Middle East and North Africa. The consumption aggregate was not available for Morocco for 2010 when this report was completed. Hence, in this figure, data for Morocco are from the 2001 Living Standards Measurements Survey (LSMS). OVERVIEW 13 For instance, among the working popula- FIGURE O.19 Probability of being in a poor household by tion in Jordan, the probability of belonging education and geographic location among the employed to the bottom-income quintiles peaks among population in Jordan, 2006–10 unskilled rural workers (figure O.19). According to the defi nition adopted here, 60 the working poor constitute a considerable share of the working population in many 50 countries in the region, ranging from about % of workers in bottom consumption quintile 10 percent in Jordan to more than 50 percent 40 in Yemen (figure O.20). The working poor 30 are primarily employed in agriculture. Consistent with the slow structural trans- 20 formation of the economy in many MENA countries, unskilled rural employment con- 10 tinues to be very prevalent even among young workers. 0 Unskilled rural workers are predominantly y sic ry ry ar da tia ba im nonwage employees. In contrast, the preva- n r y/ Te pr co ar Se w im lence of nonwage employment tends to be lo Pr Be far lower among skilled and urban workers. Urban informal Rural formal In nearly all countries, nonwage workers are Urban formal Rural informal excluded from social security coverage and are therefore informal. Data indicate that this Source: Based on Jordan’s LMPS 2010. type of self-employment most often consists of subsistence entrepreneurship. Many of the unskilled self-employed report choosing this type of employment because no other suitable job was available to them. FIGURE O.20 Education level of unskilled rural workers as a percentage of the working population, 2006–10 Where are the barriers? 60 The following sections describe in detail the barriers that prevent the region from moving 50 to a higher-productivity equilibrium. These %, working population 40 include the macro- and microeconomic bar- riers that hinder the generation of more and 30 better-quality jobs; existing labor market regulation that reinforces the status quo; 20 employment conditions in the public sector that distort workers’ incentives; training and 10 educational systems that lack incentives to build high-quality, relevant skills; and inef- 0 an q a za p. co . ep fective labor intermediation systems that isi Ira Re Ga oc rd ,R n Tu Jo or b en d further contribute to the lack of a level play- ra an M m t, A nk Ye yp ing field for qualified candidates. Ba Eg t es W Is growth a problem? Primary Below primary MENA is a region with an extraordinary het- Source: Based on the Jordan LMPS 2010, the West Bank and Gaza LFS 2008, the Iraq HSES 2006, the erogeneity in socioeconomic characteristics, Tunisia LFS 2010, the Egypt LFS 2010, the Yemen HBS 2006, and the Morocco LFS 2009. 14 JOBS FOR SHARED PROSPERITY TABLE O.2 Heterogeneity in MENA economies Labor availability Labor abundanta Resource richb Algeria, Islamic Republic of Iran, Iraq, Syrian Arab Republic, Republic of Yemen Resource poor Djibouti, Arab Republic of Egypt, Jordan, Lebanon, Morocco, Tunisia, West Bank and Gaza Labor importingc and resource rich Bahrain, Kuwait, Libya Oman, Qatar, Saudi Arabia, United Arab Emirates Population size Larged Algeria, Arab Republic of Egypt, Islamic Republic of Iran, Iraq, Morocco, Saudi Arabia Small Bahrain, Djibouti, Jordan, Kuwait, Lebanon, Libya, Oman, Qatar, Tunisia, United Arab Emirates, West Bank and Gaza, Republic of Yemen Income Low income Djibouti, Republic of Yemen Lower-middle income Arab Republic of Egypt, Iraq, Morocco, Syrian Arab Republic, Tunisia Upper-middle income Algeria, the Islamic Republic of Iran High Kuwait, Oman, Saudi Arabia, United Arab Emirates Geography or colonial heritage Maghreb Algeria, Libya, Morocco, Tunisia Mashreq Arab Republic of Egypt, Iraq, Jordan, Lebanon, Syrian Arab Republic, West Bank and Gaza Members of the Gulf Cooperation Council Bahrain, Kuwait, Oman, Qatar, United Arab Emirates a. Refers to net inflows of workers’ remittances. b. As defined by the World Bank (2004), “resource-rich countries” are those in which extractive industries account for, or are expected to soon account for, more than 50 percent of government revenue. c. Refers to net outflows of workers’ remittances. d. Population over 20 million. FIGURE O.21 Five-year moving average of real GDP growth rates and 1970s, most countries experienced solid in GCC and in Non-GCC MENA countries, 1960–2010 growth arising from public investments in infrastructure, education, and health which 25 were fi nanced by favorable oil prices. 5 The high growth rates (averaging 7 and 6 percent 20 respectively in the 1960s and 1970s) reflected Smoothed growth rate, % not only the accelerated accumulation of pro- 15 duction factors, but also the increased pro- ductivity linked to much-needed investments 10 in physical and human capital. As interna- tional oil prices plummeted in the 1980s, 5 however, the foundations of growth in earlier decades collapsed. With eroding macroeco- 0 nomic balances and growing debt burdens, 1960 1970 1980 1990 2000 2010 and despite heavy external assistance (which year permitted spending for several more years), GCC countries Non-GCC MENA countries Lower-middle-income Upper-middle-income investments declined dramatically, and the countries countries rate of growth of the physical capital stock almost halved compared to the previous Source: World Development Indicators 2012. decade. Note: GCC = Gulf Cooperation Council; MENA = Middle East and North Africa. As the government-led development model became increasingly unsustainable institutional development, and natural in the late 1980s, a number of countries resources (table O.2); yet MENA countries (including Morocco, Tunisia, and, soon share significant trends in economic perfor- after, Jordan) embarked on programs of mance. Following independence in the 1960s macroeconomic stabilization and policy OVERVIEW 15 reform, followed in the 1990s by the Gulf FIGURE O.22 Total annual employment growth in selected Cooperation Council (GCC). Reforms var- regions of the world, 1998–2008 ied markedly in timing and intensity, but all included reducing subsidies and public Middle East and spending, liberalizing trade, encouraging North Africa exports and investment, and strengthen- GCC countries ing the institutional foundations of a mar- Non-GCC ket-led economy. These structural reforms countries allowed the region to catch up on several fronts. Export growth increased. A shift away from procyclical fiscal policy and Developed countries from a government-led growth model also East Asia helped reduce macro imbalances, particu- larly in Gulf countries. As a result, average Europe and Central Asia real gross domestic product (GDP) growth World rose from 3.6 percent a year between 1996 and 1999 to around 5 percent between 2000 Southeast Asia and Pacific and 2008 (figure O.21). These changes were Latin America and the associated with increases in total factor pro- Caribbean ductivity and labor force participation, with South Asia a limited role for physical and human capi- Sub-Saharan Africa tal accumulation compared to the past.6 In particular, changes in total factor productiv- 0 1 2 3 4 5 ity became the main driver of growth in the Percent early reformers such as Jordan, Morocco, and Tunisia. Source: ILO–KILM database. Note: GCC = Gulf Cooperation Council. Disappointing employment outcomes Despite the recent extended stretch of solid growth in the past decade (which has, how- FIGURE O.23 Growth in employed and working-age ever, slowed since the 2008 crisis), employ- population in non-GCC MENA countries, 1998–2009 ment outcomes in the region continue to be of concern on two main fronts: 55% not • In the past decade, the number of jobs cre- 250 working 56% not ated, while quite responsive to the level of 200 working growth, was not suffi cient to absorb the increasing number of new entrants into 150 Millions the labor market. Employment grew— 100 and at a higher rate than in other regions of the world (figure O.22)7—but so did 50 the working-age population (figure O.23), as most MENA countries are experienc- 0 98 99 00 01 02 03 04 05 06 07 08 09 ing a “youth bulge.” 19 19 20 20 20 20 20 20 20 20 20 20 • Jobs were created in relatively low-value- Year added sectors, and public employment Employed Working-age population continued to account for a substan- tial share, especially in Gulf countries. Source: Based on the ILO’s EAPEP (Economically Active Population, Estimates and This phenomenon refl ects the slow pace Projections) database. at which structural transformation has Note: GCC - Gulf Cooperation; MENA - Middle East and North Africa. 16 JOBS FOR SHARED PROSPERITY FIGURE O.24 Changes in the composition of GDP in selected generally associated with high value world regions, 1980–83 and 2007–10 added, they are capital intensive and gen- erate little employment directly. At the 80 same time, however, oil revenues have been instrumental in increasing employ- 60 ment by fi nancing the creation of more 40 public sector jobs. 20 An unfinished reform agenda Percent 0 Many argue that the slow structural trans- –20 formation in the region is the result of an –40 unfi nished reform agenda. Sound fiscal and monetary policies—prerequisites for sus- –60 tained economic performance—are still to be implemented in MENA, as is a level play- –80 ing field that would allow the private sector sia sia c an a ca ifi ric hA fri to play a pivotal role in achieving structural be lA c Af Pa nA ib ra ut rth d ar nt ra So change. A number of specific factors inhibit an eC No Ce ha ia th Sa d nd As the needed changes: an b- d ta st an Su pe Ea as ica ro eE Eu • MENA’s real exchange rate volatility, er dl Am id M which is the highest in the world. A vola- tin La Change in share of services tile real effective exchange rate reduces Change in share of industry (including mining the incentive to invest in nonresource- and manufacturing) tradable sectors by increasing uncer- Change in share of agriculture tainty. In turn, it reduces development prospects in more dynamic sectors of the Source: Diop, Marotta, and deMelo, forthcoming. economy, which are not capital intensive Note: GDP = gross domestic product. and have a higher potential for creating employment.8 • Procyclical and discretionary fi scal poli- cies, which rely on costly universal energy occurred in MENA (although results vary and food subsidies and result in a lack of across countries) with respect to econo- high, sustained investment. mies elsewhere in the world. Manufactur- • A failure to establish rule-based modes of ing contributes little to both value added interaction with the private sector, which and employment growth (figures O.24 has limited the incentives for the private and O.25). Instead, low-value-added sector to grow and generate jobs. sectors, such as construction, have driven most of the employment creation in the As a result, while the private sector plays past decade. At the same time, agricul- a larger role in MENA today than before, it ture continues to play a relatively large is far from being a strong engine of growth. role in contrast to comparator countries For instance, the contribution of private that instead saw labor shift from agri- investment to growth is among the lowest culture into services. As discussed in the in the world (figure O.26). Except for some recent MENA Economic Development of the resource-poor countries (oil importers an d Prospec t s repor t ( World Ban k in the Maghreb), where the ratio of private 2011b), natural resources remain a major investment has actually increased owing growth engine in the typical MENA to a substantial increase in foreign direct country. While extractive industries are investment, the composition of investment OVERVIEW 17 still favors public investment in most of the FIGURE O.25 Sectoral contribution to annual employment resource-rich countries. Most private invest- growth in a typical MENA country and other selected countries, ment directed to the region has been biased average 2000s toward either capital-intensive or low-skill, labor-intensive sectors (such as oil, con- 140 struction, and tourism) rather than toward 120 more dynamic, employment-creating sectors 100 like financial services and manufacturing 80 (figure O.27). Percent 60 This evidence suggests that while GDP 40 growth was solid and some employment 20 was created, an understanding of the qual- 0 ity of growth, and more specifically of the –20 dynamics of the private sector, is neces- –40 sary for identifying the determinants of the Typical MENA Malaysia Indonesia Brazil country region’s underperformance in employment outcomes. Agriculture Manufacturing Trade, tourism, logistics, Government, administration, and communication and social services The private sector: A dynamic Mining and utilities Construction based on privilege Financial and real estate services The private sector in MENA displays lit tle dynamism: it has one of the world’s highest Source: World Bank 2011b. average fi rm age, the highest average age Note: MENA = Middle East and North Africa. FIGURE O.26 Private sector gross fixed-capital formation as a percentage of GDP in selected countries, average 2004–09 25 20 15 Percent 10 5 0 Th sia As arib ia d n Ch c Le sia n A Ara sia Sa epu . Sy gyp ala a n c nd e ca Al a a an d Tun i nt isia ut d n a am ey co Af . jib . p rth ep p t i i M by ric in di a no So an ou cif bl r b Re Re t a Yem Afri oc Isl Turk be rd e C ge lA hA y No n, R In Li ail Pa ba Jo or b ic ra D M R ra an Ce ha ria t, a ia r h n, t an Ira b- d E st an Su pe Ea as ica ro eE Eu er dl Am id M tin La Country or region MENA countries Comparator countries Source: World Development Indicators 2004–09. Note: GDP = gross domestic product. MENA = Middle East and North Africa. 18 JOBS FOR SHARED PROSPERITY FIGURE O.27 Structure of foreign direct investment in selected MENA countries and the EU, 2000–07 90 80 70 60 % of total FDI 50 40 30 20 10 0 ria p. co a n on ic a on ey isi in a bl Re oc rk rd ge an ni Ch n pu Tu nU Tu Jo or b Al b ra Re Le M ea t, A b p ra yp ro nA Eu Eg ria Sy Country and region Manufacturing Telecoms Finance Tourism and construction Energy High-tech services Source: United Nations Conference on Trade and Development, World Development Indicators 2000–07. Note: EU = European Union; MENA = Middle East and North Africa. of its chief executive offi cers, and one of Lack of a level playing field the lowest fi rm densities in the world (see The lack of a level playing field inhibits the World Bank 2009; Klapper and Love 2010) employment growth of firms. Firms’ slow (figure O.28). Recent evidence from firm growth and limited capacity to generate censuses in Morocco and Tunisia shows employment are linked to the quality of the that even when fi rms enter the market, they business environment. A comparative analy- stay small and do not grow. For example, sis of the characteristics of business environ- a comparison of representative fi rm trajec- ments worldwide shows that while MENA tories in Brazil and Jordan indicates that countries have made important advances in while fi rms start out larger in Jordan, they reforming, they still rank consistently low grow more slowly over time than in Brazil. on two key dimensions: the enforcement of Over a 10-year period, for example, a fi rm regulations and access to credit. in Brazil becomes about twice as large as Legal and regulatory ambiguity expands one in Jordan (figure O.29). Overall, the the scope for discretion in public agencies, process of creative destruction—that is, and reforms designed with little transparency the birth of innovative fi rms and the exit or consultation compound the unpredictabil- of unproductive firms that characterize ity of the investment environment. Firms in regions with higher and sustained growth MENA complain bitterly about inconsistent such as East Asia—is substantially attenu- and unpredictable policy implementation, ated in MENA. OVERVIEW 19 FIGURE O.28 Average firm entry density for selected emerging economies, 2004–09 4.00 3.50 3.00 Entry density 2.50 2.00 1.50 1.00 0.50 0 es ia tia ile lic il a ey n sia p. co a az isi di a ar Re tri ub oc rk Ch rd oa ne In n Br lg Tu un Tu Jo or b ep Cr do Bu ra M co nR In t, A 80 ica yp e in Eg ag m er Do av MENA countries Comparator countries Source: Klapper and Love 2010. Note: Entry density measures the number of newly registered limited liability firms per 1,000 working-age people (those between ages 15 and 64). Average 80 countries represents the average entry density in the 80 developing countries for which data are available. and data reveal a high variation across firms FIGURE O.29 Relation between firm size and age in Brazil (2009) in the enforcement of regulations. By raising and Jordan (2006) entry costs, the unpredictable implementa- tion of policies reduces competition, rein- forces a status quo characterized by limited 50 dynamism, skews firms’ incentives toward rent-seeking activities, lowers turnover, and 40 Number of workers diminishes the probability of innovation (World Bank 2011b). 30 On the other hand, firms in MENA—small and large—have the second-lowest access to credit in the world, surpassing only Sub- 20 Saharan Africa. Probably as a consequence, MENA’s banks have highly concentrated 10 portfolios focused on large enterprises. For example, non-GCC MENA countries have 0 5 10 15 20 Age of the firm (years) the highest ratio of top 20 loan exposures to Brazil Jordan total equity.9 Both discretionary implementa- tion of regulation and limited access to inputs are particularly constraining for young enter- Source: Based on Enterprise surveys in Jordan (2006) and Brazil (2009); private prises, which have been shown to create the domestic firms only. 20 JOBS FOR SHARED PROSPERITY majority of jobs. With limited competition, demand for labor in the short run, but also incumbent firms enjoy privileges and face lit- suppress incentives to innovate, thereby tle or no pressure to innovate and create new impeding productivity growth, which is avenues for growth. the crucial determinant of long-run labor demand. They tend to disproportionately benefit older and publicly owned firms, Effects of distorted input prices which in turn are likely to use more out- In MENA, input prices are distorted and dated technologies and consequently more create an antilabor bias. Energy subsidies, energy. which are common and substantive in the region, increase the relative cost of labor in relation to the cost of energy and thereby A regulatory environment that limit labor demand. For instance, in 2007 reinforces the status quo in the Egypt’s energy-intensive industries con- labor market sumed more than 80 percent of the coun- Labor regulation in MENA is quite hetero- try’s fuel oil and 28 percent of the diesel geneous—it is more restrictive in Maghreb oil, which together receive 50 percent of countries and more flexible in GCC coun- the total energy subsidies (Abouleinein, El- tries—but all labor markets in the region Laithy, and Kheir-El-Din 2009). In fact, contend with a relatively heavy public sector. many MENA countries figure among those The design of social insurance systems also with the highest energy content per unit of influences the dynamics of labor markets: GDP in the world. Fuel subsidies are dou- very generous, costly pension systems cover bly disadvantageous: they not only repress a small minority of workers, while the lack FIGURE O.30 Labor taxes and contributions as a percentage of base salary in world regions, 2011 30 25 20 Percent 15 10 5 0 be d As ral ca GC th GC h sia ic es ib an n- ort cif fri a ( or nt tri hA ia an C) C) nA ar ca ric N Pa no N Ce un ut Af and e C eri a ( nd d co ra d So an an th Am ric t a ha CD t as ia Af Eas Sa pe tin OE As eE b- ro e La st Su dl Eu dl Ea id id M M Source: IFC and World Bank 2011. Note: GCC = Gulf Cooperation Council; OECD = Organisation for Economic Cooperation and Development. OVERVIEW 21 FIGURE O.31 Labor taxes and contributions as a percentage of base salary in MENA economies, 2011 35 30 25 20 Percent 15 10 5 0 t, A eria . p. Le a n Re o Dj . Em in s. Ba i q ud an a m an . ar it za Isl Rep ep ic t si bi c no Ira an uwa ou te Re a t bl Ga c d Om ni Qa ,R hr ra ro ira g r ba b pu Tu Jo b ic iA Al i en nd n A Mo K ra am ka ab b Ye Sa ra yp Ar n, tB Eg Ira d es ite ria W Un Sy Source: IFC and World Bank 2011. Note: MENA = Middle East and North Africa. of well-functioning unemployment insur- The level of labor taxes can affect many ance and poverty-targeted social safety nets decisions of fi rms and workers—including makes job losses extremely costly for the whether to operate informally, whether to majority. open vacancies, or whether to accept job offers—and further exacerbates wage floors. MENA countries, especially those out- Effects of rigid employment protection side the Gulf, have relatively high labor taxes on labor mobility and social contributions, ranking only after In some countries, important wage rigidities countries in the Organisation for Economic contribute to unemployment, queuing, and Co-operation and Development (OECD) informality. In Tunisia, for example, collec- and Eastern Europe, where, however, social tive wage agreements work by levels, reflect- insurance and public services have much ing employees’ education and competencies. higher coverage and scope on average than in These agreements assume that a university MENA (figures O.30 and O.31). education is needed for the highest levels. In addition to wages, contract duration Wage floors are then set accordingly, often (temporary, fixed term, or open ended), work- above what workers with less than a univer- ing hours, and dismissal procedures (includ- sity degree, who would still be qualified for ing severance pay) define the employment those jobs, are paid on the market. relationship. In the past decade, a number of As a result, the salary differences are quite countries (Egypt and Morocco, for example) remarkable in most sectors, and unemploy- reformed their labor codes; in Egypt, this ment is likely at each of these levels whenever reform resulted in substantially more flexible the mandated wage exceeds the actual pro- hiring procedures. Indeed, with the exception ductivity of the workers that are available to of Algeria and Djibouti, fixed-term contracts fill the position. for permanent tasks are now allowed in all 22 JOBS FOR SHARED PROSPERITY countries. Dismissal (from the formal sector), generally low pressure to be productive, however, remains more difficult in MENA among others) compete with and may even than in any other region of the world. Most crowd out employment conditions in the MENA countries (all but most Gulf coun- private sector. Generous public employment tries, Lebanon, and Morocco) require the conditions provide a de facto reservation employer to notify a third party (such as a wage, if not a wage floor, especially for those government agency) for the dismissal of just with higher education, thus making the pub- one worker, and severance pay is particularly lic sector the implicit rule setter for the labor high in Egypt. markets in the region. In addition to advantageous wages and benefits, extremely low risk of dismissal, Important role of the public sector and relatively low demands for productiv- The public sector has traditionally played ity, “double dipping” is a possibility. For an important role in the region, employ- example, evidence indicates that in Egypt ing anywhere between 14 and 40 percent one-quarter of the personnel in public health of all workers (figure O.32). Employment facilities is absent on an average day (Grun, conditions in the public sector (job safety, Etter, and Jillson 2010). As a result of all access to social security, higher wages, and these factors, the overall employment pack- age continues to determine the preference of job seekers, including youth, for working in FIGURE O.32 Average employment shares in the public sector in MENA and selected comparator economies, averages 2000s the public sector. a. MENA economies Effects of generous social insurance 45 systems 40 35 The design of pension systems constitutes a 30 further barrier to mobility in labor markets Percent 25 in the region. Pension systems in MENA are 20 15 relatively new, having recently evolved from 10 covering only civil servants (figure O.33). 5 Their coverage is still limited, leaving about 0 60 percent of the labor force without protec- p. lam o Em p. es Eg Rep r ra c it q ria ud za a an a li bi Un n, Is rocc Ira wa Re e ra Qat at ub Ga rd tion against old-age risk. ge Ar ic R ra r Ku Jo b i iA Al nd o M Notwithstanding the fact that pension ka t, A ab b Sa an yp nA tB systems in MENA notionally have a favor- Ira d ite ria es W Sy able ratio of contributors to retirees, they 45 are increasingly unsustainable financially. b. Comparator economies 40 In part, the generosity of benefits drives the 35 imbalance between contributions and liabili- 30 25 ties of the system. Internal-rate-of-return Percent 20 computations indicate returns between 6 and 15 17 percent, significantly higher than those 10 5 of other investment instruments. With such 0 high returns, workers would want to partici- sia ey M o ia il ile ain d Hu a y a pate in pension systems if they could. This az i in ar ic lan ys ar rk Ch ne ex Sp Ch ng Br lg ala Tu Po do M Bu indirect evidence suggests that in many cases In lack of participation in social insurance is due to exclusion rather than to opting out. Source: Laborsta 2012. With a large share of the private sector Note: MENA = Middle East and North Africa. engaged in low-value-added production, OVERVIEW 23 FIGURE O.33 Percentage of labor force contributing to social security in MENA and other regions, 2000–07 a. World regions b. Selected MENA economies Libya MENA non-GCC countries Egypt, Arab Rep. Tunisia Algeria Developed countries Iran, Islamic Rep. Economies Lebanon Region Europe and Central Asia Jordan Latin America and the Caribbean Morocco East Asia and Pacific West Bank and Gaza Iraq South Asia Syrian Arab Republic Sub-Saharan Africa Yemen, Rep. 0 20 40 60 80 100 0 10 20 30 40 50 60 70 Percent Percent Source: Gatti et al. 2012. Data on social security contributions in GCC countries are available only for Bahrain, 2007 (20 percent), and Qatar, 2008 (4.4 percent) (Pallares-Miralles, Romero, and Whitehouse 2012). Note: GCC = Gulf Cooperation Council; MENA = Middle East and North Africa. social security contributions often become incentives to cultivate relevant skills can unaffordable for firms, resulting, in the be distorted. In particular, in moving from aggregate, in a limited contributory base for school to work, youth in MENA need to pension systems (thus further increasing pen- make a successful “double transition.” sion liabilities) and making public sector jobs First, they need to obtain skills, competen- all the more desirable. cies, and credentials of sufficient quality to Overall, by segmenting the market into become employable; second, they need to a small set of protected insiders and large position themselves in a labor market char- pools of outsiders who have little or no pro- acterized by poor signaling and substantial tection and security, the relatively rigid wage segmentation. structure, overall high social security con- tributions and strict dismissal constrain the Challenge of becoming employable development of a healthy, dynamic labor market. Over the past decades, MENA countries have significantly expanded access to edu- cation, with substantial growth in enroll- Skill gaps and asymmetric ment in secondary and tertiary education information in job search (see World Bank 2008). However, employ- In economies where (1) the private sec- ability—defined as the capital of skills, tor has limited dynamism, (2) the public competencies, academic certificates, and sector offers attractive employment con- professional qualifications, as well as the ditions, and (3) relatively rigid labor regu- capacity to function in a job—remains a lations maintain labor market divides, the challenge in the region. 24 JOBS FOR SHARED PROSPERITY The quality of learning in MENA as learning; (2) poor links between the edu- measured by international standardized cational system and the private sector; and tests, for example, is still below the level (3) the perception that the public sector is still expected given MENA countries’ per cap- the main client of the educational system. ita income (figure O.34). At the same time, evidence points to pervasive mismatches of A prevailing logic of selection skills. Compared to fi rms elsewhere in the A “logic of selection”—manifested in world, more fi rms in MENA contend that rigid tracking in secondary education and inadequate labor force skills, both techni- high-stakes examinations — dominates cal and soft, impede growth (fi gures O.35 educational systems instead of a “logic of and O.36).10 learning.” An ample literature shows that Three key factors rooted in the incen- early tracking and streaming11 can have tives prevalent in MENA’s training and negative consequences for subsequent edu- educational systems contribute to these out- cation and labor market outcomes, particu- comes: (1) a logic of selection rather than larly for pupils from lower socioeconomic backgrounds who tend to perform less well in early selection (Ireson, Hallam, and Hur- FIGURE O.34 Quality of education as measured by TIMSS and PISA, ley 2005; OECD-PISA 2006; Jakubowski et 2011 and 2009 al. 2010). In particular, one disadvantage of a. 2011 highly stratifi ed systems is that transition- ing from a lower to a higher track is dif- 600 ficult and thus not common. In educational TIMSS mathematics scores, 2011 systems in MENA, tracking happens rela- tively early in pupils’ lives, which substan- 500 tially limits transition pathways and viable United Arab second chances. For example, in Egypt, Lebanon Emirates Tunisia 400 Iran, Islamic Rep. Bahrain Jordan Qatar only 5 percent of graduates from technical Saudi Arabia Morocco Syrian Arab Republic Oman secondary schools transitioned into postsec- ondary vocational education in 2008–09, 300 down from 8 percent in 2003–04. The low 7 8 9 10 11 quality of the training provided by the tech- Natural log GDP per capita 2010 nical and vocational education and training b. 2009 (TVET) system reinforces the low perceived status of technical training and exacerbates 550 PISA mathematics scores, 2009 the effects of tracking. As such, this type of system perpetuates and reinforces exclu- 500 sion from labor market opportunities and 450 United segmentations. Arab Emirates 400 Private sector links with the educational Jordan and training sector Tunisia Qatar 350 When the private sector and the educational and training sector operate in isolation, 8 9 10 11 information and signaling failures occur on Natural log GDP per capita both sides, resulting in skill gaps and mis- World Middle East and North Africa Fitted values matches (IFC and ISDB 2011; World Bank 2008). In MENA, educational and training Source: TIMSS and IMF World Economic Outlook (database), January 2013 update. systems lack the information to respond to Note: For panel a, the GDP per capita of year 2010 is an IMF estimate, not final figures, for Botswana, Chile, Georgia, Ghana, Honduras, Lebanon, Oman, South Africa, the United Arab the needs of the private sector, whereas the Emirates, and the United Kingdom. GDP = gross domestic product; MENA = Middle East and private sector lacks the capacity or the inter- North Africa; PISA = Programme for International Student Assessment; TIMSS = Trends in International Mathematics and Science Study. est in playing its role in a demand-driven OVERVIEW 25 skill development system. This miscom- FIGURE O.35 Share of firms identifying inadequately educated munication is a particular concern for the workforce as a major constraint to growth, by region and selected TVET system, in which employers have a countries, 2005–11 crucial role in ensuring that skills acquired through that system are relevant for the 40 labor market. 35 Perception of the public sector as the main 30 client of education and training The public sector continues to be the main 25 client of educational and training systems Percent 20 and thus the main shaper of students’ choices and expectations. Some countries 15 have already undergone structural adjust- ment (Morocco, for example), and others 10 have considerably slowed public sector hir- 5 ing (Egypt). Even so, the fact that govern- ments continue to play an important role 0 re y a p. a in providing jobs in MENA and that public Ge ain a sia co ica nt ean sia e O an ific an Car ca ut s d ld So rie Ko man Re pe e Afri hA gh Su ia an Wor r lA c Af t Sp d ibb co Sah Pa un sector jobs (although more scarce) are still a, ra r Eu and orth ar Ce D the most coveted by youth distorts incen- N EC ro th ica d As er an -in b- tives for skill formation. Students strive to st Am ast m Ea tin e E attain degrees that fit with public employ- La iddl M ment, but those degrees are increasingly irrelevant to the private sector, leaving grad- hi Region or country uates without the skills demanded by the Source: Investment Climate Assessment (ICA) Enterprise Surveys (global dataset) 2012. market. In Tunisia, for example, about half Note: OECD = Organisation for Economic Cooperation and Development. a. 2005. of university graduates opted to study the humanities and social sciences, curricula FIGURE O.36 Mismatches in hard and soft skills of newly hired favored for entering the public sector (fi g- graduates in selected MENA countries, 2010 ure O.37). A recent tracer study in Tunisia found that almost 50 percent of graduates in the humanities and law had still not found 50 a job 3.5 years after graduation, while more than 80 percent of medical school graduates 40 had found a job by that time (Ministry of 30 Percent Employment and Professional Insertion of Youth and World Bank 2009). 20 10 A meritocracy deficit in the labor market? 0 p. . a n co ep bi a Re oc rd ,R ra The increasing demand for transparency b Jo or en iA ra M m ud t, A and equality of opportunity in accessing Ye Sa yp Eg jobs heightens the importance of merit and clear rules for job seeking and hiring. University graduates’ hard skills University graduates’ soft skills Young people and their families have high Vocational graduates’ hard skills Vocational graduates’ soft skills expectations for the future, they invest heavily in education and skills, and they Source: IFC and ISDB 2011. expect these investments to pay off even- Note: MENA = Middle East and North Africa. Numbers show the percentage of human resource tually. Yet from what students, graduates, managers agreeing that graduates hired in the past year have the appropriate skills. 26 JOBS FOR SHARED PROSPERITY FIGURE O.37 Distribution of university graduates by field of study of job opportunities, putting those from less in MENA economies, 2004–10 affluent families at a disadvantage. Professional services can help fi rms tap a large pool of talent and find the best matches West Bank and Gaza Qatar in skills and competencies for any given Lebanon vacancy; yet fi rms in MENA make limited Algeria use of such recruitment practices. Indeed, Saudi Arabia non-GCC countries score the lowest com- Jordan pared to other regions in the meritocracy of Oman United Arab Emirates hiring (see figure O.39). In contrast, GCC Iraq countries score extremely high, probably Morocco because of the large recruitment of expatriate Tunisia workers. Iran, Islamic Rep. The prevalence of informal job-matching Djibouti 0 20 40 60 80 100 methods is consistent with a number of Percent explanations. Small or family- ow ned fi rms—which are the norm in the region’s Scientific, technical, and engineering Health private sector—might have little or no incen- Education and humanities Social sciences tive to invest in searching wide pools of tal- Other ent. The structure of product markets (little competitive pressure) or the pervasiveness Sources: Edstats (database); Tunisian data from L’enseignement supérieur en chiffres, Année univer- of arbitrary rule enforcement, which makes sitaire 2010–2011; Bureau des Etudes, de la Planification et de la Programmation. Note: MENA = Middle East and North Africa. Data from Tunisia refer to students enrolled during the other qualities such as trust more relevant, academic year 2010–11. could also dictate this choice.12 Although improvements in competition and transpar- ency are likely to increase incentives to fi nd and employers in MENA countries report, better matches, the infrastructure and regu- it is clear that educational credentials are lation that could increase information flows widely perceived to play a minor role in are also lacking. employers’ hiring decisions. For example, while in Egypt, Tunisia, and, to a lesser Limited role of formal labor extent, Syria, young people identify the intermediation lack of good jobs as the main constraint to being hired, a substantial share of youth in The widespread use of informal networks MENA thinks that jobs are given to con- is a sign of information asymmetry and of nected people (figure O.38). signaling failure. Information asymmetries When formal degrees have limited signal- in labor markets can be overcome to some ing value, other factors (such as trust and extent through intermediation services, and personal connections) can dominate job public and private employment agencies can search and hiring decisions. Since individual play an important role in job matching. For ability is unobservable, informal networks employers, these agencies facilitate contacts can be efficient in resolving the information with job seekers and even provide assistance asymmetries in labor markets. Anywhere with screening and selection. In non-GCC between 40 percent (in Yemen) and 80 per- MENA countries, public intermediation cent (in Lebanon) of private sector employ- services are still relatively new and under- ees in countries with available data report developed. Most display low capacity and having found their job through friends or effi ciency, with public employment service relatives. However, since informal networks workers carrying excessive caseloads and usually develop within limited socioeco- adopting a generally passive approach to job nomic strata, they can affect the distribution matching. OVERVIEW 27 What policy options? FIGURE O.38 Perceptions of youth in selected MENA economies of the constraints to getting a job, 2009 The employment challenge in MENA is structural. The previous sections underscored how the rules that govern the distribution of Libya growth dividends—including limited compe- Morocco tition in the private sector and restrictive reg- Qatar ulation of labor markets—have also limited Tunisia economic growth and employment creation. Egypt, Arab Rep. To move M E NA towa rd a h ig her- Kuwait productivity equilibrium with more and Saudi Arabia better-quality jobs (figure O.40), a medium- Djibouti United Arab Emirates term agenda is needed to shift the current West Bank and equilibrium along two dimensions: a more Gaza Jordan efficient and fluid allocation of resources Lebanon and a more equitable access to opportuni- Syrian Arab Republic ties. Realigning incentives for fi rms to invest Algeria and grow will be essential to unlocking the Yemen, Rep. potential of the private sector. At the same Iraq time, a rebalancing of employment protec- Bahrain tion regulation and incentives in the public 0 20 40 60 80 100 sector will be needed to promote a more Percent dynamic labor market and to capture the Jobs given based on No good jobs benefits of growth in greater job creation connections available more fully. Finally, better governance will Lack of training Other be needed to equip educational and training systems to respond to the new signals com- ing from the labor market. Source: Based on Gallup World Poll Survey 2009. Note: MENA = Middle East and North Africa. Aligning incentives to invest, innovate, material inputs) is likely to stimulate job and generate employment creation. Since tax and subsidy schemes in With burdensome business regulations and MENA place labor at a disadvantage, reduc- discretionary enforcement, poor access to ing labor taxes or reducing subsidies on credit, and subsidies that distort energy other factors of production should create prices, the private sector—particularly more employment opportunities and, in the the high-productivity segment—is subop- medium term, accelerate growth through timally small and lacks the incentives to better allocative efficiency and innovation. diversify, innovate, and invest. These con- These actions would free fiscal resources to straints can be addressed, among other further stimulate job creation. ways, by lifting barriers to competition, The current subsidy system favors energy- making access to production inputs more intensive production processes that rely on equitable, and promoting a culture of outdated technology. Ensuring that firms entrepreneurship that encourages experi- internalize the true cost of energy would mentation and learning. enhance competitive pressure and stimulate the adoption of more advanced, energy- Removing distortions that repress labor efficient modes of production. Such a reform demand would level the playing field for small and Reducing the cost of labor relative to other medium enterprises and private sector firms, factors of production (such as capital and given that their competitors (large and public 28 JOBS FOR SHARED PROSPERITY FIGURE O.39 World scores on proxies of meritocracy in firms) currently benefit disproportionately hiring, 2011 from energy subsidies. Without a doubt, the removal of subsidies would have political and economic reper- 6 cussions. To create the political support to 5 enact and sustain reform, governments must compensate the losers adequately and cred- 4 ibly and sustain their transition to more Score 3 energy-efficient production. Another option 2 for simultaneously increasing employment and fiscal space would be to reexamine 1 agricultural subsidies. For example, remov- 0 ing wheat subsidies (which account for C) C) an sia ific es ca sia GC a large share of agricultural subsidies in GC be tri fri lA ac hA un nA n- a( rib dP ra MENA) could have important impacts on ut no ric co nt Ca ra an So Ce a( Af ha CD he agricultural employment. Wheat produc- ia ric th Sa nd OE As dt Af or b- ea an st tion is relatively capital intensive, unlike the dN th Su Ea op ica or an r dN Eu er production of other crops, such as olives, st Am an Ea which require more labor. In lieu of pro- le st tin idd Ea La le viding wheat subsidies, governments could M idd M fi nance investments to increase agricultural Region productivity. The removal of energy and agricultural subsidies could have especially Source: Based on Executive Opinion Survey. powerful effects in labor-abundant coun- Note: GCC = Gulf Cooperation Council; OECD = Organisation for Economic Co-operation and tries such as Egypt and Syria. Development. See World Economic Forum (2011, 75) for a detailed description of the Global Executive Survey. Most questions in the survey ask respondents to evaluate, on a scale of 1 to 7, one particular aspect of their operating environment. One represents the worst possible situation; seven represents the best. Enabling firms to compete and invest Reduce the costs of entry, exit, and adjustment Creating more jobs requires existing fi rms to expand, new fi rms to start operations, and, paradoxically, the destruction of FIGURE O.40 Toward more and better-quality jobs in MENA unproductive fi rms that use inputs ineffi - ciently. All of these developments become more feasible when competitive pressure is high unleashed and contestability is enhanced. Efficient and fluid allocation of resources equilibrium more and Most M E NA cou nt r ie s , pa r t ic u la rly better-quality jobs those with burdensome business environ- ments, have considerable scope for adapt- ing business regulations to encourage fi rm entry and facilitate the exit of unproduc- low how to shift from tive firms (through more efficient and equilibrium rapid bankruptcy procedures, for exam- low to high equilibrium? ple). In addition, access to many markets MENA now remains restricted. For instance, the legal or accounting professions in Morocco and Tunisia are subject to licensing processes Equality of opportunities; even playing field that are not clearly spelled out and leave room for discretion. OVERVIEW 29 Create a level playing field access to credit. The limited development The case for regulatory reform is especially of the financial sector reflects a stark pronounced in countries where discretionary absence of legislation, probably owing enforcement of regulation has been used more to insufficient interest among incum- as a vehicle for rent distribution. Institutional bent institutions in widening access to reforms that improve the accountability of the credit. A new set of policies focused on public administration can foster a consistent strengthening financial infrastructure, implementation of rules and regulations and increasing bank competition, and devel- may include several elements. First, transfer- oping nonbank fi nancing (such as venture ring responsibilities and decision making to capital) would not only improve access to lower tiers of the public bureaucracy while fi nance in the MENA region, but also lay making (lower-tier) civil servants accountable the foundation for more sustainable eco- to a broader base could enhance accountabil- nomic growth.13 ity and reduce the scope for opportunistic behavior. Second, it is important to increase the capacity of the lower ranks of the bureau- Rebalancing the social contract to cracy. Third, restructuring recruitment and promote more dynamic labor markets promotion schemes in public administra- How can countries in MENA foster labor tions based on merit or commitment to a markets that give all groups an equal development strategy instead of regional and chance, enable the labor force to move sectarian considerations would progressively toward the most productive sectors, and reduce incentives for rent seeking. eventually support a dynamic private sec- In parallel, empowering antitrust agencies tor? This question is not unique to MENA, could also effectively bolster competitiveness. as ma ny ot her econom ie s , i nclud i ng Recently established authorities on com- advanced ones, are grappling with similar petition would benefit from clear-cut legal issues. While the answer must be coun- autonomy from the executive branch, which try specifi c, striking a productive balance currently approves decisions on competition between flexibility and the protection of cases while presiding over the very public workers will be key to promoting inclu- enterprises that often maintain monopolistic sive and dynamic labor markets. Attaining positions. this balance will involve decisions about the minimum wage, wage agreements, Promote trade to relax demand constraints and public sector wage scales. It will also Opening up economies to trade can also involve decisions about the regulation of enhance competition, especially if coupled contracts, including hiring and dismissal with an improvement in the investment cli- procedures and working hours in both the mate. Increased openness could pay a double private and the public sectors. Finally, it dividend, because tapping into the global will involve decisions about the character- demand pool would be an effective way of istics of social insurance systems. complementing domestic demand, which is To achieve inclusiveness, move toward a manifestly insufficient. It would also spur higher-productivity equilibrium, and fos- productivity growth both by serving as a ter a more equitable distribution of oppor- disciplining device for inefficient fi rms and tunities in labor markets, these decisions by allowing firms to capitalize on knowledge need to rest on a broad-based consensus spillovers. that involves the whole citizenry. In this sense, not only the regulatory framework Expand access to credit to help firms grow around labor markets but also the social and invest dialogue needs to be rebalanced, so that Another precondition for the efficient func- those who have been traditionally outside tioning of markets is fair and equitable the decision-making process— informal 30 JOBS FOR SHARED PROSPERITY workers, youth, the unemployed, and contract employees. Emerging examples women—can participate in a bargaining of newly designed unemployment insur- that, through wage and rule setting, affects ance systems include the Jordanian Unem- them directly. Several specific measures ployment Insurance Savings Accounts: could improve the design of social insur- such accounts reduce work disincentives ance and labor regulation in MENA: by allowing recipients to keep their own unused unemployment contributions and • Wages. Algeria, Tunisia, and, to a lesser ex- offer the possibility of extending coverage tent, Jordan have relatively well-organized, to informal sector workers. traditional social partners (trade unions, • Pensions. Many countries will need to employers, parliaments, and the govern- redesign their pensions systems. Para- ment), but they do not yet include rep- metric reforms or shifts to defi ned con- resentation of new social partners (civil tribution systems could prevent pension society organizations, youth, women, debt from spiraling up. At the same time, the unemployed, and informal workers). social partners are increasingly demand- Opening up the dialogue on collective ing greater coverage, something that could wage agreements might result in a more substantially close the divide between inclusive wage-setting process. system insiders and outsiders. Sequenc- • Contract provisions and worker protec- ing will be very important to success, as tion. Most countries in the region would reforms to improve the sustainability of benefit from reviewing their employment current systems will need to be enacted legislation and instituting more mod- before coverage can be extended. erate worker protection (especially for • The role of the public sector. It is unlikely dismissal, which includes a rethinking that countries such as Egypt, Jordan, or of the strong divides between fi xed-term Syria will inject much dynamism in labor and open-ended contracts), combined markets or promote incentives to cre- with better income security through ate better jobs without a reform of civil provision of effective unemployment service employment. Most countries insurance. will need to review the risks and returns • Unemployment insurance. In labor mar- offered by the employment package in the kets with less rigid and less costly dis- public sector by revisiting career and per- missal regulations, appropriately designed formance incentives, accountability mech- unemployment insurance schemes can anisms, and wage and job security. provide adequate protection to workers. • Barriers to women’s participation in With the gradual disappearance of life- labor markets. Explicit attention is long jobs and an increasing need for job needed to reduce the barriers to female mobility, income protection for workers labor market participation and assist in transition between jobs will be increas- those women who wish to enter the labor ingly important. Only a few countries in force. Policies that could help include the region have unemployment insurance ensuring that women can travel to work systems (Algeria, Bahrain, Egypt, the safely and that the workplace itself is Islamic Republic of Iran, and Kuwait). safe; increasing the supply of and access However, even the systems that are in to affordable care for children; reform- place are underused owing to a lack of ing and improving parental leave poli- public awareness, restrictive eligibil- cies without reducing incentives to hire ity conditions, the difficulty of and the women (such policies also have several stigma attached to documenting a “just- positive externalities); increasing women’s cause” firing decision, and low overall capacity to start and run their own busi- layoff risks among covered open-ended nesses and obtain credit;14 and revising OVERVIEW 31 laws that still limit women’s access to the large information and knowledge gaps productive assets.15 related to learning outcomes and the sup- ply of and demand for skills must be closed. This effort entails monitoring and evalu- Realigning incentives for skills that ating the quality of education and, at the matter same time, reforming assessment and cer- In a region that generates far too few jobs, tification systems. Addressing knowledge and far too few high-quality jobs, and where and information failures, including through merit influences the allocation of jobs to counseling throughout secondary school, only a limited extent, the educational and will allow employers to better communi- training systems receive scrambled signals cate what they require from the educational that do not emphasize skill acquisition as system, schools, and educators to improve the route to employment. Instead, the sig- quality and to better assess and certify nal is to acquire the “right” degree from the learning, and families and students to make “right” university and then queue for a pub- better decisions and form realistic expecta- lic sector job. The results are little pressure tions about the transition from education to to change traditional patterns of pedagogy work. Overall, more systematic, accurate, (which maintain high selectivity and rigid and publicly available information on what tracking), a declining TVET system, and is accomplished by education and training eventually a system with relatively poor out- will lead to more solid accountability of all comes at the end of the compulsory school- institutions and actors involved. ing cycle. The key to increasing productivity and Value learning and problem solving restoring a sense of dignity and agency The ultimate challenge for education in the among MENA’s young people is for educa- 21st century is a paradoxical one: to design tional and training systems to become more a race where everybody can win and where accountable to people, more responsive to the everyone has multiple ways to win. Inter- private sector, and more capable of promot- national experience shows that this goal ing labor markets that value and recognize can be achieved by making educational individual skills and competencies in a trans- and training systems, particularly those at parent, merit-based manner. If those goals the secondary and TVET level, more inclu- are met, the incentives of students and educa- sive, more directed toward learning, and tional systems will be reoriented toward the less directed toward selecting and exclu- acquisition and provision of more relevant sively rewarding the academically able skills, provided that competition in the pri- while leaving the rest behind. The priority vate sector becomes stronger, a level playing is to prevent students from dropping out field is created between public and private and to identify alternative training itin- employment, and labor markets in general eraries so that everyone can acquire rel- are revitalized. At the same time, a number of evant skills for a successful school-to-work institutional features can affect the ability of transition. educational and training systems to produc- tively capture and respond to these changes. Empower the private sector and realign Closing stakeholders’ information and incentives for public sector hiring knowledge gaps Making employability count through a More access to education and training is renewed partnership between educational not enough to realign incentives to teach and training systems and the private sec- and acquire the most relevant skills. To tor will help align incentives. A new and know which skills are required, and where, more effective public-private partnership 32 JOBS FOR SHARED PROSPERITY framework in education and skill develop- process? And what strategies can help coun- ment should help create a common language tries face the post–Arab Spring challenges and incentives for addressing the signaling in the short term? failures from both sides. A complex historical legacy Improve efficiency through active labor market policies Historically, the social contract in many A better balance between publicly funded countries in the region anchored the sup- and private intermediation services will port of often autocratic, long-lived regimes likely improve job search processes and lead to a system of rent sharing among minority to better matches between labor demand coalitions. In this system, elites benefited and supply. Some countries may also need from rents derived from their leading posi- to adopt legislation to enable and regulate tions in organizations that controlled certain private intermediation services. Most coun- sectors of the economy or from rich natural tries will need active labor market programs resources. To maintain their dominant posi- (that is, programs aimed at improving ben- tion, elites used different strategies to limit eficiaries’ employment prospects or earn- access to these rents, including discretion- ings capacity) for those who falter in the ary application of regulation and signifi- transition from education to work and for cant labor market rigidities. For example, serving the workers who are most difficult the lack of access to credit limited the entry to place or most in need of improving their of new firms and increased privileges for skills. Although active labor market pro- incumbent fi rms. grams currently serve urban areas and the At the same time, elites constrained the unemployed (some of whom can afford to ability of other social groups to challenge be unemployed) almost exclusively, they will their dominant position by restricting the need to reach out to rural areas and infor- right of association, press freedom, the role mal workers. To increase effectiveness, ser- of civil society organizations, and access to vices must also be rationalized and focused data and information. In many countries, on case management. Finally, in all coun- this combination resulted in narrow-based tries, better measurement and tracking of social dialogue that limited accountabil- effectiveness—for example, of the insertion ity to the majority of the population. This rate (that is, the proportion of program ben- system reinforced divides both within the eficiaries who fi nd employment) — and feed- private sector (where a few privileged fi rms back mechanisms from beneficiaries and the thrive while many struggle to survive and fail private sector will contribute to improved to grow) and within labor markets (where performance. only a minority of insiders enjoys gainful employment). The stability of this equilibrium relied also The political economy of on elites’ ability to share these rents selec- inclusion tively through subsidies and public employ- The analysis so far has shown that breaking ment. For example, public employment has the cycle of privilege and lifting the barri- been actively used as an instrument for guar- ers to more and higher-quality employment anteeing secure jobs for middle- and upper- will involve a complex set of reforms. Yet class university graduates and subsidies for the question of jobs in MENA is not new, ensuring a minimum standard of living for and many of the “technical” recommenda- the whole population. However, such heavy tions discussed so far are well known. If interventionism in the economy and in labor this is the case, why have these reforms markets has muted incentives for the private not been enacted before? Conversely, what sector to grow, innovate, and generate high- could finally trigger the needed reform quality jobs. OVERVIEW 33 Looking ahead in post–Arab Spring could concentrate on generating early gains MENA that both respond to immediate concerns and are compatible with a medium-term agenda The powerful demand for democracy and of inclusive growth. Figure O.41 summarizes voice of the Arab Spring carry unprece- the links among enabling conditions that dis- dented opportunities to move toward a more rupt the low-productivity equilibrium, poten- inclusive development model. Engendering tial short-term actions, and reform objectives. true change in the rules of the game will be challenging, however. Governments are Leveraging new political forces for an under tremendous pressure to deliver results inclusive dialogue rapidly, which might lead to populist, fis- A new political landscape continues to shape cally costly measures that continue to rein- the region, affecting the roles of both tra- force existing privileges. Long-time outsiders ditional social partners (trade unions and might see the current political changes as employers, for example) and new social their opportunity to fi nally become insiders, partners (civil society organizations repre- which would merely imply an alternation senting outsiders such as youth and women, of elites. Moreover, evidence indicates that the unemployed, and informal enterprises). it is especially difficult for new democracies An inclusive social dialogue that gives these to credibly commit to policies with broad- new partners a voice can help build a shared based benefits at the expense of highly orga- view of labor market challenges and con- nized small interest groups, something that sensus around reforms that are beneficial to can be even more daunting amid the current all. Such reforms include a minimum wage drop in growth and fiscal revenues. that protects workers without harming the Yet the Arab Spring is testimony to a pow- job entry chances of others; collective wage erful demand for change, one that has toppled agreements that balance inequality concerns decades-old autocratic regimes. It is, how- with the entry chances of inexperienced ever, important to accompany the progressive youth; the establishment of the features of forces that are working to disrupt the status a social insurance system that can progres- quo with measures that facilitate transition- sively cover the entire workforce; or the ing to a new equilibrium—that is, measures improvement of governance in educational that build credibility for the new govern- and training institutions. ments and mitigate the future cost of reforms. Given the pressure that governments receive Improving data openness to repeat “more of the same,” it is important Access to data and information is crucial for that these transitory measures be compatible accountability, transparency, and effective with the new midterm agenda of reform. citizen participation in policy formulation. The following sections explore three com- How should governments open up access to plementary approaches that would allow gov- data? First, they can eliminate legal barriers ernments to respond to short-term challenges that prevent the public from accessing ano- while preparing the ground for more struc- nymized microdata.16 A number of coun- tural, medium-term reforms. First, improv- tries in Latin America and Eastern Europe ing access to data and information and have made important strides in this area; in allowing for public expression would enable MENA, countries such as Tunisia are pro- governments to inform the people about the viding a leading example that things can necessity of reform and to improve the com- change. Second, building a critical mass of mon understanding of key issues. Second, expertise in and outside of statistical agen- leveraging the dialogue with newly emerg- cies will allow for consistent use of data for ing social forces could change the nature of policy making and will support the creation the policy dialogue from a narrow base to a of a shared view of policy priorities. In other more inclusive process. Finally, governments countries, institutions such as labor market 34 JOBS FOR SHARED PROSPERITY FIGURE O.41 Linking short-term measures to medium-term objectives Disrupting the low equilibrium Generating credible interventions with early Reform objectives measurable gains • Making access Increase competition, to credit more reduce privilege democratic • Youth employment Leverage new social Rethinking programs, Early forces for a renewed and the social Childhood Rethink the social contract inclusive social dialogue contract Development interventions • Giving voice to service clients Improve governance in skills (like parents systems and students) observatories have contributed productively to the labor force. A developed industry to developing this type of expertise. for child care education could directly cre- ate a large number of jobs among women, Generating early measurable gains while also allowing many more parents Governments in MENA are now operating (young mothers) to be economically active. in a fluid and complex political landscape. In addition, international evidence (MIT Investing in early measurable results will help 2005) suggests that high-quality early consolidate their credibility and build the ter- childhood education and care help prepare rain for the game-changing reforms to come. young children to succeed in school and Examples of these interventions follow: eventually in life. • Filling important infrastructure gaps for • Targeted programs to improve opportu- direct and indirect job creation. Invest- nities and employability of young people. ment in infrastructure could be a natural Instead of creating more permanent civil “early gains” measure for creating jobs in service jobs in response to popular pressure, MENA. Depending on design, this inter- rigorously evaluated publicly funded youth vention can have high labor content and and “cash for work” programs can contrib- play a strategic role in connecting people ute to employment creation in the short run to markets. and promote skills building as a bridge to • Reform credit regulations to benefit future employment, when designed accord- credit-starved small firms. Evidence sug- ing to international best practices. gests that in the short run, increased • Early childhood development interven- access to fi nance, including microfi nance, tions to facilitate female participation in can foster new business growth and OVERVIEW 35 especially investment growth among the This report has shown in detail how the credit-starved fi rms. This reform would incentives for formal private sector job cre- require the political will to increase ation, productivity growth in fi rms, skills competition between banks, allow new formation, and the efficient allocation of entries, and reduce state ownership of resources and human talent are stifled in financial institutions. While the imme- MENA. Not only has this curbed the dyna- diate gains in job creation might not be mism of the economy, but has also produced large, improving access to credit could inequitable outcomes that have been espe- provide an important signal of the govern- cially detrimental to women, young people, ment’s commitment to fostering a more and the poor. The public sector and the state open and inclusive business environment. have a vital enabling role to play, but this is By improving the conditions of fi rm entry, a very different role from that seen in the access to credit could also endogenously majority of the region’s countries at present. expand the constituency for change and Part of the underperformance in MENA reform in the private sector. is grounded in the complex political econ- • Produce tangible signs of improved omy of the region, which also explains the social services to citizens. The demand inability of countries to embrace the reforms for greater inclusion can also be satisfied that could really change their development by helping citizens monitor the delivery trajectory. The Arab Spring disrupted the of local services and in this way improve political equilibrium in the region and many service providers’ performance (Reinikka countries are undergoing important and and Svensson 2004). For instance, in complex transitions. This presents a unique education, countries in other regions are opportunity to break the system of privileges providing greater autonomy to schools and move toward a new and more inclusive and encouraging stronger local scrutiny model of development. However, the new from parents through the empowerment governments will need to walk the fi ne line of parents’ councils. Publication of data between responding to political pressure of on performance and resource flows and different groups and building ownership for the use of scorecards that allow citizens to reform. evaluate delivery are additional methods Many structural reforms will be needed of increasing accountability. to unburden countries from the legacies of the past and promote job creation. These include creating a level-playing field among Conclusion: Time for action in the fi rms, establishing a fair de facto investment Middle East and North Africa climate, extending access to credit, rebal- With their natural resource wealth, young ancing employment conditions between the and increasingly well-educated populations, private and the public sector, improving and strategic location near mature markets, governance in education and training sys- countries in the Middle East and North tems, and removing energy subsidies while Africa have tremendous potential. However, protecting the poor and vulnerable. the great promise these countries hold has The time for action is now. Global expe- yet to be realized. rience has shown that it is possible to pro- A comparative analysis of the region’s duce tangible improvements in the lives of employment dynamics suggests that the citizens in the short run without resorting development model adopted by many coun- to populist policies that might be harmful tries in MENA has failed to deliver enough to the economy in the longer run. Improved good quality jobs. Demographics and global social service delivery, short-term employ- economic transformations more than ever ment and skills development program with make a dynamic private sector the indis- well-designed temporary interventions for pensable factor for sustained job creation. youth and the poor, and well-functioning 36 JOBS FOR SHARED PROSPERITY social safety nets are but a few of the policies ability grouping in selected curriculum areas that can deliver early, visible and measur- or subjects. able results. They can help new governments 12. The use of informal networks for hiring is less gain credibility, while a more open dia- prevalent among firms with foreign or mixed ownership. In Tunisia, for example, 34 percent logue with the population, strengthened by of employees with a university degree work- open access to public data and information, ing in private firms with Tunisian ownership builds consensus for the needed, longer-term report having found their job through friends structural reforms. or relatives compared with 24 percent in And if not now, when? firms with foreign or mixed ownership (based on a Graduate Tracer Survey of the gradua- Notes tion cohort of 2004). 13. World Bank (2011a). 1. The expected wage is predicted based on a 14. See World Bank (2010). simple Mincerian regression. Interestingly, 15. World Bank (2011b) discusses these issues in wage differentials persisted even when the depth. direct measure of ability was accounted for 16. Microdata is information at the level of indi- (in Lebanon and Syria, where available data vidual respondents. Anonymized means that made this possible). it is not possible for the data user to track the 2. For a discussion on agglomeration, see World information back to individuals, that is, to Bank (2010). identify the identity of individuals in the data. 3. Protected is defined as having social security, and high-paying is defined either as yielding a wage above two-thirds of the wage distri- References bution or as self-employed and belonging to Abouleinein, S., H. El-Laithy, and H. Kheir- a household in the top two income quintiles. El-Din. 2009. “Phasing Out Energy Subsidies 4. This definition leaves out pockets of the work- in Egypt.” Working Paper 145, Egyptian ing poor in urban areas, but the prevalence of Center for Economic Studies, Cairo. poverty is not as correlated with being low- Angel-Urdinola, D., and K. Tanabe. 2012. “Micro- skilled in urban areas as it is in rural areas. Determinants of Informal Employment in the 5. The growth strategy during these decades Middle East and North Africa Region.” Social was financed mainly by high oil revenues Protection and Labor Discussion Paper 1201, in resource-rich countries and by workers’ World Bank, Washington, DC. remittances and public borrowing in the Angel-Urdinola, D, A. Kuddo, and A. Semlali. resource-poor countries. 2013. Public Employment Services in the 6. Changes in the drivers of growth, and par- Middle East and North Africa. Washington, ticularly in total factor productivity growth DC: World Bank. rates, conceal stark differences between sectors Barros, R., F. Ferreira, J. Molinas Vega, and within the same country, which warrant a more J. Saavedra. 2008. “Measuring Inequality detailed country-level analysis to better identify of Opportunities in Latin America and the and determine the impact of this change. Caribbean.” World Bank, Washington DC. 7. See also discussion in Gatti et al. (2012). Bodor, A., D. Robalino, and M. Rutkowski. 8. As analyzed in Diop, Marotta, and deMelo 2008. “How Mandatory Pensions Affect (forthcoming). Labor Supply Decisions and Human Capital 9. World Bank (2011, fig. 4.8, 103). Accumulation? Options to Bridge the Gap 10. Soft skills include teamwork, leadership, and bet ween E conom ic T heor y and Polic y problem solving. Analysis.” MPRA Paper 12046, University 11. Tracking and streaming are alternative Library of Munich, Germany. denominations of similar practices of the abil- Bureau des Etudes de la Planification et de la ity grouping of students. While tracking usu- Programmation. 2010, 2011. Répartition des ally refers to totally different study programs étudiants par domaine d’étude (CITE). [Tunisia for different groups of students of the same Ministry of Higher Education and Scientific age (for instance, vocational and educational Research, Higher Education in Numbers training programs in upper secondary educa- 2010 –2011], http://www.mes.tn/francais/ tion versus general ones), streaming refers to donnees_de_base/depliantfr10_11.pdf. OVERVIEW 37 D iop, N . , D. M a rot t a , a nd J . d e M elo. 1999 Education Reform in Poland.” Policy Forthcoming. Natural Resource Abundance, Research Working Paper 5263, World Bank, Growth and Diversification in M EN A . Washington, DC. Directions in Development Series. Washington, Klapper, Leora, and Inessa Love. 2010. “The DC: World Bank. Impact of the Financial Crisis on New Firm EAPEP (Economic Alive Population, Estimates Registration.” Policy Research Working Paper and Projections) database. 2011. 6th ed. 5444, World Bank, Washington, DC. International Labour Organization, Geneva, Laborsta (database). 2012. International Labour Switzerland, http://laborsta.ilo.org/applv8/ Organization, Geneva, Switzerland, http:// data/EAPEP/eapep_E.html. laborsta.ilo.org/. E d s t at s (d at ab a s e). 2 01 2 . World B a n k , Ministry of Employment and Professional Washington, DC, http://go.worldbank.org/ Insertion of Youth and World Bank. 2009. ITABCOGIV1. “Dynamique de L’Emploi et Adéquation de la Enterprise Survey Brazil. 2009. IFC and World Bank, Formation Parmi les Diplômés Universitaires. Washington, DC, http://www.enterprisesurveys Deuxième Rapport.” Tunis, Tunisia. .org/data/exploreeconomies/2009/brazil/. MIT (Massachusetts Institute of Technology). Enterprise Survey Jordan, 2006. IFC and World Bank, 2005. Early Childhood Education for All: Washington, DC, http://www.enterprisesurveys A Wise Investment. Cambridge, MA: MIT .org/Data/ExploreEconomies/2006/jordan. Workplace Center. Executive Opinion Survey. 2011. World Economic Morocco LSMS (Living Standards Measurements Forum, Geneva, Switzerland. Survey). 2001. World Bank, Washington, DC. Gatti, Roberta, Diego Angel-Urdinola, Joana OEC D – PISA (Organisation for E conomic Silva, and Andras Bodor. 2012. Striving for Co-operation and Development–Programme Better Jobs: The Challenge of Informality in for I nter nat iona l St udent A sse ssment) the Middle East and North Africa. Directions database. 2006, 2009. OECD, Paris, France, in Development Series. Washington, DC: http://pisa2006.acer.edu.au/downloads.php. World Bank. Pa l l a r e s - M i ra l l e s , M . , C . Rom e ro , a nd Grun, R., L. Etter, and I. Jillson. 2010. “Arab E. Whitehouse. 2012. “International Patterns Republic of Egypt: Management and Service of Pension Prov ision I I : A Worldw ide Quality in Primary Health Care Facilities in Overview of Facts and Figures.” Social the Alexandria and Menoufia Governorates.” Protection and Labor Discussion Paper 1211, Mimeo. World Bank, Washington, DC. World Bank, Washington, DC. ICA (Investment Climate Assessment) Enterprise Reinikka, R., and J. Svensson. 2004. “The Power Surveys (global dataset). 2012. IFC and of Information: Evidence from an Information World B a n k , Wa sh i ng ton , D C , w w w Campaig n to Reduce Capt u re.” Policy .enterprisesurveys.org. Research Working Paper 3239, World Bank, IFC (International Finance Corporation) and Washington, DC. ISDB (Islamic Development Bank). 2011. United Nations Conference on Trade and Education for Employment: Realizing Arab Development. 2000–07. Geneva, Switzerland, Youth Potential. Washington, DC: IFC. http://unctad.org/en/Pages/Home.aspx. IFC (International Finance Corporation) and Wo rl d B a n k . 2 0 0 4 . Jo b s , G r o w t h a n d World Bank. 2011. Doing Business 2012: Gove r n anc e in M E N A : Unlocking the Doing Business in a More Transparent World. Potential for Prosperity. Washington, DC: Washington, DC: IFC and World Bank. World Bank. ILO-KILM (International Labour Organization– ———. 2008. The Road Not Travelled: Education Key Indicators of the Labour Market) database Reform in the Middle East and North Africa. (data for 1980–2010). 7th ed. International Washington, DC: World Bank. Labour Organization, Geneva, Switzerland. ———. 2009. From Privilege to Competition: http://kilm.ilo.org/kilmnet/. Unlocking Private-Led Growth in the Middle Ireson, J., S. Hallam, and C. Hurley. 2005. “What East and North Africa. Washington, DC: Are the Effects of Ability Grouping on GCSE World Bank. Attainment?” British Educational Research ———. 2010. Poor Places, Thriving People: How Journal 31 (4): 443–58. the Middle East and North Africa Can Rise Jakubowski, M., H. A. Patrinos, E. E. Porta, above Spatial Disparity. Washington, DC: and J. Wisniewski. 2010. “The Impact of the World Bank. 38 JOBS FOR SHARED PROSPERITY ———. 2011a. Financial Access and Stability: World Development Indicators (database). A Road Map for the Middle East and North World Ban k, Wash ing ton, DC , ht t p: // Africa. Washington, DC: World Bank. d a t a .w o r l d b a n k . o r g /d a t a - c a t a l o g / ———. 2011b. Economic Development Prospects: world-development-indicators. Investing for Growth and Jobs. Washington, Wo rld E c o n o m i c F o r u m . 2 011. G l o b a l DC: World Bank. C o m p e t i t i v e n e s s R e p o r t 2 0 11 – 2 0 1 2 . ———. 2013. Opening Doors: Gender Equality G eneva, Switzerland: World E conom ic in the Middle E ast and Nor th Afric a . Forum, http://www.weforum.org/reports/ Washington, DC: World Bank. global-competitiveness-report-2011-2012. Labor Markets in the PART 1 Middle East and North Africa: A Low-Productivity Equilibrium And I say that life is indeed darkness save when there is urge, And all urge is blind save when there is knowledge, And all knowledge is vain save when there is work, And all work is empty save when there is love; And when you work with love you bind yourself to yourself, and to one another, and to God. — Khalil Gibran, The Prophet T he first part of this book lays the foundations for the analysis of the process of job creation and job allocation in the Middle East and North Africa (MENA). Chapter 1 presents the main characteristics of labor markets in MENA and then analyzes systematically the extent to which its mechanics favor economic devel- opment and a productive allocation of the available human resources. It lays out the evidence that labor markets are both inefficient and inequitable and that they oper- ate at a low-productivity equilibrium. Three groups are most likely to be negatively affected by this low-productivity equilibrium: women, youth, and unskilled workers in rural areas. Chapter 2 examines the specific situation of each of these three groups more closely and identifies some of the constraints to fulfilling their potential in the labor market. It analyzes the specific barriers that women face in entering into productive employment, the different dimensions of labor market disadvantage among young people, and the main factors associated with working poverty. 39 Inefficient and Inequitable Labor Markets: A Low-Productivity 1 Equilibrium Main findings • MENA has a large share of untapped human resources that is not participating in economic activities. Most workers are employed in low-value-added sectors, and well-remunerated jobs in the private sector have hardly increased over time. • The most qualified workers desire working in the public sector, which is under little pressure to be productive and whose expansion is costly for society at large. • Low mobility and large unexplained wage gaps across sectors indicate that labor markets are inefficient in allocat- ing human talent and function at a “low productivity equilibrium.” • Attaining a desirable job is largely due to “unearned” circumstances, such as gender and parental characteristics, because labor markets do not allocate opportunities equitably. • A large share of households agglomerates around one member who has access to a well-paid or protected job. This factor constitutes a de facto safety net that makes reforms much more difficult politically. Labor markets in MENA: Key facts High rates of joblessness H igh and persistent rates of jobless- MENA has extensive untapped human ness and low-productivity infor- resources: that is, many willing and able peo- mal employment, together with a ple are not participating in economic activ- formal sector dominated by public employ- ity. When compared to joblessness (defi ned ment, suggest that the Middle East and as the sum of the unemployed and the inac- North Africa (MENA) is not putting human tive) in other middle-income regions, such capital to its best use. Multiple barriers pre- as Latin America and the Caribbean (LAC) vent individuals either from participating in and Europe and Central Asia (ECA), jobless- the labor force or from obtaining a job with ness in MENA is relatively high: 54 of every the productivity and pay that correspond to 100 individuals between the ages of 15 and their human capital and effort—especially if 64 in MENA do not participate in the labor such jobs are scarce. force. A notable feature is that the inactive 41 42 JOBS FOR SHARED PROSPERITY working-age population is much higher in participation in the labor force, which can MENA than in LAC and ECA (at 48 per- be as low as 15 percent in many countries, cent versus 31 and 38 percent, respectively) compared to the world average of 60 percent (figure 1.1). Also, of these regions, MENA in 2005–10 (see figure 1.2). In past decades, has the lowest share of the working-age pop- women in the region have participated very ulation with a formal job (19 percent, versus little in the labor force, especially married 27 percent in LAC and 40 percent in ECA). women with children. An increasing number H igh rates of inactivit y in M ENA of women want to work, but they face a series are largely explained by women’s low of constraints (beyond their own preferences) FIGURE 1.1 Composition of the working-age population in the Middle East and North Africa, Latin America and the Caribbean, and Eastern Europe and Central Asia, 2010 percent a. Selected MENA countries b. Selected LAC countries c. Selected ECA countries 19 31 27 38 40 48 27 5 37 6 6 16 Formal workers Informal workers Unemployed Inactive Source: Based on the International Labour Organization’s Key Indicators of the Labour Market (ILO-KILM) database. Note: ECA = Europe and Central Asia; LAC = Latin America and the Caribbean; MENA = Middle East and North Africa. FIGURE 1.2 Labor force participation rates of the working-age population in selected economies in MENA, by gender, 2009–10 100 90 80 70 60 Percent 50 40 30 20 10 0 te b ta b a . p. p. q ic za ria an on co sia ya ti ep s Ira ou a r bl Re Re Ga Lib oc rd ge ,R an ni bi pu ira jib Qa Tu ra Jo or b ic en Al b nd ra Re am Em Le D iA M m t, A ka ud ab Ye Isl ab an yp Ar Sa n, Ar tB Eg Ira n d ria es te W Sy i Un Females Males Source: Based on ILO-KILM 2012, Kuwait’s 2009 Annual Statistical Abstract, Oman’s Labor Statistics 2009, Saudi Arabia’s 2009 Saudi Manpower Survey, and the United Arab Emirates’ Labor Force Survey (LFS) 2009. See the appendix for more information on some of these surveys. a. Refers to national, nonimmigrant population. b. Includes immigrant population. INEFFICIENT AND INEQUITABLE LABOR MARKE TS: A LOW-PRODUC TIVIT Y EQUILIBRIUM 43 that make it hard for them to find acceptable as entrepreneurs. These constraints are dis- forms of employment. Unemployment rates cussed in detail in chapter 2. for women in MENA are considerably higher than unemployment rates for men. In some High and persistent unemployment localities, and for certain types of academic degrees, more than half of female gradu- Unemployment in MENA is persistently ates with a tertiary education remain unem- higher than in other regions and is over- ployed after a year of job search. Women whelmingly a youth phenomenon (see also face difficulties fi nding employment in figure 1.3a, b, c, and d). Nearly all the the informal sector, whether as employees or unemployed are young, not only because FIGURE 1.3 Unemployment rates in various world regions, by age group and gender, 2006–11 a. Total population b. Youth 30 30 25 25 20 20 Percent Percent 15 15 10 10 5 5 0 0 ld D ia a n st ca or As an EC Ea fri sia a t be ld CD as ric l be or O A lA W e tra ib dl eE OE Af rth W ar ib ra id dl n ar rth nt Ce C No id M eC he Ce No M d t th an d nd an d e a an op pe r ica ica ro Eu er Eu er Am Am tin tin La La c. Males d. Females 25 25 20 20 15 15 Percent Percent 10 10 5 5 0 0 an an sia sia a a t t ld ld CD CD as as ric ric be be or or lA lA eE eE OE OE Af Af W W ib ib ra ra dl dl ar ar rth rth nt nt id id eC eC Ce Ce No No M M th th nd d an d d ea an an pe op ica ica ro r Eu Eu er er Am Am tin tin La La 2006 2008 2011 Source: Based on ILO-KILM 2012. Note: Youth refers to those ages 15–24. OECD = Organisation for Economic Co-operation and Development. 44 JOBS FOR SHARED PROSPERITY u nemploy ment is fa r more prevalent working-age population (see also the dis- among the young (as shown for instance in cussion on demographics in chapter 3). In figure 1.4) but also because of the demo- addition, unemployment among women, graphic weight of this age group within the currently well above that of men, rep- resents the absolute hig hest reg ional FIGURE 1.4 Unemployment rates in MENA, by age, 1998–2010 unemployment rate in the world (see figure 1.3c and d). The educational profile of the unem- Lebanon ployed is heterogeneous in the region. In Tunisia the Arab Republic of Egypt, Jordan, and Morocco Tunisia, unemployment rates are very high Jordan among the better educated, particularly Egypt, Arab Rep., 2009 among women. However, in economies Egypt, Arab Rep., 1998 like some of the Gulf States, Iraq, and the Iraq West Bank and Gaza, unemployment rates tend to be higher among the low-skilled United Arab Emirates population (see figure 1.5). In Morocco, UAE nationals unemployment of the tertiary-educated West Bank and Gaza population fell in recent years to the level 0 10 20 30 40 of those with a secondary education, while Percent in Egypt, the opposite occurred between 55–64 35–54 25–34 15–24 1998 and 2006. Within the stock of the unemployed (as distinguished from unem- ployment rates), the majority in all coun- Source: Based on the Arab Republic of Egypt’s Labor Market Panel Survey (LMPS) 1998 and the Survey of Young People in Egypt (SYPE) 2009, Iraq’s Household Socioeconomic Survey (HSES) 2006, tries in the region are low-skilled workers Jordan’s LMPS 2010, Lebanon’s Employer-Employee Survey 2010, Morocco’s LFS 2009, Tunisia’s LFS who attained at most a secondary education 2010, the West Bank and Gaza’s LFS 2008, and the United Arab Emirates’ LFS 2009. See the appendix for more information on these surveys. (figure 1.6). Note: MENA = Middle East and North Africa; UAE = United Arab Emirates. FIGURE 1.5 Unemployment rates by education in selected economies in MENA, 1998–2010 30 25 20 Percent 15 10 5 0 za ls s 8 9 co a n te is i 99 00 q da na Ga oc Ira ira n ,1 ,2 r io Jo or Tu d Em at p. p. an M Re Re En ab k UA an b b Ar ra ra tB ,A ,A d ite es pt pt W Un y y Eg Eg Primary or below Secondary Tertiary Source: Based on the Arab Republic of Egypt’s LMPS 1998 and SYPE 2009, Iraq’s HSES 2006, Jordan’s LMPS 2010, Morocco’s LFS 2009, Tunisia’s LFS 2010, the United Arab Emirates’ LFS 2009, and the West Bank and Gaza’s LFS 2008. See the appendix for more information on these surveys. Note: The unemployment definition varies by survey type. MENA = Middle East and North Africa; UAE = United Arab Emirates. INEFFICIENT AND INEQUITABLE LABOR MARKE TS: A LOW-PRODUC TIVIT Y EQUILIBRIUM 45 High informality FIGURE 1.6 Percentage of the unemployed by education in selected economies in MENA, 2006–10 Informal employment in MENA is preva- lent and associated with low-productivity and low-quality jobs. A typical country 100 in MENA produces about one-third of its 90 80 gross domestic product (GDP) and employs 70 67 percent of its labor force informally. As a 60 Percent result, more than two-thirds of all workers 50 40 in the region may not have access to health 30 insurance or may not be contributing to a 20 10 pension system (two common proxies for 0 informality). From a fiscal perspective, about za q n a ls p. co si a Ira na Re Ga oc rd ni one-third of total economic output in the tio Tu Jo or b nd ra na M ka t, A region remains undeclared, with consider- E UA an yp tB Eg able implications for government revenue. es W Tertiary Secondary Primary and below The difference between the share of the labor force engaged in informal employment and the share of GDP produced informally (as Source: Based on the Arab Republic of Egypt’s SYPE 2009, Iraq’s HSES 2006, Jordan’s LMPS 2010, Morocco’s LFS 2009, Tunisia’s LFS 2010, the United Arab Emirates’ LFS 2009, and the West Bank and captured by the Schneider index) is larger in Gaza’s LFS 2008. See the appendix for more information on these surveys. MENA, for example, than in Latin America Note: The unemployment definition varies by survey type. MENA = Middle East and North Africa; UAE = United Arab Emirates. (figure 1.7). Although there are many caveats to these comparisons, this evidence suggests that informal jobs in MENA are, on average, associated with lower productivity than in informal output as a share of GDP (about other regions at comparable levels of devel- 20 percent, as measured by the Schneider opment. This interpretation is supported by index; see figure 1.8). This gap likely indi- evidence on earnings, showing that informal cates that a small formal workforce in the workers systematically earn less than formal capital-intensive energy sector produces workers in MENA (Gatti et al. 2012; see most of the output of these countries, while also figure 1.7). the majority of the informal workforce is The prevalence of informality varies engaged in low-productivity jobs. substantially across countries in the region. Their heterogeneity—in size, role of the Widespread subsistence self- public sector, availability of resources and employment and limited labor, economic development, and produc- entrepreneurship among the highly tive and demographic structure—influences skilled the size of the informal economy. Informal- ity is higher in countries such as Morocco The magnitude and qualitative features of and the Republic of Yemen, where the entrepreneurship (comprising here both self- share of agricultural employment relative employed and employers) in MENA are dif- to total employment is high, and lower in ferent from those in other regions in the Egypt and Jordan, where the public sec- world. About one-third of the workers in tor accounts for 30–35 percent of overall MENA are nonwage workers (figure 1.9), a employment (Elbadawi and Loayza 2008).1 share that varies across countries accord- A country’s productive structure is also an ing to their per capita income and the important driver of informality: energy- importance of agriculture in the econ- rich countries (such as the Islamic Repub- omy. Globally, the share of wage work- lic of Iran and the Syrian Arab Republic) ers (nonentrepreneurs) across countries tend to have high informal employment tends to increase with the level of income (about 80 percent) but have rather low (Gindling and Newhouse 2012), refl ecting 46 JOBS FOR SHARED PROSPERITY FIGURE 1.7 Informality in MENA and other regions, 2001–07 a. Undeclared output Middle East and 27.7 North Africa GCC countries 20.1 non-GCC countries 36.2 Developed countries 16.6 East Asia and Pacific 32.8 South Asia 37.0 Eastern Europe and 38.9 Central Asia Sub-Saharan Africa 42.2 Latin America and 43.9 the Caribbean 0 10 20 30 40 50 60 70 80 90 100 % of GDP (Schneider index) b. Labor force not contributing to social security non-GCC countries 65.0 Developed countries 8.9 Eastern Europe and Central Asia 39.4 Latin America and 61.3 the Caribbean East Asia and Pacific 76.1 South Asia 91.0 Sub-Saharan Africa 94.7 0 10 20 30 40 50 60 70 80 90 100 Percent Source: Gatti et al. 2012. Note: The periods covered are the average of 2000–07 for pension scheme and 2001–04 for Schneider index, respectively. GCC = Gulf Cooperation Council. the structural transformation (that is, the small fi rms. The educational distribution movement of workers out of agriculture of workers also suggests that MENA has into wage work) of the countries as well as a large number of poorly educated subsis- how the size distribution of fi rms evolves tence entrepreneurs and a small number with development. In this respect, MENA of highly educated fi rm owners. Globally, countries are atypical, as they have a lower employers tend to be more educated than share of wage workers than middle-income self-employed and wage employees (Gin- regions such as ECA and LAC, denoting dling and Newhouse 2012), but not in the prevalence of employment in relatively MENA. Across all countries considered, INEFFICIENT AND INEQUITABLE LABOR MARKE TS: A LOW-PRODUC TIVIT Y EQUILIBRIUM 47 FIGURE 1.8 Informality rates for selected non-GCC economies in MENA, 2000–07 b. Labor force not contributing to a. Undeclared output social security Iran, Islamic Rep. 20 Libya 35 Jordan 21 Egypt, Arab Rep. 45 Tunisia 50 Syrian Arab Rep. 21 Algeria 63 Yemen, Rep. 28 Iran, Islamic Rep. 65 Algeria 35 Lebanon 67 Jordan 67 Egypt, Arab Rep. 36 Morocco 80 Lebanon 36 West Bank and Gaza 81 Morocco 37 Iraq 82 Syrian Arab Rep. 83 Tunisia 39 Yemen, Rep. 90 0 10 20 30 40 50 0 20 40 60 80 100 % of GDP (Schneider index) Percent Source: Gatti et al. 2012. Note: Time periods are as follows: Schneider index (1.8a), average 2001–04; not contributing to social security, average 2000–07. MENA = Middle East and North Africa; GCC = Gulf Cooperation Council; GDP = gross domestic product. fewer employers than wage employees FIGURE 1.9 Distribution of work status across regions, 1999–2008 have a tertiary education (see figure 1.10), reflecting the role played by the public sec- Work status of employed individuals 100 tor in absorbing high-skilled individuals. 90 High-productivity entrepreneurs are 80 few, as measured by household welfare. 70 Figure 1.11 shows the distribution of 60 Percent 50 employed males by work status and by 40 quintile of their household wealth or con- 30 sumption in Egypt, Jordan, and Morocco. 2 20 About half of informal wage workers live 10 in poor or near-poor households, while 0 e l A nd sia b- Pac and ca tin Afric nd be d m entrepreneurs are distributed more evenly ib an fri sia ha ic a ra a a hA co an if nA nt pe ia rth ast ar ca in ut As across quintiles. Formal wage workers are Ce ro i No e E ra e C er gh So Eu st th m dl Hi Ea A rn Sa id the most likely to be in the top income M ste La Su Ea quintiles. A more detailed disaggrega- Self-employed Employer tion would show that the employers are Unpaid Wage earner those most likely to be in higher-income quintiles—they are the most successful of Source: Gindling and Newhouse 2012, based on the International Income Distribution Database (I2D2) (World Bank 2012) (repository of household surveys). entrepreneurs—while the self-employed are relatively more likely to be poor, but less so than informal wage workers. Another a large public sector, which accounts for fact that makes MENA countries stand out about 29 percent of overall employment is the gender predominance among entre- in the region. In countries like Egypt, for preneurs, the majority of whom are men. example, growth in the civil service was the result of a social contract in the 1970s and 1980s, whereby the government effec- Dominance of public employment in tively offered employment guarantees to the formal sector university graduates and to graduates of Labor markets in many M ENA coun- vocational secondary schools and train- tries are still influenced by the legacy of ing institutes (World Bank 2004). As a 48 JOBS FOR SHARED PROSPERITY consequence of that inheritance, the civil FIGURE 1.10 Distribution of employment status service in many MENA countries is larger by educational attainment in selected economies than in other countries with similar income in MENA, 2008–10 levels and economic structure (Elbadawi and Loayza 2008). Wage employee D e spite t he fac t t h at employ ment Employer Jordan growth in the public sector has slowed Self-employed dramatically in recent years, government Unpaid/other Ag worker and public enterprises in most countries Wage employee still account for more than 60 percent of Employer Arab Rep. Egypt, all formal sector employment, while the Self-employed formal private sector (generally the most Unpaid/other productive in the economy) is rather small Ag worker (figure 1.12). Employment in the formal Wage employee Employer private sector is almost nonexistent in Tunisia Self-employed Iraq and the Republic of Yemen and below Unpaid/other 10 percent of total employment in Egypt Ag worker and Morocco, while it is somewhat larger in Wage employee Jordan and Tunisia. As seen earlier, this situ- Employer ation reflects a number of factors, including Morocco Self-employed the structure of a country’s production, the Unpaid/other Ag worker large size of the public sector (which effec- Wage employee West Bank and tively competes for resources and talent with Employer the private formal sector), and the design Gaza Self-employed of pension systems (which in Iraq and the Unpaid/other Republic of Yemen do not extend to the Ag worker private sector) (see also the discussion in 0 25 50 75 100 Percent chapters 4 and 5). Below primary Primary/basic Secondary Tertiary Source: Based on the Arab Republic of Egypt’s SYPE 2009, Jordan’s LMPS FIGURE 1.11 Work status of employed males in the Arab Republic 2010, Morocco’s LFS 2009, Tunisia’s LFS 2010, and the West Bank and of Egypt, Jordan, and Morocco by household per capita wealth and Gaza’s LFS 2008. Note: ag = agricultural; MENA = Middle East and North Africa. consumption quintile, 2009–10 Formal wage earner MENA labor markets: Inefficient Arab Rep. Egypt, Entrepreneur Informal wage earner and inequitable Unpaid/other Labor markets in MENA can be charac- Formal wage earner terized as both inefficient and inequitable. Jordan Entrepreneur Informal wage earner They are inefficient because human capital Unpaid/other is not allocated where it could get a higher return (that is, so that the highest output Formal wage earner can be achieved with the available inputs). Morocco Entrepreneur Informal wage earner And they are inequitable because desirable Unpaid/other labor outcomes, such as getting access to a 0 20 40 60 80 100 protected or well-paid job, are determined Percent Q1 Q2 Q3 Q4 Q5 by circumstances (like gender and parental education) that do not necessarily depend on Sources: The Arab Republic of Egypt’s SYPE 2009, Jordan’s LMPS 2010, and Morocco’s LFS 2009. individual effort. The inefficient and inequi- Note: Q = quintile. table nature of labor markets in the region INEFFICIENT AND INEQUITABLE LABOR MARKE TS: A LOW-PRODUC TIVIT Y EQUILIBRIUM 49 has created a deep sense of exclusion among FIGURE 1.12 Employment composition by sector in selected young people. economies in MENA, 2005–10 UAE nationals Symptoms of inefficiency UAE expatriates Iraq Three main symptoms of inefficiency pre- Jordan vail in MENA’s labor markets: (1) wage Egypt, Arab Rep. differentials across sectors and individuals Tunisia are not explained by differences in human West Bank and Gaza capital endowments; (2) mobility of workers Morocco from low-productivity to high-productivity Yemen, Rep. employment is low; and (3) geographical dif- 0 20 40 60 80 100 Percent ferentials in labor market outcomes are large Public wage earner Private formal wage earner and persistent. Private informal wage earner Self-employed and unpaid Wage differentials unexplained by Source: Based on the Arab Republic of Egypt’s LFS 2010, Iraq’s HSES 2006, Jordan’s LMPS 2010, differences in human capital endowments Morocco’s LFS 2009, Tunisia’s LFS 2010, the United Arab Emirates’ LFS 2009, the West Bank and Neoclassical economic theory states that in Gaza’s LFS 2008, and the Republic of Yemen’s Household Budget Survey (HBS) 2005. See the appen- dix for more information on these surveys. unified and efficient labor markets, firms Note: Formal employment, defined through affiliation with social security, is not available among create vacancies according to their produc- private sector firms in the West Bank and Gaza and for migrant workers in the United Arab Emirates. MENA = Middle East and North Africa; UAE = United Arab Emirates. tive needs and workers compete to fi ll these vacancies according to their human capital endowments. In such labor markets, varia- tions in the wages and conditions of differ- tertiary education, respectively) would earn ent workers arise from individual differences wages that are twice to three times larger in their human capital (skills, experience, than those earned by an informal worker or formal education) or preferences. Symp- who also has 12–16 years of education. This toms of ineffi ciency exist when differences result indicates that wage differentials across in compensation and other employment con- sectors are not necessarily explained by dif- ditions originate on the demand side rather ferences in human capital. than being explained by individual workers’ While higher returns to human capital in productivity. Often such inefficiency is char- the formal sector are not uncommon in devel- acterized by the existence of a sector (or sec- oping economies (for instance, see World tors) that rewards human capital better than Bank 2009, for a discussion of Turkey), what others or in cases where labor market insti- is unique to MENA is that returns to human tutions (such as a minimum wage) oblige capital in the informal sector (on average) are employers to pay wages above productivity. strikingly low. Figure 1.13 plots the expected wage rate of Finally, as expected, the formal private an average worker (not including the effect sector pays higher wages to individuals who of experience) according to his or her years have attained higher levels of education, of education by applying the exponential which probably reflects higher productivity in function to the coefficients obtained from the this sector. It is worth mentioning, however, Mincer model.3 Results indicate that returns that in many countries, public sector jobs to education (and especially of primary and are still preferred over private formal jobs, secondary education) are higher in the pub- because, first, private formal jobs are scarce lic and formal private sector and very low in and, second, public sector jobs might require the informal sector. For instance, a typical less effort, have better nonmonetary benefits, formal worker in Egypt and Jordan with no and are more secure (as discussed in detail in experience but with 12–16 years of educa- chapter 5). Only about one-third of all youth tion (equivalent to complete secondary and with a tertiary education fi nds public sector 50 JOBS FOR SHARED PROSPERITY FIGURE 1.13 Expected wage by years of education) tend to select themselves into education in the Arab Republic of Egypt, 2006, and public sector and services sector employ- Jordan, 2010 ment. As indicated in table 1.1, a signifi- cant share of employed women work in the a. The Arab Republic of Egypt, 2006 public sector (from 40 percent in Tunisia to 4.0 77 percent in Iraq) and in the services sec- 3.5 tor (from 16 percent in Iraq to 25 percent Log hourly wage 3.0 2.5 in Jordan). Because working women are on 2.0 average more educated than their male coun- 1.5 terparts, one would not especially expect to 1.0 see a gender wage differential in the public 0.5 sector. In the private sector, women may 0 0 2 4 6 8 10 12 14 16 18 20 self-select into lower-paying service indus- Years of education tries (as occurs in many countries), which could partly explain the wage differentials b. Jordan, 2010 4.0 seen between men and women. 3.5 For instance, estimates for Egypt in 2006 indicate that the majority of women in the Log hourly wage 3.0 2.5 private sector work in the agricultural and 2.0 personal service sectors (about 60 percent 1.5 of all employed women), while most men 1.0 0.5 work in transport, retail, tourism, and manu- 0 facturing (which are generally better-paid 0 2 4 6 8 10 12 14 16 18 20 industries) (see World Bank 2010). Even after Years of education controlling for self-selection into particular Public sector Private formal sectors and industries, however, results from Private informal Egypt and Jordan indicate that women earn lower wages than their male counterparts, Source: Based on the Arab Republic of Egypt’s LMPS 2006 and Jordan’s LMPS 2010. See the appendix for more information on these surveys. net of education and experience, even in the Note: E(w|Yrs of edu) = exp(C + β1 Yrs of Edu. + β2 Experience + β3 Experi- public sector (figure 1.15). These gaps can be ence squared). Estimations based on Mincer model. Sample includes wage earners in urban areas working between 30 and 60 hours per week. attributed to discrimination (which in theory is inefficient, but firms may be able to afford these practices) or to other unobservable fac- employment (see figure 1.14b); yet data from tors, such as ability and effort.4 At the same Gallup indicate that in most countries in the time, results from Egypt and Jordan differ region, the majority of young people (50–60 significantly, which indicates that wage dif- percent in most cases) prefer to work in the ferentials between men and women are likely public sector (figure 1.14a). The implications to vary greatly across countries in the region. for labor market efficiency are important: In Jordan, wage gaps are rather low (vary- the most talented young individuals prefer to ing between 10 and 12 percent in the public work in the public sector, which provides less and private sectors), whereas in Egypt, they value added to the economy than the high- are much larger (ranging from 19 percent in productivity private sector. the public sector to 77 percent in the private Important wage differentials that are informal sector). not fully explained by differences in human Finally, wage differentials also exist capital or industry also persist among men between migrant and nonmigrant popula- and women. Employment among women tions and are also not explained by differences in MENA is likely to reflect important self- in human capital endowments. Labor market selection patterns. That is, the few women outcomes differ greatly between national and who choose to participate in the labor foreign workers, particularly in the Gulf coun- force (generally those with higher levels of tries. As a result of these countries’ liberal de INEFFICIENT AND INEQUITABLE LABOR MARKE TS: A LOW-PRODUC TIVIT Y EQUILIBRIUM 51 FIGURE 1.14 Preference for and attainment of public sector employment among youths ages 15–34 in selected economies in MENA, 2005–10 a. Youths that would prefer to work in the public sector b. Youths in the labor force with tertiary education 90 30 80 25 70 60 20 Percent Percent 50 15 40 30 10 20 5 10 0 0 sia p. ra blic Ye Ku es en it ain 6 09 an Mo on nd co Al za Tu ria Sa jib sia i A uti Eg rab Jord a yp Re an ab Q q ira r Ba ep. an q . co za ep Em ata bi 00 Ira Ira m wa Re t Ga Ga k a roc oc 20 rd ge n ni ni ud o ,R ,R hr ra t, A pu ., 2 ba Tu Jo or b en d p., p Le D n M Re Re m ka Ye b b an Ar ra ra nA tB tB t, A t, A d ite ria es es yp yp W W Un Sy Eg Eg Youths with tertiary education Youths with tertiary education that work in the public sector Source: Based on the Arab Republic of Egypt’s SYPE 2009 and LMPS 2006, Iraq’s HSES 2006, Jordan’s LMPS 2010, Morocco’s LFS 2009, Tunisia’s LFS 2010, the West Bank and Gaza’s LFS 2008, the Republic of Yemen’s HBS 2005, and the Gallup World Poll Survey 2010. See the appendix for more information on these surveys. Note: MENA = Middle East and North Africa. TABLE 1.1 Composition of urban employment by industry in the Arab Republic of Egypt, Iraq, Jordan, and Tunisia, 2006–10 percent Jordan Egypt, Arab Rep. Iraq Tunisia Women Men Women Men Women Men Women Men Primary 3.63 3.20 6.07 4.71 1.04 3.47 2.81 6.63 Secondary 11.28 23.82 11.38 29.43 5.77 29.95 35.23 34.63 Tertiary 24.72 44.58 24.10 41.48 16.13 38.95 21.68 35.94 Public administration 60.37 28.40 58.45 24.38 77.07 27.62 40.28 22.80 Source: Based on the Arab Republic of Egypt’s SYPE 2009, Iraq’s HSES 2006, Jordan’s LMPS 2010, and Tunisia’s LFS 2010. Note: Primary sector (agriculture); secondary sector (manufacturing and construction); tertiary sector (wholesale, transport, services); and public administration and social services (including education and health). facto immigration policies, the abundant sup- Limited labor market mobility into high ply of low-wage foreign labor has depressed value-added jobs the immigrants’ wages, pushing them to the L ow mobi l it y of hu ma n capit a l i nto bottom of the wage scale. Taking the example high-productivity sectors and out of low- of the United Arab Emirates, results from the productivity sectors is also a symptom of World Bank (2011) suggest that wage differ- labor market ineffi ciency. In this context, entials persist after controlling for education a desirable social and economic outcome and skills and indicate that nationals earn would be one in which workers move out wages that are two to three times higher than of sectors with low value added (the infor- those of migrants. As one might expect, these mal sector) and low pay (the public sector) gaps are even greater for low-skilled workers and into sectors with high value added and (figure 1.16). (Chapter 5 will explain this issue high pay (the formal private sector). Panel in greater detail.) data and retrospective questions from 52 JOBS FOR SHARED PROSPERITY FIGURE 1.15 Male-female wage differentials by sector in the Arab Republic of Egypt and Jordan, 2006 and 2010 a. The Arab Republic of Egypt b. Jordan 80 80 70 70 60 60 50 50 Percent Percent 40 40 30 30 20 20 10 10 0 0 Public Private formal Private informal Public Private wage earner wage earner wage earner wage earner wage earner Source: Based on the Arab Republic of Egypt’s LMPS 2006 and Jordan’s LMPS 2010. Note: Data are net of education, experience, and industry. The selection sample includes wage earners in urban areas working between 30 and 60 hours per week. FIGURE 1.16 Education and wage differentials among migrants and nationals in the United Arab Emirates, 2011 a. Education distribution of the employed population b. Ratio of average wages (nationals vs. expatriates) 50 4.0 45 3.5 40 3.0 35 30 2.5 Percent Ratio 25 2.0 20 1.5 15 10 1.0 5 0.5 0 0 d d led led d d w n l l y ra na ille ille ille ille ar io elo ne kil kil io rti at -sk -sk sk sk rb at i-s i-s ge uc Te w- Un gh gh oc em m yo ed ry Lo yv Se hi Hi da ar s sic p gh ar im on To Ba nd Hi Pr c Se co Se Expatriates Nationals Source: World Bank 2011. Note: Skill levels here are defined on the basis of different one-digit International Standard Classification of Occupations (ISCO) job categories, in which the highest ISCO classification reflects the highest skill level. Egypt, Jordan, and Lebanon allow for and the majority of those who left the pub- measurement of the extent of mobility in lic sector became inactive, probably due to the labor market. retirement. Only a minority (1–2 percent) In the case of Egypt (table 1.2), most work- of all workers in the public sector in 1998 ers employed in the public sector in 1998 moved to the formal private sector. In con- (79 percent among unskilled workers and trast, movement into the public sector was 87 percent among skilled workers) contin- substantial: a significant share (about 20 ued to work in the public sector in 2006, percent) of all high-skilled workers in the INEFFICIENT AND INEQUITABLE LABOR MARKE TS: A LOW-PRODUC TIVIT Y EQUILIBRIUM 53 TABLE 1.2 Employment transition matrix in the Arab Republic of Egypt, 1998 and 2006 Formal Informal wage wage Self- All workers, Inactive, Unemployed, Public, earner, earner, Employer, employed, Unpaid, Total, 1998 2006 (%) 2006 (%) 2006 (%) 2006 (%) 2006 (%) 2006 (%) 2006 (%) 2006 (%) 2006 (%) N Workers with tertiary education Public wage earner 11 1 87 1 1 0 0 0 100 710 Formal wage earner 5 0 20 65 6 2 2 0 100 86 Informal wage earner 11 4 21 23 26 9 7 0 100 57 Employer 7 2 10 0 2 67 11 2 100 61 Self-employed 3 0 0 3 3 41 48 0 100 29 Workers with at most secondary education Public wage earner 12 0 79 2 2 2 1 0 100 971 Formal wage earner 6 1 7 52 17 9 9 0 100 172 Informal wage earner 6 2 9 9 42 20 11 1 100 598 Employer 7 0 4 2 5 69 12 1 100 297 Self-employed 11 1 5 2 10 27 40 5 100 320 Source: Based on the Arab Republic of Egypt’s LMPS, various years. Note: The highlighted text measures the share of workers whose employment status did not change between 1998 and 2006; for example, 87 percent of workers with tertiary education who were public employees in 1998 were still public employees in 2006. N = number. Sample data are for individuals between 31 and 64 years of age. formal private sector in 1998 held public mobility out of informal (presumably low- sector jobs in 2006. productivity) activities. As in Egypt, in Jordan With regard to informal workers, data a sizable share of high-skilled workers uses the for Egypt (table 1.2) show that their mobil- informal sector as a bridge to formal employ- ity prospects are associated with their educa- ment (mainly in the public sector). tional level. Among workers with a tertiary Lebanon’s case is somewhat different education, working as a wage earner in the from the previous two, among other reasons informal sector is a transitional path to for- because the available data show short-term mal employment, while informal wage earn- mobility patterns (see table 1.4). Because of ers without a university education have quite the short time span, even small differences in limited mobility into higher-value-added probabilities indicate important mobility dif- jobs. Indeed, 42 percent of informal wage ferences over longer periods. Table 1.4 shows earners without a university diploma in that in Lebanon, hardly any worker who is in 1998 still had a similar employment status informal wage work or is self-employed has in 2006 (against only 26 percent among the a chance to move into a formal wage job, as highly skilled), while 31 percent had become was the case in Egypt, signaling that mobil- employers or self-employed (probably still in ity into higher-value-added wage positions is the informal sector). constrained. However, the nonwage sector in To a large extent, employment transitions Lebanon appears to be particularly open to in Jordan between 1999 and 2010 (table 1.3) entry and is attractive on many levels: self- mirror patterns observed in the data from employment is a very important avenue into Egypt. Mobility out of the public sector the labor market for unemployed and inactive (other than to become inactive or retired) is individuals (relative to other work status), limited, and low-skilled workers have little and even many formal wage workers move to 54 JOBS FOR SHARED PROSPERITY TABLE 1.3 Employment transition matrix in Jordan, 1999 and 2010 Formal Informal wage wage Self- Inactive, Unemployed, Public, earner, earner, Employer, employed, Unpaid, Total, All workers, 1999 2010 (%) 2010 (%) 2010 (%) 2010 (%) 2010 (%) 2010 (%) 2010 (%) 2010 (%) 2010 (%) N Workers with tertiary education Public wage earner 15 4 68 5 3 1 4 0 100 262 Formal wage earner 13 7 14 57 4 5 0 1 100 117 Informal wage earner 16 3 18 14 33 7 9 0 100 110 Employer 15 6 0 0 0 73 7 0 100 40 Self-employed 8 3 14 4 5 27 40 0 100 35 Workers with at most secondary education Public wage earner 24 4 55 5 4 2 6 0 100 977 Formal wage earner 22 4 7 54 6 4 2 0 100 380 Informal wage earner 12 5 7 11 44 9 11 0 100 888 Employer 10 6 3 0 3 64 13 0 100 185 Self-employed 15 8 5 4 9 6 52 0 100 292 Source: Based on Jordan’s LMPS 2010, with retrospective data. Note: The shaded text measures the share of workers whose employment status did not change between 1999 and 2010; that is, 68 percent of workers with tertiary education who were public employees in 1999 were still public employees in 2010. N = number. The sample consists of individuals between 33 and 64 years of age. TABLE 1.4 Monthly employment transitions of the working-age population in Lebanon, 2010 percent Formal Informal Employee Current month → next month Self-employed employees employees unknown Unemployed Inactive Total Self-employed 99.96 0.00 0.00 0.03 0.01 0.00 100 Formal employees 3.13 94.06 0.32 1.94 0.43 0.11 100 Informal employees 3.02 0.19 93.69 1.89 0.91 0.30 100 Employee—unknown 0.17 0.00 0.06 99.72 0.06 0.00 100 Unemployed 6.19 0.26 2.01 0.00 91.54 0.00 100 Inactive 4.50 0.35 1.04 0.25 0.69 93.18 100 Total 83.70 1.74 4.97 3.72 2.15 3.72 100 Source: World Bank 2013, using Lebanon’s Employer-Employee Survey 2011. See the appendix for more information on this survey. Note: This table should be read as follows: for example, 99.96 percent of respondents who were self-employed during one month were self-employed the following month. self-employment. The quality of these jobs is markets, skill mismatches, and limited very heterogeneous, and the high persistence mobility of women may prevent individuals of self-employment could be a combination from migrating from one region to another of preference and lack of better alternatives to obtain a better job (see Grunwald et al. (as discussed in chapter 2). 2009). Data for Tunisia indicate that some of the Geographical disparities regions with the highest unemployment rates Another symptom of inefficient alloca- in 2005 (such as Gafsa and Tataouine) also tion of resources in labor markets is the experienced the highest increase in unem- marked geographical variation in labor ployment rates between 2005 and 2011. market outcomes that persists over time. In This suggests that mobility constraints mobile markets, workers tend to migrate to prevent workers from taking advantage of regions where employment opportunities employment opportunities in other regions are better. Yet important factors such as (figure 1.17). Labor force data for Tunisia the cost of housing, poorly developed rental show that among the lower-skilled and INEFFICIENT AND INEQUITABLE LABOR MARKE TS: A LOW-PRODUC TIVIT Y EQUILIBRIUM 55 FIGURE 1.17 Levels and trends in regional unemployment in Tunisia, 2005–11 a. Unemployment, 2005 b. Change in unemployment IBRD 39569 Mediterranean Mediterranean SEPTEMBER 2012 Sea BIZERTE Sea BIZERTE ARIANA ARIANA NABEUL NABEUL TUNIS MANOUBA TUNIS JENDOUBA JENDOUBA BEJA BEJA ZAGHOUAN ZAGHOUAN SILIANA Gulf of Hammamet SILIANA Gulf of Hammamet LE KEF SOUSSE LE KEF SOUSSE KAIROUAN MONASTIR KAIROUAN MONASTIR KASSERINE KASSERINE MAHDIA MAHDIA SIDI BOUZID SIDI BOUZID SFAX SFAX Kerkenna Kerkenna Islands Islands GAFSA GAFSA Gulf of Gulf of Gabes Gabes TOZEUR TOZEUR Djerba Djerba Chott el Jerid GABES Island Chott el Jerid GABES Island MEDENINE MEDENINE KEBILI KEBILI ALGERIA ALGERIA TATAOUINE TATAOUINE Percent L I B YA % of change L I B YA 15.6–22.2 12.1–24.5 12.2–15.6 4.4–12.1 10.2–12.2 0.1–4.4 8.1–10.2 –5.4 to –0.1 0 25 50 75 KILOMETERS 0 25 50 MILES Source: Based on Tunisia’s LFS, various years. female unemployed, the distance to available external circumstances beyond their control jobs was among the top reasons for refusing and their individual efforts. This is measured job offers (see chapter 2). here using the approach developed by John Roemer (1998) to capture the extent to which opportunities are equally distributed in the Symptoms of inequity in the labor labor market. market Inequality of opportunities is measured This section highlights how an individual’s by the dissimilarity index (D-index), which access to desirable jobs—defined as jobs indicates how equitably outcomes are dis- that offer either protection (notably social tributed among groups with different “life insurance) or high pay—is not necessar- circumstances,” such as gender, race, family ily transparent or determined by his or her background, and place of birth, which may own effort or merit. Indeed, to a large extent, affect their capacity to gain advantages but labor market outcomes for individuals in do not depend on individual effort or merit many MENA countries depend more on (see Barros et al. 2009; also Bourguignon 56 JOBS FOR SHARED PROSPERITY et al. 2007). Box 1.1 explains in detail the The present analysis includes data from index construction. Egypt, Jordan, and Morocco for individuals The index can be interpreted as the share between 21 and 34 years old. The exogenous of the “outcome” that needs to be reallo- circumstances considered in the analysis cated among the groups with different life include the person’s gender, his or her loca- circumstances to ensure equality of oppor- tion (urban or rural), the wealth quintile of tunities. It ranges between 0 and 1, with his or her household, and the level of educa- higher values suggesting higher inequality tion of his or her mother and father. To con- (see box 1.1). For instance, a D-index equiv- trol for individual effort, the analysis includes alent to 0.3 for the outcome variable “hav- educational attainment and age (as a proxy ing formal employment” would suggest that for years of working experience). It is worth 30 percent of all individuals having formal noting that ability and effort are largely unob- employment are disproportionately repre- servable, and thus the analysis is likely to sented in one or more circumstance group; underestimate the effect of individual effort thus, to achieve equal opportunities across on outcome inequality. Since this framework different circumstance groups, one would requires the identification of advantages or need to reallocate 30 percent of formal jobs. outcomes deemed to be favorable, the results BOX 1.1 The dissimilarity index In Roemer’s framework (Roemer 1998), social fair- From the estimation of this logistic regression, we ness exists when an individual’s efforts prevail over will obtain coefficient estimates. his or her circumstances in determining the main 2. Given these coeffi cient estimates, determine for outcomes that defi ne his or her life. In some cases, each individual in the sample the predicted prob- to achieve social fairness, government interventions ability of access to the basic good or service in might be required to alleviate the “additional” bur- consideration, pi , based on the predicted relation- den that these circumstances impose on a certain ˆ ,and a vector of their circumstances xki: ship, b k group of people, to delink desirable outcomes from ˆ 0 + ∑k m ˆ privileged backgrounds, and provide everyone real- Exp( b = 1 xki b k ) ˆi = p . istic possibilities of success. The Dissimilarity Index 1 + Exp( bˆ 0 + ∑ k = 1 xkib m ˆk) is used to capture the extent to which opportunities are distributed equally in society. 3. Compute the overall coverage rate C (where wi = 1/n The first step in constructing the index is to or some sampling weights): identify a dichotomous variable that captures the n outcomes deemed to be socially desirable, such as C = ∑ wi p ˆ i. having more income, more human capital, or high- 1 quality high-paying employment. The second step is to identify a set of variables 4. Compute the dissimilarity index (D): that captures individual circumstances and individual n ˆ = 1 efforts. As described in Bourguignon et al. (2007), D ∑ 2C i =1 Wi | pi − C |. circumstances are factors that are out of the individu- al’s control, such as gender, race, family background, The index has several advantages: it is easy to calcu- and place of birth, but that may affect their capacity late, allows progress to be monitored over time, and is to gain advantages. Effort, however, can be affected particularly useful for comparing inequality of oppor- by individual choice. The following basic steps are tunities across different countries and sectors of the used to calculate the dissimilarity index (D): population. Another advantage of the index is that it 1. Estimate a separable logistic model on whether can be easily decomposed so that the contribution of individual i had access to a given basic good or all circumstances to the dissimilarity index adds up to service as a function of his or her circumstances. 100 percent (see Hoyos and Narayan 2011). INEFFICIENT AND INEQUITABLE LABOR MARKE TS: A LOW-PRODUC TIVIT Y EQUILIBRIUM 57 look mainly at employment outcomes that FIGURE 1.18 Decomposition of the dissimilarity index by are desirable (in this case, formal employ- circumstance and effort for men ages 21–34 in the Arab Republic of ment or public employment and having a Egypt, Jordan, and Morocco, 2009 and 2010 permanent or high-paying job). Results for young men and women are 0.7 presented in figures 1.18 and 1.19. In nearly 0.6 0.5 all cases, 60 to 75 percent of the D-index is Index 0.4 0.3 explained by circumstances outside an indi- 0.2 vidual’s control. Results for men indicate 0.1 0 that the D-index is generally higher for all Morocco Morocco Morocco Morocco Jordan Jordan Jordan Jordan Jordan Egypt, Arab Rep. Egypt, Arab Rep. Egypt, Arab Rep. Egypt, Arab Rep. Egypt, Arab Rep. selected outcomes in Morocco, suggesting that the inequality of opportunities plays a higher role in explaining youth employ- ment outcomes in that country compared to Egypt or Jordan. Inequality in the out- Being Formal Public sector Permanent High- employed employment employment job wages job comes “formal employment” and “public sector employment” (which are somewhat Effort Circumstances correlated) is rather high for all countries (the D-index swings between 0.2 and 0.7) Source: Based on the Arab Republic of Egypt’s SYPE 2009, Jordan’s LMPS 2010, and Morocco’s House- hold and Youth Survey (HYS) 2010. See the appendix for more information on these surveys. but is consistently lowest in Jordan and Note: Morocco’s HYS 2010 does not contain wages, which is why there are no D-index values for highest in Morocco. The inequality in the “high-wages jobs.” outcome “having a permanent job” seems low in Egypt (D-index at 0.1) and rather high in Morocco (D-index at 0.5) and is also explained mainly by circumstances rather than effort. Finally, data from Egypt and Jordan indicate that the inequality of “holding a high-wage job” is moderate (the D-index oscillates between 0.1 and 0.2) and FIGURE 1.19 Decomposition of the dissimilarity index by is explained mainly by circumstances in circumstance and effort for women ages 21–34 in the Arab Republic of Egypt, Jordan, and Morocco, 2009 and 2010 Egypt (and to lower extent in Jordan). Results for female youth, on the other hand, suggest that the unequal distribu- 0.7 tion of opportunities is higher in Jordan 0.6 and Morocco than in Egypt. In addition, it 0.5 0.4 Index is interesting to see that in most cases the 0.3 weight of circumstances is consistently lower 0.2 among women than among men. This result 0.1 is true, for instance, in Jordan, where most 0 inequality in access to public and permanent Egypt, Arab Rep. Morocco Egypt, Arab Rep. Morocco Egypt, Arab Rep. Morocco Egypt, Arab Rep. Morocco Egypt, Arab Rep. Jordan Jordan Jordan Jordan Jordan employment among men is explained by cir- cumstances, whereas for women a signifi - cant share is explained by effort. This result probably reflects self-selection: because few women enter the labor force—and they Being Formal Public sector Permanent High- are more educated than men who enter employed employment employment job wages job the labor force—their background (family Effort Circumstances income and socioeconomic situation) tends to be more homogeneous and is thus less Source: Based on the Arab Republic of Egypt’s SYPE 2009, Jordan’s LMPS 2010, and Morocco’s HYS 2010. likely to explain differences in employment Note: Morocco’s HYS 2010 does not contain wages, which is why there are no D-index values for outcomes. “high-wages jobs.” 58 JOBS FOR SHARED PROSPERITY MENA labor markets: A low-productivity previous discussion do not affect all indi- equilibrium viduals in the same way. The chances of attaining a “desirable job” may vary across MENA’s economies offer several paths socioeconomic groups and lead to the ineq- to “success” in the labor market: a well- uitable labor market outcomes that we have protected job (especially in the public sector), observed. a well-paid job in the private sector, or To quantify this distribution, figure 1.20 high-earning self-employment. 5 The dis- m aps t he ex tent to wh ich d i f ferent tortions of the labor market laid out in the FIGUR E 1.20 Share of workers with high-paying or protected jobs among the working-age population in the Arab Republic of Egypt, Jordan, and urban Morocco, 2009 and 2010 a. Arab Republic of Egypt, 2009 b. Jordan, 2010 25 25 20 Rur-F-Primary 20 Rur-M-Primary % of population % of population 15 15 Urb-M-Primary-Y Rur-M-Skill Urb-F-Primary-Y Urb-F-Skill-Y Urb-M-Skill-Y 10 10 Urb-F-Primary-A Urb-M-Skill-A Rur-F-Skill Urb-F-Skill-Y Urb-M-Skill-Y Urb-F-Skill-A Urb-M-Primary-A Urb-F-Primary-A Urb-M-Skill-A Rur-F-Primary Rur-M-Primary 5 Urb-F-Primary-Y Urb-F-Skill-A 5 Rur-F-Skill Rur-M-Skill Urb-M-Primary-Y Urb-M-Primary-A 0 0 0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100 % in well-paid or protected job % in well-paid or protected job c. Urban Morocco, 2010 25 Urb-F-Primary-A 20 Urb-F-Primary-Y Urb-M-Primary-A % of population 15 Urb-M-Primary-Y Urb-M-Skill-Y 10 Urb-F-Skill-Y Urb-M-Skill-A 5 Urb-F-Skill-A 0 0 10 20 30 40 50 60 70 80 90 100 % in well-paid or protected job Less than 50% in well-paid or protected job 50% or more in well-paid or protected job Source: Based on the Arab Republic of Egypt’s SYPE 2009, Jordan’s LMPS 2010, and Morocco’s HYS 2010. Note: The x-axis represents the share of workers in group J (say, urban-male-skilled-adult) holding desirable jobs, as a percentage of the working-age population; the y-axis represents the share of that group in the working-age population. Group characteristics are abbreviated as follows: Urb = urban; Rur = rural; F = female; M = male; Primary = primary education or below; Skill = secondary education or above; A = adult (35–64); Y = youth (15–34). For instance, 79 percent of urban men ages 35–64 with secondary education or above were in well-paid or protected jobs in 2009 Egypt, and they represented 6 percent of the working-age population. INEFFICIENT AND INEQUITABLE LABOR MARKE TS: A LOW-PRODUC TIVIT Y EQUILIBRIUM 59 socioeconomic groups in the working-age FIGURE 1.21 Percentage of individuals and households with population of Egypt, Jordan, and Morocco “desirable” jobs in the Arab Republic of Egypt and Jordan, 2009 hold high-paying or protected jobs. Groups and 2010 are constructed along clusters combining percent four dichotomous dimensions: gender, urban or rural location, young or prime age, and a. Egypt, Arab Rep. b. Jordan high versus low skilled. Individuals in labor force Individuals in labor force The figures suggest that while the alloca- with desirable jobs with desirable jobs tion of “desirable jobs” is not uniform across the population and displays important varia- tions across countries, young people, women, 25.5 27.2 and rural, low-skilled workers in general are less likely to have such a job. In contrast, 74.5 prime-age men are most likely to have a 72.8 desirable job. The reasons for this outcome are specific to each group and will be exam- ined in chapter 2. Using the same defi nition, only a minor- Households with at least one Households with at least one member having a desirable job member having a desirable job ity of the labor force has access to desirable jobs (from 27 percent in Jordan to 26 percent in Egypt; see figure 1.21), but, interestingly, the fact that male adult breadwinners are those who benefit most from this situation 39.4 43.3 ensures that the benefits from these scarce 60.6 56.7 jobs accrue to a large number of house- holds through cohabitation patterns. It is also important to point out that many of these coveted jobs, while desirable from With Without an individual standpoint, may actually be suboptimal for society. In MENA coun- Sources: The Arab Republic of Egypt’s SYPE 2009; and based on Jordan’s LMPS 2010. tries, most of these “desirable” jobs are in the public sector, which is character- ized by weak governance and low intrinsic productivity. with workers’ individual incentives, move This equilibrium is a mixed blessing for the MENA labor markets into a low-productiv- region. On the positive side, it keeps inequal- ity equilibrium. To increase the chances of ity as a whole moderate in MENA countries securing a public sector job, young people and can be considered as offering a safety net, choose higher education degrees that are given that about 6 of every 10 households in not relevant to the private sector; then those Egypt and Jordan have at least one individual who can afford to wait spend time queuing with a desirable job (figure 1.21). The draw- for a public sector job, with the expecta- back is that it sustains relatively high reser- tion of relatively high pay for low intrinsic vation wages among both women and youth productivity. Many women who have made in a large section of the population,6 which important investments in human capital is also reflected in high unemployment rates. choose not to participate in the labor mar- The specific determinants of this behavior ket at all. Given the fi scal constraints that among young people and women will be countries are facing and the rising share explored in detail in chapter 2. of highly educated people in the popula- In sum, the evidence shows that the tion, this model of labor market success is inefficient use of human capital, along unsustainable. 60 JOBS FOR SHARED PROSPERITY Notes Bourguignon, F., F. Ferreira, and M. Menéndez. 2007. “Inequality of Opportunity in Brazil.” 1. Given the weight of the public sector in over- Review of Income Wealth 53 (4): 585–618. all formal employment, changes in the size of Elbadawi, I., and N. Loayza. 2008. “Informality, the public sector are likely to affect overall Employment and Economic Development in informality trends, especially considering that the Arab World.” Journal of Development and growth in private formal employment remains Economic Policies 10 (2): 25–75. limited and has been insufficient to offset the Gatti, R., D. F. Angel-Urdinola, J. Silva, and A. recent downsizing of the public sector in many Bodor. 2012. “The Challenge of Informality countries (Rodman 2007). in the MENA Region.” Overview, World 2. This imperfect measure is used to compare Bank, Middle East and North Africa Region, workers across work status, in absence of Washington, DC. earnings data for the self-employed. Gindling, T. H., and D. Newhouse. 2012. “Self- 3. Because of the likely presence of omitted Employment in the Developing World.” Policy variables, the estimates of the Mincer regres- Research Working Paper 6201, World Bank, sion are probably biased. In particular, if Washington, DC. (unobservable) ability is positively correlated Grunwald, E., G. Lotz, K. Nitschke, and N. with formality, this coefficient is likely biased Sakr. 2009. “Vocational Education and upward. Gatti et al. (2012) use direct measures Training in the Context of Labour Mobility— of individuals’ cognitive and noncognitive Country Report: Egypt.” Manuscript, German ability for Lebanon and Syria in the attempt to Technical Cooperation, Cairo, Egypt. reduce these concerns. Their results show that, Hoyos, A., and A. Narayan. 2011. “Inequalities even after controlling for these variables, wage of Opportunities among Children: How Much gaps between the formal and the informal sec- Does Gender Matter?” Background papers tors persist and that the size of the coefficient for the World Development Report 2012, remains largely unchanged. manuscript, World Bank, Washington, DC. 4. Sample includes wage earners in urban areas ILO (International Labour Office). 2012. Global working between 30 and 60 hours per week. Employment Trends 2012: Preventing a Wage gaps were assessed through a simple Deeper Job Crisis. Geneva: ILO. http://www Mincer regression controlling for years of .ilo.org/wcmsp5/groups/public/@dgreports/@ education and experience. The model includes dcom m / @ publ /do c u ment s /publ ic at ion / a Heckman correction for selection into dif- wcms_171571.pdf ferent industries. Statistical analyses based ILO-KILM (International Labour Organization– on those nonrandomly selected samples can Key I ndicators of the Labour Market) lead to erroneous conclusions. The Heckman (database) (data for 1980 –2010). 7th ed. correction, a two-step statistical approach, International Labour Organization, Geneva, offers a means of correcting for nonrandomly Switzerland. http://kilm.ilo.org/kilmnet/. selected samples. Kuwait’s Annual Statistical Abstract. 2009. Central 5. Protected is defined as having social security; Statistical Office, Kuwait. “Unemployment well paid is defined as yielding either a wage in Kuwait: Facts and Figures” (database). above two-thirds of the wage distribution or http://mopweb4.mop.gov.kw/portal/page/ as being self-employed and residing in a house- portal/Mop_Portal/Home_En/ReleaseSector/ hold in the top two income quintiles. Statistical%20Studies%20Analysics. 6. This term is used in labor economics to define O m a n i L ab or S t at i s t ic s . Va r iou s ye a r s . the lowest wage rate at which a worker would LaborMaster.com. http://www.nationmaster be willing to accept a particular kind of job. .com/country/mu-oman/lab-labor. Rodman, S. 2007. “Good Jobs, Bad Jobs, and Economic Performance: A View from the References Middle East and North Africa Region.” In Barros, R., F. Ferreira, J. Molinas Vega, and J. Employment and Shared Growth, ed. Pierella Saavedra. 2009. Measuring Inequality of Paci and Pieter Serneels, 37–51. Washington, Opportunities in Latin American and the DC: World Bank. Caribbean. Latin America Development Roemer, J. 1998. Equality of Opportunity. Forum Series. Washington, DC: World Bank. Cambridge, MA: Harvard University Press. INEFFICIENT AND INEQUITABLE LABOR MARKE TS: A LOW-PRODUC TIVIT Y EQUILIBRIUM 61 Saudi Manpower Survey. 2009. Kingdom of Saudi ———. 2011. “Policy Report on Reforming Arabia, Central Department of Statistics and the Labor Admission System to the UAE”. Information. http://www.cdsi.gov.sa/english/ Manuscript, Washington, DC. index.php?option=com_content&view=article ———. 2012. International Income Distribution &id=50&Itemid=130. Database (I2D2) (database). Development World Bank. 2004. The Road Not Traveled: Research Group, Povert y & I nequalit y Education Reform in the Middle East and Unit. http://econ.worldbank.org/ WBSITE/ North Africa. Washington, DC: World Bank. EXTERNAL/EXTDEC/EXTRESEARCH/0 ———. 20 09. “Turkey Country E conomic ,,contentMDK:20699070~pagePK:64214825 Memora ndu m: I n for ma l it y: Causes , ~piPK:64214943~theSitePK:469382,00.html. Consequences, Policies.” Report No. 48523- ———. 2013. “Republic of Lebanon. Good Jobs TR, Washington, DC. Needed: The Role of Macro, Investment, ———. 2010. “The Arab Republic of Egypt: Education, Labor and Social Protection Gender Assessment 2010.” World Bank (MILES).” Washington, DC. and Egypt’s National Council of Women, Washington, DC. Women, Youth, and the Working Poor 2 Main findings • Progress in female participation in the labor force has been slow in MENA, particularly among the less educated. • Youth unemployment rates underestimate the extent of youth labor market disadvantage in MENA, because of high rates of discouragement among the lower educated and among young women. • Many factors such as limited decision making, an imperfect legal framework, and lack of enforcement of norms that protect women’s safety keep many women willing to work out of the labor force. • Lower-educated youth on average experience a longer transition to work than the higher educated. • Reservation wages above market wages and preferences for public sector jobs partially explain high unemployment rates, especially among better educated youth and educated women. • Working in poverty is most common among low-skilled rural workers. These workers represent 10 percent to 50 percent of the working population in the region; they tend to be informal and subsistence entrepreneurs. Women in recent decades has been substantial. Where MENA falls considerably short is on Female labor force participation indicators of women’s economic participa- in MENA tion (World Bank 2003). At 25.4 percent in T he achievements of the countries in the Middle East and 28.1 percent in North the Middle East and North Africa Africa, women’s labor force participation (MENA) in many areas of women’s (LFP) is lower than the 2008 world average well-being compare favorably with those of 51.6 percent. of other regions. Indicators such as female Over the past two decades, women’s labor education, fertility, and life expectancy force participation has risen in many MENA show that MENA’s progress in those areas countries (table 2.1), and in most countries 63 64 JOBS FOR SHARED PROSPERITY TABLE 2.1 Percentage increase in female labor force participation in selected economies in MENA by age group, 1990–2010 percent Economy All (ages 15–64) Ages 25–34 Ages 35–54 Jordan 71.6 78.0 48.0 West Bank and Gaza 52.5 72.4 44.1 Lebanon 37.0 55.3 45.0 Iraq 32.2 13.3 65.9 Tunisia 24.4 55.7 51.1 Morocco –2.6 7.1 12.7 Egypt, Arab Rep. –9.3 –18.7 9.7 Syrian Arab Republic –27.9 –17.2 0.6 Source: Based on International Labour Organization–Key Indicators of the Labour Market (ILO-KILM) database. Note: MENA = Middle East and North Africa. the highest increase occurred among the age economies in MENA have lower female group of younger women who have likely LFP rates than those expected given the completed their education (25–34), whose education and demographic structure of the participation rates are much higher than population. If female LFP in these coun- those of older women (35–64). tries were to rise to the level predicted by Despite this progress, rates of female women’s age and educational structure, participation in the labor force in MENA household earnings could increase by up to are not necessarily catching up with inter- 25 percent (World Bank 2003). These effects national levels. Figure 2.1 depicts rates in occur because female employment brings MENA in year 2010 (figure 2.1a) and their extra income to the household, decreases evolution compared to Europe and Central women’s dependence on their spouses and Asia (ECA) and Latin America and the other family members, and increases their Caribbean (LAC) (figure 2.1b). Results indi- intrahousehold decision-making power. cate that female LFP rates in MENA have At the same time, important externalities risen at the same speed as in LAC and only are linked to women’s economic empower- slightly faster than in ECA. If rates continue ment. The literature has found that working to rise at this pace, it will take the region women generally are more involved than 150 years to attain the current world average nonworking women in making the right (World Bank 2013). decisions about their children’s education, nutrition, and health (Angel-Urdinola and Wodon 2010). Why is female labor force participation It is important to emphasize that this important? chapter does not advocate for increasing Higher rates of female LFP are instrumen- female participation at all costs. Rather, it tal to generating economic growth and explores the constraints faced by women reducing poverty. International experience who would like to participate in the labor indicates that greater economic equality force. This analysis is particularly impor- between women and men is associated with tant because many women in the region poverty reduction, higher gross domes- want to work. Younger and more educated tic product (GDP), and better governance women have positive views about the suit- (Bardhan and Klasen 2000; World Bank ability of engaging in paid work and exhibit 2012b). Recent studies indicate that many a stronger desire to join the labor market. WOMEN, YOUTH, AND THE WORKING POOR 65 FIGURE 2.1 Female participation in the labor force in three world regions by age, 1990–2010 a. Labor force participation in ECA, MENA, b. Change in labor force participation in 80 and LAC, 2010 40 ECA, MENA, and LAC, 1990–2010 70 30 60 20 50 Percent Percent 40 10 30 0 20 –10 10 0 –20 lA d eq be d eb lA d eq be d eb ra an ib an ra an ib an hr hr hr hr sia an sia an nt e ar ca nt pe ar ca as ag as ag Ce urop e C ri e C eri M M Ce uro M M th me th Am E E A tin tin La La 25–34 35–54 Source: Based on the ILO-KILM. Note: Calculation based on averages. Mashreq refers to the Arab Republic of Egypt, Iraq, Jordan, Lebanon, the Syrian Arab Republic, and the West Bank and Gaza. Maghreb countries included are Morocco and Tunisia. ECA = Europe and Central Asia; LAC = Latin America and the Caribbean; MENA = Middle East and North Africa. Box 2.1 reflects some of these views and investments in education. Less educated aspirations. women, in contrast, generally look for jobs in the informal economy. Simple profi les of female unemployment for a selected group Women’s labor market outcomes of countries in MENA confi rm that female beyond participation unemployment is generally higher among Unemployment younger women, single women, and more As illustrated in figure 2.2, unemploy- educated women (table 2.2). Young, edu- ment rates are much higher among women cated women often queue for public sector than among men in almost all countries employment, as the private (often informal) in the region (reaching up to 50 percent in sector offers conditions that are below their some countries). In most of the developing reservation wages1 or quality (see Assaad world, women’s unemployment rates are 2006; Frankel 2010; Keddie 2007; World higher than men’s. Nevertheless, the gap Bank 2013). between female and male unemployment Skill mismatches also explain why unem- rates is particularly wide in MENA and ployment is disproportionately higher has doubled over the past 25 years (World among women. In MENA, the type of tech- Bank 2013). High female unemployment nical skills acquired by men and women rates can be explained largely by the fact differ considerably (figure 2.3). Women that women, especially university graduates are much more likely to specialize in the who are single, are willing to search for humanities or education—degrees that are employment (and can afford doing so for required for many public sector jobs but not longer periods) to obtain a return on their highly valued by the private sector—while 66 JOBS FOR SHARED PROSPERITY BOX 2.1 Women in MENA want to work “If I could use my time differently, I would trade free “We have ambitions and we would like to start time in the afternoon to study to become literate or our own projects; we want to succeed by ourselves, to learn a craft or work skill.” Female, urban Yemen not counting on the government or seeking help “ The main characteristic of a strong woman is to from associations.” Female, peri-urban Morocco be confident and participate. A strong woman can “I want to start a business; I am ambitious and lead and create an economic project.” Female, urban creative and would love to lead something of my Yemen own. I have no training though which might be a “I want to work, it is very important in my life. problem.” Young female, urban Egypt I cannot stay home watching television.” Female, “I want to be an entrepreneur; I want to do a rural Morocco master’s degree and then open a private project with “Women want to work even when married; we my friend.” Young female, Gaza want to have the fi nancial independence because men “Most of us [women] want to work but a problem cannot be trusted.” Female, peri-urban Morocco is commuting between home and work at night.” “Palestinian women have a long history of high Female, rural Morocco education and of working; a woman can work and Source: The quotations in this section draw on multiple sources. As part of is often even expected to work (but there are no jobs the background work for this report, focus groups were carried out in March at the moment). She can combine motherhood and 2011 in the West Bank and Gaza (Brodmann et al. 2012). Other focus groups and interviews were conducted as background work for La Cava et al. (2012a, work.” Female, West Bank 2012b). FIGURE 2.2 Unemployment rates by gender in selected economies in MENA, 2006–10 a. Youth unemployment rates, ages 15–25 b. Unemployment rates, ages 15–64 50 50 45 45 40 40 35 35 30 Percent Percent 30 25 25 20 20 15 15 10 10 5 5 0 0 t, A aza p. ic ria an sia co n ria , A aza Re p. ic ria an sia co n no no bl bl Re ra b Re oc oc rd rd ge ge ni ni G G pu pu ba ba Tu Tu Jo Jo or or b Al Al nd Eg and Re ra n A ra Le Le M M ka b b k ra Sy ypt an an yp nA tB tB Eg ria es es W W Sy Men Women Source: Based on ILO-KILM 2012 for Algeria, Iraq, Lebanon, the Syrian Arab Republic, the West Bank and Gaza, and the Republic of Yemen and on the Arab Republic of Egypt’s Labor Market Panel Survey (LMPS) 2006, Jordan’s LMPS 2010, Morocco’s Labor Force Survey (LFS) 2009, and Tunisia’s LFS 2010. See the appendix for more information on these surveys. Note: MENA = Middle East and North Africa. more men tend to study science and engi- Wage gaps neering (World Bank 2013). Differences in Although the average employed woman is technical skills between men and women more educated than the average employed lead to occupational segregation, hindering man (World Bank 2013), data on wages from women’s ability to fi nd suitable employment the Arab Republic of Egypt, Jordan, and the in the private sector. West Bank and Gaza indicate that—net of WOMEN, YOUTH, AND THE WORKING POOR 67 age, education, and experience—a gender TABLE 2.2 Basic unemployment profile of women ages 15–64 for the wage gap exists in favor of male workers. 2 Arab Republic of Egypt, Jordan, Morocco, and Tunisia, 2009–10 While the magnitude of the gender wage gap Egypt, Arab appears to be highly heterogeneous across Tunisia Rep. Jordan Morocco the region (figure 2.4), it is generally much Strata larger in the private sector (reaching up to All 19.1 21.6 19.7 9.73 40–80 percent in Egypt and the West Bank Urban 19.6 23.2 17.1 19.93 and Gaza). Likely explanations for the gen- Rural 17.9 20.3 30.8 1.75 der wage gap in the private sector include Age groups 15–24 32.7 44.2 42.9 16.21 occupational segregation (women work dis- 25–34 26.1 29.4 18.9 15.51 proportionately in low-paying sectors) and 35–64 5.2 3.1 6.4 33.48 discrimination (see discussion in chapter 1 Marital status and World Bank 2013). Married 8.5 15.4 14.2 2.13 As one might expect, gender wage gaps Not married 28.5 32.5 24.9 17.49 are lower in the public sector where there Education level are more concerns about equity and less Primary or below 12.9 1.8 13.0 2.92 tolerance of discrimination. In fact, in the Preparatory/ public sector of the West Bank and Gaza, secondary, general/ secondary, vocational 15.5 34.9 19.9 24.80 women actually earn more than men on Tertiary 32.0 29.1 22.5 23.44 average, possibly because the best female workers self-select into public sector jobs. Source: Based on the Arab Republic of Egypt’s Survey of Young People in Egypt (SYPE) 2009, Jordan’s LMPS 2010, Morocco’s LFS 2009, and Tunisia’s LFS 2010. See the appendix for more information on these surveys. What determines women’s low participation in the labor force? FIGURE 2.3 Educational specializations in countries in MENA, 2010 Economic factors One body of literature claims that economic Social sciences, business and law reasons account for the low levels of female Services participation in the labor force: women’s employment decisions are based on the Science quality of available opportunities, includ- ing market wages. According to this argu- Humanities and arts ment, women would enter the labor market Health and welfare if their market wage were above their res- Engineering, manufacturing, ervation wage (see Rauch and Kostyshak and construction 2009; World Bank 2013), corresponding Education to the value of their household chores and caretaking. Agriculture Consistent with this explanation, a 0 20 40 60 80 100 number of studies have pointed out that Percent low levels of female LFP in MENA are Female dominated Male dominated explained mainly by very low participation rates among women without tertiary educa- Source: World Bank 2011. Note: MENA = Middle East and North Africa. tion, who still constitute the largest share of the working-age population (Paterno, Gabrielli, and D’Addato 2008; Chamlou, LFP rates are very low for all other educa- Muzi, and Ahmed 2010). In fact, while LFP tional groups (figure 2.5). Women without rates among educated women in MENA university education, especially in urban are comparable to those in more developed areas, can obtain only jobs that offer low economies (at approximately 60 percent), wages, require long and hard hours, and 68 JOBS FOR SHARED PROSPERITY FIGURE 2.4 Male-female wage gap in selected economies Beyond high reservation wages, employ- in MENA, various years, 2006–10 ment quality is also a major determinant of female participation in the labor force. 100 Safety concerns about the workplace, for instance, limit women’s employment 80 options relative to men’s. Safety issues make 60 women reluctant to travel or live alone to access employment. These limitations dis- Percent 40 proportionately affect women in remote, 20 rural, or less prosperous areas. As a young woman in rural Yemen notes, “Distance 0 does not play a role in finding a job for men, because they can travel if they want. It –20 does play a role for a woman” (World Bank All workers Public sector Private sector Egypt, Arab . Rep. Jordan West Bank and Gaza 2013). Also, as a woman in rural Yemen observes, “even if we wanted to we cannot Source: Based on the Arab Republic of Egypt’s LMPS 2006, Jordan’s LMPS 2010, and the West Bank work because not all these jobs are suitable and Gaza’s LFS 2008. See the appendix for more information on these surveys. for women’s nature, nor does it suit customs Note: The sample is urban workers working between 30 and 60 hours per week. MENA = Middle East and North Africa. and traditions” (World Bank 2013). Ha rassment in the workplace also remains a serious problem in some coun- provide no social security. Indeed, in coun- tries. In their report on the major challenges tries like Egypt and in urban Morocco, related to youth employment in Egypt, about 70 percent of all unskilled women Assaad and Barsoum (2007) illustrate how hold jobs in the informal sector. 3 fears of workplace harassment discour- While the expected market wages for age many women from joining the labor most women remain low, the structure of force, especially in smaller workplaces and households in MENA contributes to keep- in less traditional jobs. Only four countries ing reservation wages high. In the particular in the region (Algeria, Iraq, Morocco, and case of oil-producing countries, reservation Tunisia) have legislation that deems sexual wages are high, thanks to a wide array of harassment in the workplace a criminal subsidies. In the case of middle-income coun- act and allows women to prosecute claims tries like Egypt and Jordan (as discussed in in court. These legal provisions are a good chapter 1), the majority of households have start, but implementation remains a chal- at least one individual (generally a prime-age lenge. Legal systems may not be strong male) in a protected or well-paid job. That enough to create an effective deterrent job serves as a safety net for the whole fam- against crimes, and many women are reluc- ily and contributes to keeping reservation tant to report harassment. wages high. Other economic factors, such as As a result, women tend to prefer pub- urbanization and the business cycle, affect lic sector jobs as they are more conducive female employment. For instance, Paterno, to family life, often offering fl exible work Gabrielli, and D’Addato (2008) show that hours, leave, and job security. Such jobs female participation is generally higher in may also be perceived as safer than private urban areas, because women have opportu- sector jobs. Indeed, qualitative research nities to access wage work in the service and from Upper Egypt indicates that for many industry sectors (for a review of the litera- women, government jobs are the only ture, see Assaad 2006; Clancy-Smith 1999; “acceptable” form of employment (World Frankel 2010; Keddie 2007; Moghadam Bank 2011). As discussed in chapter 1, how- 1998; Rodary 2007; Ross 2008; and World ever, the demand for public sector jobs far Bank 2013). outstrips supply. The comments of a woman WOMEN, YOUTH, AND THE WORKING POOR 69 FIGURE 2.5 Female participation in the labor force in selected countries in MENA by educational attainments, 2006–10 a. Female labor force participation b. Female working age population 100 rate by education level 90 80 100 70 80 Percent 60 Percent 50 60 40 30 40 20 20 10 0 0 Tunisia Iraq Egypt, Arab Jordan Morocco Tunisia Iraq Egypt, Arab Jordan Morocco Rep. Rep. Primary or below Secondary or below Preparatory/secondary, general/secondary, vocational Tertiary Tertiary Source: Based on the Arab Republic of Egypt’s LFS 2010, Iraq’s Household Socioeconomic Survey (HSES) 2006, Jordan’s LMPS 2010, Morocco’s LFS 2009, and Tunisia’s LFS 2010. See the appendix for more information on these surveys. Note: Rates and composition are given as a percentage of the working-age population. MENA = Middle East and North Africa. participating in a focus group in Morocco studies mention religion and culture as illustrate the point: “Many female graduates determinants of low female labor force only want public jobs. They refuse to look participation in Muslim countries (see for other options and need to change this Keddie 2007; Rauch and Kostyshak 2009; mentality since the government’s capacity to Rodary 2007); yet evidence seems to iden- recruit is limited.” tify other social factors as equally impor- Finally, women usually have smaller pro- tant determinants. Indeed, for most women fessional networks than men and are less in the region, “family reasons” are a main active job seekers. Families generally call explanation for inactivity (table 2.4). The upon their wasta (or social connections) share of women claiming that they are to help their male relatives find employ- inactive due to family reasons ranges from ment (Furstenberg and Kaplan 2004). Data 85 percent in urban Morocco to 96 percent from the 2009 Survey of Young People in in Egypt. Of course, some of this infor- Egypt show that men and women use very mation may leave out important nuances different approaches to finding employ- in women’s choices, preferences, and con- ment (table 2.3). Women were dispropor- straints. Most surveys in the region do not tionately likely to enter lotteries for public explicitly ask inactive women whether they sector jobs and to register in labor offices, would be willing to work. An exception while men were more likely to actively was the Morocco Household and Youth respond to advertisements, contact employ- Survey (La Cava et al. 2012a), in which ers directly, and ask friends or relatives for young women reported the lack of parental advice or help. or spousal permission as the main reason that they did not want or did not expect to Social factors work (figure 2.6). Social factors are also a fundamental Nevertheless, a woman’s path to success determinant of female employment in the and self-realization in MENA is not neces- region (see Bourqia 2010; Offenhauer sarily associated with participation in the 2005; Moghadam 2004; Rachik 2006; labor force. Results from the World Values World Bank 2013; Youssef 1972). Some Survey suggest that women in the MENA 70 JOBS FOR SHARED PROSPERITY TABLE 2.3 Job search methods used by young people in the Arab region are more likely than women in other Republic of Egypt, 2009 parts of the world to view the role of full- percent time domestic work as favorably as that Method Men Women of working for pay (World Bank 2013). A young female from rural Morocco observes, Entered government job lottery competition 9.7 29.3 “In our modern society, girls and boys face Sent job application 26.3 24.8 Inquired at work location 20.7 8.8 the same difficulties because we need to Applied to a job advertised in newspapers 13.1 5.7 work and help the family. But boys are more Asked friends or relatives for help 40.4 22.1 pressed because he is the man and he needs Contacted employer 17.6 3.9 to fi nd a job to support the family….A girl Registered at a government or labor office 9.7 21.1 can always marry” (World Bank 2011). Used a regular phone 47.2 24.5 Traditional norms see men as the primary Used a mobile phone 44.0 19.7 breadwinners in the family unit. While many Source: The Arab Republic of Egypt’s SYPE 2009. See the appendix for more information on this families may support a woman’s desire to survey. work, her work is not seen as mandatory. Note: This table shows major job search methods or those in which there are significant gender differences. On the contrary, a man’s identity is strongly linked to his role in the labor market, and unemployment is more stigmatized for men than for women. For this reason, women can rely on other family members (par- ents, husbands) and wait for a suitable job. TABLE 2.4 Reasons for being out of the labor force among Results from the World Values Survey also working-age individuals in the Arab Republic of Egypt, Iraq, indicate that a majority of MENA women urban Morocco, and Tunisia, 2006–10 agreed with the statement that “when jobs percent are scarce, men should be given priority Women (ages 15–64) over women.” Egypt, Arab Urban The following excerpt from a mixed focus Rep. Iraq Tunisia Morocco group in Morocco illustrates two aspects of Family reasonsa 96.1 94.0 91.1 85.5 the ongoing cultural transition of women’s Old age/retired 1.2 0.3 — 8.4 role in the labor market: Illness/disability 1.2 1.3 4.7 3.7 Female participant: Girls are com- Other 1.5 4.4 4.2 2.4 pletely different. T hey are more Prime age males (ages 35–54) responsible than boys. And many girls Egypt, Arab Urban are engaged or married before the high Tunisia Rep. Morocco Iraq school diploma. So the unemployment Illness/disability 91.8 26.4 24.1 8.8 is not really a big problem. Because she Old age/retired — 63.2 67.1 39.6 will stay home [and] take care of hus- Family reasons 4.7 2.2 5.6 0.8 band and kids. Only the youth from Other 3.6 8.1 3.2 50.8 rich …[families] can work outside and Young males (ages 15–34) have high positions. Urban Egypt, Arab Female par ticipant: In my case, I hate Morocco Tunisia Rep. Iraq to be unemployed. I feel excluded, all Illness/disability 51.4 34.4 12.9 8.0 my friends are busy and I stay home Family reasons 30.7 7.0 12.3 1.2 watching TV and doing home duties. Other 17.9 58.5 74.8 90.8 I feel like I am already an old woman. Source: Based on the Arab Republic of Egypt’s LMPS 2006, Iraq’s HSES 2006, Morocco’s Household (La Cava et al. 2012a) and Youth Survey (HYS) 2010, and Tunisia’s LFS 2010. See the appendix for more information on these surveys. Note: Data apply to those who have left school. — = not available. Marriage may also limit some women’s a. “Family reasons” groups answers such as pregnancy, taking care of children, and housewife. ability to make their own choices about WOMEN, YOUTH, AND THE WORKING POOR 71 whether to work. Women who still live at brief careers. Results from surveys con- home with their parents may enjoy greater ducted in Amman, Cairo, and Sana’a (see freedom to work than women who are mar- Chamlou, Muzi, and Ahmed 2010) confi rm ried. Another explanation could include the these results. The authors fi nd a significant social preference for the male as the main or negative association between being married sole provider for a couple or family (Paterno, Gabrielli, and D’Addato 2008). FIGURE 2.6 Main reason given by females ages 15–29 in Morocco Figure 2.7 illustrates the important effect for not wanting to work, 2010 of marital status on female employment. percent With the exception of Egypt, single women are noticeably more likely to participate in the labor force than married women. While it is not uncommon to see a nega- tive effect of marriage on female employ- 12.2 ment, this effect is much stronger in MENA 30.6 than in other regions (see Jaumotte 2003). Similarly, a greater number of infants in 22.9 the household will likely reduce women’s participation. A female in urban Morocco explains the phenomenon in the following 11.1 23.2 way: “The majority of girls say that their objective for fi nding a job is to be able to buy better clothes so that she eventually will fi nd a good husband. Once she has found Husband won't allow Parents won't allow a husband she will mostly quit her job” Social norms Busy at home (World Bank 2013). One common expec- Other tation is that women will leave the work- force upon marriage and will have only Source: La Cava et al. 2012a, using Morocco’s HYS 2010. FIGURE 2.7 Female participation in the labor force, by marital status and number of infants in the household, in the Arab Republic of Egypt, Iraq, Jordan, and Tunisia, 2006–10 a. Female labor force participation rate b. Female labor force participation rate by number of by marital status infants (ages 0–5) in the household 45 45 40 40 35 35 30 30 Percent Percent 25 25 20 20 15 15 10 10 5 5 0 0 Tunisia Iraq Egypt, Jordan Tunisia Iraq Egypt, Jordan Arab Rep. Arab Rep. Never married Married Divorced/widowed 1 2 3+ Source: Based on the Arab Republic of Egypt’s LMPS 2006, Iraq’s HSES 2006, Jordan’s LMPS 2010, and Tunisia’s LFS 2010. See the appendix for more informa- tion on these surveys. 72 JOBS FOR SHARED PROSPERITY and having children and female participa- “work” based on whether or not she makes tion in the labor force, highlighting that the her own decisions about her work status. “marriage factor” is much stronger than the The difference between these two prob- “children factor.” abilities is used as a proxy of the “extra” Data from Morocco make it possible to likelihood that a woman will work because identify women’s decision-making patterns she can make her own decisions. Our in relation to employment. Figure 2.8a plots fi ndings indicate that, all things equal, if the share of women (ages 15–45) who claim women can make their own decisions, they that they can determine their own employ- are between 10 and 15 percentage points ment status by age and marital status. After more likely to work. Results hold for both marriage, only a minority of women (about single and married women. Nevertheless, 15 percent) can make their own work- female decision-making patterns are likely related decisions. Husbands or other family to vary across countries in MENA and thus members make decisions about work for the these results from Morocco should not be remaining 85 percent. For single women, generalized. the pattern is somewhat different: not only Finally, cross-country evidence sug- are they more likely to make decisions on gests that the female role model at home the own, but their decision-making power largely determines a woman’s decision to increases rapidly with age. Interestingly, participate in the labor force. Data from even for married women without children, Egypt, Iraq, Jordan, urban Morocco, and decision-making power remains weak, sug- Tunisia all point to an important asso- gesting that the marriage effect is stronger ciation between female employment and than the children effect.4 the employment status of the female role Figure 2.8b suggests that when women model in the household (proxied by the are allowed to decide by themselves on their household head’s spouse). Women who labor market status, they tend to partici- belong to households where the spouse pate more in the labor force. First, we esti- of the male household head is employed mated the probability that a woman will are highly likely to be active (they exhibit FIGURE 2.8 Women’s decision making and labor force participation by marital status and age in Morocco, 2010 a. Labor force participation b. Women’s decision to work 100 60 90 50 80 in the labor force % participating 40 70 60 Percent 30 50 20 40 10 30 0 20 10 cid t cid t es es de s no de s no cid id e e ec 0 e e de do do d e: d: 14 17 20 23 26 29 32 35 38 41 44 e: d: gl rie gl rie Sin ar age Sin ar M M Can decide, single Active, single Ages 15–24 Ages 25–34 Ages 35–44 Can decide, married Active, married Source: Based on Morocco’s HYS 2010. See the appendix for more information on this survey. WOMEN, YOUTH, AND THE WORKING POOR 73 pa r t ic ip at ion rat e s b e t we e n 18 a nd to return to the workplace after child- 31 percent), whereas women who live with birth, these policies have the unintended female role models who are not in the consequence of raising the cost of hiring labor market will generally remain out of women. First, fi rms are often required to the labor force, too (in all countries but pay for part or all of a woman’s maternity Tunisia) (fi gure 2.9). This fi nding is con- leave, and fi rms must also hold a position sistent with that of Fernández, Fogli, and open for her until she returns to work. By Olivetti (2004), who show that the work- raising the cost of employing women rela- ing behavior of a man’s mother has a large tive to men, these laws may discourage and significant impact on the likelihood fi rms from hiring women. Jordan recently that his wife works. Farré and Vella (2007) amended its social insurance law to also fi nd that a woman’s attitudes toward address this problem; maternity benefits female employment have a statistically sig- are now be provided by social security, nificant effect on how her children perceive funded by a payroll tax on both male and women engaged in work and in turn influ- female employees. Second, fi rms in some ence female labor market decisions. countries are required to pay maternity benefits only when they employ more than Legal and institutional factors a certain number of female workers. This Finally, some laws and regulations also law can have the perverse incentive of dis- discourage or prevent women from having couraging a fi rm from hiring additional access to the labor market: women employees in order to remain below the threshold. • Laws guaranteeing highly protective • Laws restrict women from working at maternity leaves and child care benefi ts night and limit the industries in which impose extra costs on firms that hire they can work. Of 113 countries surveyed women. All countries in the MENA region by the Economist Intelligence Unit, 11 had provide for some form of maternity leave, restrictions on female employment in jobs and some also have provisions for child that were considered “against women’s care.5 Although designed to enable women FIGURE 2.9 Female participation in the labor force, by employment status of the household head and spouse, in the Arab Republic of Egypt, Iraq, Jordan, urban Morocco, and Tunisia, 2006–10 a. Female labor force participation rate by employment b. Female labor force participation rate by employment status of the head of the household status of the spouse of the head of the household 45 45 40 40 35 35 30 30 Percent Percent 25 25 20 20 15 15 10 10 5 5 0 0 Tunisia Iraq Egypt, Jordan Urban Tunisia Iraq Egypt, Jordan Urban Arab Rep. Morocco Arab Rep. Morocco Not in the labor force In the labor force Source: Based on the Arab Republic of Egypt’s LMPS 2006, Iraq’s HSES 2006, Jordan’s LMPS 2010, Morocco’s HYS 2010, and Tunisia’s LFS 2010. See the appendix for more information on these surveys. 74 JOBS FOR SHARED PROSPERITY morals” (UN Women 2011). Many of approval it would be very difficult” these countries are in the MENA region. (World Bank 2012b). Egypt, Kuwait, Morocco, Saudi Arabia, • Women have limited ownership rights Syria, the United Arab Emirates, and the to household capital. The most common Republic of Yemen all have restrictions property ownership arrangement in the on work that may be “morally harmful” region is the separate property regime, or affect women’s social standing. Some whereby a woman owns only the personal countries also restrict female employ- assets that she brings to, or acquires dur- ment during night shifts. Such restrictions ing, the marriage. This arrangement typi- are intended to protect women, but in cally excludes land and housing, which industries like accommodation and res- are purchased and owned by men. If taurants they greatly curtail women’s job women do not have ownership rights over opportunities. these fi xed family assets, they are limited • G u a rd i a n shi p l a ws l i m i t wo m e n’s in the extent to which they can borrow employment options. Some countries in for their own enterprises. Some coun- MENA require women to obtain a male tries now provide the option of a “com- guardian’s express permission to accept munity property regime” under which all a job; incorporate a business; apply for a property acquired during the course of passport, driver’s license, or loan; travel; the marriage is jointly owned.6 However, or even leave the house (table 2.5). Even this regime is normally only optional, and if permission can be obtained easily in women may be reluctant to request it at most cases, the process limits women’s the time of marriage. freedom of movement and autonomy Even where law s have been reformed and makes it more diffi cult for them to or introduced to level the playing field for work effectively, look for work, or carry women, implementation is often selec- on a business. As a woman in urban tive or absent. The majority of countries Yemen notes: “If a woman wants to in the MENA region have laws against start a business without her husband’s TABLE 2.5 Summary of legal restrictions on women in selected economies in MENA, 2012 Unmarried women Married women Women are not able to do the following in the same way as men: Apply for a passport Saudi Arabia Egypt, Arab Rep.; Iran, Islamic Rep.; Jordan; Kuwait; Oman; Saudi Arabia; United Arab Emirates; Yemen, Rep. Travel internationally Saudi Arabia Iran, Islamic Rep.; Saudi Arabia; Syrian Arab Republic Travel domestically Iran, Islamic Rep.; Saudi Arabia; Yemen, Rep. Get a job or pursue a trade Kuwait, Oman Iran, Islamic Rep.; Jordan; Oman; United Arab Emirates Sign a contract ¸ ¸ Register a business ¸ ¸ Open a bank account ¸ ¸ Women must obey their husbands n.a. Egypt, Arab Rep.; Jordan; Lebanon; Saudi Arabia; Syrian Arab Republic; United Arab Emirates; West Bank and Gaza; Yemen, Rep. The constitution: Bans gender discrimination Algeria, Oman, West Bank and Gaza Guarantees gender equality Algeria; Egypt, Arab Rep.; Iran, Islamic Rep.; Jordan; Kuwait; Lebanon; Morocco; Oman; Syrian Arab Republic; Tunisia; United Arab Emirates; West Bank and Gaza; Yemen, Rep. Source: Women, Business and the Law Database 2012 (http://wbl.worldbank.org). Note: n.a. = not applicable; MENA = Middle East and North Africa; ¸ = Permitted in all MENA countries. WOMEN, YOUTH, AND THE WORKING POOR 75 discrimination in the workplace and guaran- FIGURE 2.10 Framework for interpreting the labor market tees of equal pay for equal work. But with- disadvantage out enforcement, these laws are powerless to protect women. Entrenched social prejudices about women’s abilities, the safety of the Queuing workplace, and so on may continue to deter women from taking jobs in the private sector (or their family from allowing them to do so). Unemployed Only visible enforcement of these protections with no job offer will change attitudes. Labor Discouraged Working poor Youth Willingly market disadvantage inactive The previous chapter showed that unem- Inactive Informal ployment, out-of-school inactivity, and Forbidden informality tend to affect young people dis- proportionately. This chapter analyzes these Student Entrepreneur states, taking into account the evidence on individual preferences and constraints, and shows that for the large majority of young people in MENA, working informally or being out of work is not the result of choice but of lack of alternatives that are either the labor force and could disguise invol- better in a strict economic sense or socially untary inactivity. Figure 2.11a shows that acceptable. NEET rates in MENA countries are per- The analytical approach adopted in this vasive (fi gure 2.11a) and that they can be section acknowledges that each outcome much higher than unemployment rates (see could be the result of a process of selec- fi gure 2.11b), particularly for those with tion or rather of outright exclusion (see little education. For instance, while the figure 2.10). For instance, some youth may unemployment rate among youth with pri- choose to work in the informal sector to mary education (ages 15–24) in Egypt and become a successful entrepreneur, while Iraq is 6 and 18 percent, respectively, 58 others may be constrained in their ability and 57 percent in the same reference group to join wage employment. are NEET. Figure 2.12 shows that within this large group of NEET youth, those out of the labor NEET rates best define young people’s force are more than those in unemploy- labor market disadvantage ment. While this fact is not surprising for Unemployment is conventionally singled young women—given historical trends in the out as evidence of youth disadvantage in region—it is also a prevailing pattern among the labor market. However, the share of out-of-school young males in many coun- youth neither in education, employment, tries, including Egypt, Iraq, Lebanon, and nor training (NEET) is now considered a the Republic of Yemen. more accurate measure of labor market According to available evidence, much disadvantage, because it simultaneously of the inactivity among youth who are not captures the extent to which young people in school is not voluntary. Results from a are neither participating in the labor mar- recent Youth Survey in Morocco showed ket nor building skills for participating that about 60 percent of all inactive young in the future.7 Unemployment rates, for men and about 23 percent of all inactive instance, are sensitive to participation in young women were actually discouraged 76 JOBS FOR SHARED PROSPERITY FIGURE 2.11 NEET and unemployment rates in selected economies in MENA for individuals ages 15–24 a. NEET rates by educational level, b. Unemployment rates by educational 2005–10 level, 2005–10 Morocco Morocco Yemen, Rep. Yemen, Rep. West Bank West Bank and Gaza and Gaza Tunisia Tunisia Jordan Jordan Iraq Iraq Egypt, Arab Rep. Egypt, Arab Rep. 0 10 20 30 40 50 60 70 0 10 20 30 40 50 60 70 Percent Percent Tertiary Secondary Primary Tertiary Secondary Primary Source: Based on the Arab Republic of Egypt’s SYPE 2009, Iraq’s HSES 2006, Jordan’s LMPS 2010, Morocco’s LFS 2009, Tunisia’s LFS 2010, the West Bank and Gaza’s LFS 2008, and the Republic of Yemen’s Household Budget Survey (HBS) 2005. Note: NEET = neither in education, employment, nor training; MENA = Middle East and North Africa. FIGURE 2.12 Labor market status of NEET youth in selected workers (La Cava et al. 2012a). Similarly, economies in MENA, by gender, 2005–10 the survey on youth in Upper Egypt shows a high rate of discouragement among inac- 80 tive youth (La Cava et al. 2012b). Although the additional detrimental effects of job- % out of school / out of work (ages 15–24 ) 70 lessness in MENA are not covered in detail 60 in this report, it is important to highlight them; they include delays in marriage; 50 increase in risky behaviors, such as sub- stance abuse for young men and prostitu- 40 tion for young women; greater exposure to 30 violence, including political violence; and unsafe migration (Assaad and Barsoum 20 2007; Boudarbat and Ajbilou 2007; Kab- bani and Kamel 2007; Silver 2007; World 10 Bank 2007, 2012a, 2012b). 0 F M F M F M F M F M F M F M A slow school-to-work transition Iraq Yemen, Egypt, Jordan West Bank Tunisia Lebanon Rep. Arab Rep. and Gaza High NEET rates among youth are mainly the result of a slow and incomplete transi- Inactive, out of school Unemployed tion from school to work. A proxy mea- Source: Based on the Arab Republic of Egypt’s LMPS 2006, Iraq’s HSES 2006, Jordan’s LMPS 2010, sure of the duration of the transition from Lebanon’s Employer-Employee Survey 2010, Tunisia’s LFS 2010, the West Bank and Gaza’s LFS 2008, school into employment in absence of cross- and the Republic of Yemen’s HBS 2005. See the appendix for more information on these surveys. Note: NEET = neither in education, employment, nor training; MENA = Middle East and North Africa; country data on individual work histories F = females; M = males. is given by the age in which a youth cohort WOMEN, YOUTH, AND THE WORKING POOR 77 has achieved a set employment rate (like to result from dynamics on both the labor 75 percent or 90 percent), 8 minus the pre- supply and the labor demand side. On the sumed age of completing education. Such demand side, historically sluggish demand a measure for MENA countries shows that for labor in the private sector and falling the transition to work is long overall, and hiring rates in the public sector have cur- relatively longer for youth with secondary tailed the number of jobs available relative education than for those with higher educa- to the number of labor market entrants. tion. For instance, in West Bank and Gaza, On the supply side, a subset of the youth secondary male graduates achieved only a population may be exclusively seeking a 75 percent insertion rate nine years after public sector or a highly paid private sec- graduation from secondary school com- tor job. pared to three years for university graduates Relatively high unemployment rates (see figure 2.13). among more highly educated youth are During the years of transition, unemploy- probably the result of their greater abil- ment seems to be more pervasive among ity to afford unemployment: they can wait the more highly educated individuals, while for the appropriate job, compared to the inactivity is higher for the less-educated less educated. As a fi rst indication of this ones. One explanation is that after complet- fact, the ratio of the unemployment rates of ing their studies, those with a tertiary edu- individuals with tertiary education to those cation appear to be searching for jobs more with secondary education is above 1 for intensely and to be choosier in the jobs they youth in many economies (see table 2.6), accept. But after some years, they have the suggesting that unemployment is higher ability to settle with a job. In contrast, job among the more highly educated. However, seekers with lower education appear to have among adults, the ratio is below 1. This a much harder time finding a job, and the finding may be a signal that labor mar- prevalence of unemployment decreases only kets are not systematically biased against partially even many years after they have tertiary-educated workers and that a subset completed their studies. of the young higher educated may actually prefer unemployment to some of the jobs available. Wage and job quality expectations In addition, a recent Gallup survey High unemployment rates—and their dif- shows that in 2009 in most countries in the ferentials by educational attainment—seem region (even in a context of job scarcity) FIGURE 2.13 Gap between school and work for young men in the Arab Republic of Egypt, Jordan, Tunisia, and the West Bank and Gaza, 2006–10 a. Average years for 90% of graduates to get a job b. Average years for 75% of graduates to get a job Egypt, Arab Rep. 7 Egypt, Arab Rep. 4 6 3 Tunisia 15 Tunisia 7 10 7 Jordan 5 Jordan 3 3 1 West Bank and Gaza 16 West Bank and Gaza 9 7 3 0 5 10 15 20 0 2 4 6 8 10 Secondary Tertiary Source: Based on the Arab Republic of Egypt’s LMPS 2006, Jordan’s LMPS 2010, Tunisia’s LFS 2010, and the West Bank and Gaza’s LFS 2008. See the appendix for more information on these surveys. 78 JOBS FOR SHARED PROSPERITY TABLE 2.6 Ratio of unemployment rates of tertiary educated and secondary educated in several economies in MENA, 2005–10 Iraq West Bank and Gaza Yemen, Rep. Egypt, Arab Rep. Morocco Jordan Tunisia Adults 0.62 0.44 0.56 0.61 0.86 0.78 0.91 Young 0.92 1.19 1.21 1.34 1.40 1.54 1.77 Source: Based on the Arab Republic of Egypt’s SYPE 2009, Iraq’s HSES 2006, Jordan’s LMPS 2010, Morocco’s HYS 2010, Tunisia’s LFS 2010, West Bank and Gaza’s LFS 2008, and the Republic of Yemen’s HBS 2005. See the appendix for more information on these surveys. Note: MENA = Middle East and North Africa. FIGURE 2.14 Prevalence of refusing a job in selected economies in MENA, by educational level, 2009 70 60 50 Percent 40 30 20 10 0 ti s p. . ria ain sia za a co q it ic an r n ya ep ta te bi no Ira ou wa bl Re Ga Lib oc rd ge ni Qa ira hr ,R ra pu ba ib Ku Tu Jo or ab iA Al Ba nd en Em Dj Re Le M Ar ud m ka ab ab Ye t, Sa an yp Ar Ar tB Eg d n ite ria es W Un Sy Primary Secondary Tertiary Source: Based on Gallup World Poll. See the appendix for more information on the poll. Note: The polling question was, Have you ever refused a job? MENA = Middle East and North Africa. more than a third of individuals with ter- women reported for refusing a job offer is tiary education reported refusing a job the location of the job, which is evidence offer (figure 2.14), and these rates increase of how mobility constraints continue to uniformly with educational achievement. u nderm ine oppor t u nities for women. Among those who refused a job, concerns Finally, for the higher educated, find- with low wages play an important role for ing work with appropriate qualifications all educational categories; however, con- increases in importance relative to other cerns with job quality and the sector in groups of unemployed, but salary continues which the job is increase with the level of to prevail over quality as a main concern education (figure 2.15). for refusing a job. Detailed data on unemployed Tunisian In sum, both the reservation wage and youth (see appendix for LFS 2010) indicate the nonpecuniary attributes that make a that refusals for inadequate pay are par- job valuable increase in importance with ticularly an issue for males with primary education. education, who probably receive the low- Figure 2.16 shows that the minimum sal- est wage offers among job seekers. How- ary at which young Moroccans would accept ever, the top reason that young unemployed work increases with the level of education, WOMEN, YOUTH, AND THE WORKING POOR 79 FIGURE 2.15 Reason for refusing a job in resource-rich and resource-poor economies in MENA, by education and gender, 2010 a. By education b. By gender 50 50 45 45 40 40 35 35 30 30 Percent Percent 25 25 20 20 15 15 10 10 5 5 0 0 Primary Secondary Tertiary Primary Secondary Tertiary Primary Secondary Tertiary Women Men Women Men Women Men Wage was too low Below Constraints: Wages/salary Below Constraints: quality, education, location, hours too low quality/career location, hours expectations expectations MENA resource rich MENA resource poor Source: Based on Gallup World Poll. Note: MENA resource rich = Algeria, Bahrain, Iraq, Kuwait, Qatar, Saudi Arabia, and the United Arab Emirates. MENA resource poor = Djibouti, the Arab Republic of Egypt, Jordan, Lebanon, Morocco, Syria, Tunisia, the West Bank and Gaza, and the Republic of Yemen. MENA = Middle East and North Africa. even after controlling for household wealth, FIGURE 2.16 Average self-reported reservation wages for males location, and marital status. Men tend to and females ages 15–29 in Morocco, by educational level, 2010 report higher reservation wages than women. Focus group research (La Cava et al. 2012a) 4,500 has shown the strong differences in the mean- 4,000 ing of work for young men and young women 3,500 Moroccan dirhams (men feel compelled to be breadwinners, for 3,000 example), which can explain the systematic 2,500 differentials between men and women in 2,000 wages expectation. 1,500 A second observation is that across edu- 1,000 cational groups, young people on average 500 continue to expect higher wages than those 0 observed in the market. Young women with ne y l ry y y na ar ar t da rsi No tio im nd less than a high-school diploma have the wid- ive on Pr ca co c Un Vo se Se est gap between actual earning potential and wer reservation wages, with a ratio of 1.6 (see Lo figure 2.17).9 Although returns to education Females Males do not seem to increase until an individual reaches the high-school level, young women Source: Based on Morocco’s HYS 2010. The figure reports the predicted self-reported reservation with basic education tend to have higher wage among youth ages 15–29 willing to work, controlling for household wealth, location, and wage expectations than women without marital status. 80 JOBS FOR SHARED PROSPERITY education and generally above what the mar- guaranteed public sector employment to ket would offer them. their most educated population, driving up The prevalence of high reservation wages education rates but pushing public budgets has much to do with how perceptions of to the limit. Following structural adjust- public sector employment have shaped wage ment, the guaranteed employment policy and employment preferences among edu- was revoked, and public jobs were sub- cated labor market entrants. World Bank jected to hiring freezes. Figure 2.18 shows (2004) provides a comprehensive account that this adjustment occurred largely at of the public employment policy followed the expense of new entrants. In Egypt in by M ENA governments over the past 2006, a relatively larger share of prime- decades. Until the early 1990s, governments age males (above age 40) held formal sec- tor jobs compared to young people. Given the limited mobility across the formal and FIGURE 2.17 Relation of reservation wages to actual wages by informal sectors, it would appear that for- gender, males and females ages 15–29 in Morocco, 2010 mal employment has become relatively less frequent among the current youth genera- tion than it was for past generations. 1.8 6,000 Ratio of reservation wage 1.7 The current generous conditions of pub- Median market wage, 5,000 Moroccan dirhams 1.6 lic employment make queuing for a public to actual wage 1.5 4,000 1.4 job a rational, although frustrating, strat- 1.3 3,000 egy for educated job seekers.10 However, 1.2 1.1 2,000 there are indications that young people’s 1 1,000 perceptions of public sector jobs are start- 0.9 0.8 0 ing to shift at last, especially in countries with sustained reductions in public sector ne y l ry ry ty oo ar rsi da da No im ch ive n on hiring. Assaad, Binzei, and Gadallah (2010) ls Pr co ec Un na se rs io note that in Egypt the now-permanent r we pe c at Up Lo Vo reduction in prospects for attaining a pub- Reservation/actual, females Median wage, females lic job is translating into shorter unemploy- Reservation/actual, males Median wage, males ment rates among more highly educated groups. Similarly, opinion surveys suggests Source: Based on Morocco’s HYS 2010. that in almost all countries the preference FIGURE 2.18 Male and female age cohorts by labor force status in the Arab Republic of Egypt, 2006 a. Labor force status of males b. Labor force status of females 1,700 1,700 1,500 1,500 1,300 1,300 Thousands Thousands 1,100 1,100 900 900 700 700 500 500 300 300 100 100 0 0 16 20 24 28 32 36 40 44 48 52 56 60 16 20 24 28 32 36 40 44 48 52 56 60 Age Age Out of labor force Unemployed Informal wage earner Formal wage earner Source: Based on the Arab Republic of Egypt’s LMPS 2006. WOMEN, YOUTH, AND THE WORKING POOR 81 for private jobs, while low, is greater among (see figure 2.21). Surprisingly, in several young people than among their parents, countries unskilled rural employment is especially in resource-poor countries (see more prevalent among youth than among figure 2.19). adults. For instance, about 50 percent of employed youth in Morocco and 34 percent of employed youth in Tunisia are rural The Working Poor Mostly rural, low-skilled, and informal FIGURE 2.19 Percentage of youth ages 15–34 that prefer private A significant share of MENA’s workers live in sector jobs and gap between youth and adult preferences for poverty. Poverty is obviously present among private sector employment in selected economies in MENA, 2010 different social and geographic groups; yet in countries where recent measures of consump- Gap between youths and adults preferring private sector jobs tion and poverty are available, the data sug- 100 15 % of youths preferring 90 private sector jobs gest that in MENA working poverty peaks 80 10 70 5 among rural workers with primary educa- 60 50 0 tion or less (figure 2.20). Clearly this defi ni- 40 tion leaves out pockets of working poverty 30 –5 20 –10 in urban areas. There, however, poverty is 10 0 –15 less correlated with low skills than it is in ir ti h s Ku rain Re Ira t pu q ud Li lic i A bya Ye A abia ,R a Tu ep. k a Q sia G r yp M ord a t, A or an b o ba p. n nd ata i Ba ate en ri J az ra occ no Em ou wa Le Re b m lge rural areas, because urban areas have con- ni ab jib r Ar D siderable numbers of low-skilled workers b Sa ra an nA tB who are not poor. Eg d ite ria es W Un Sy Taken together, the rural low-skilled Youths, ages 15–34 (left axis) population constitutes much of the work- Youth-adult preference gap (right axis) ing population in many countries in the region, ranging from about 10 percent in Source: Based on Gallup World Poll. Jordan to more than 35 percent in Morocco Note: MENA = Middle East and North Africa. FIGURE 2.20 Share of workers in the Arab Republic of Egypt, Jordan, and Morocco who live in poor households, by urban or rural location, 2009–10 a. Informal workers, Jordan b. Informal workers, Morocco c. Informal workers, Egypt, Arab Rep. 60 60 60 % of workers in bottom % of workers in bottom % of workers in bottom consumption quintile consumption quintile consumption quintile 50 50 50 40 40 40 30 30 30 20 20 20 10 10 10 0 0 0 y co c y y y co c y y y co c y y si si si ar ar ar ar ar ar ar ar ar ba ba ba im nd rti im nd rti im nd rti y/ y/ y/ Te Te Te pr pr pr ar ar ar Se Se Se w w w im im im lo lo lo Pr Pr Pr Be Be Be Rural informal Rural formal Urban informal Urban formal Source: Based on the Arab Republic of Egypt’s SYPE 2009, Jordan’s LMPS 2010, and Morocco’s HYS 2010. See the appendix for more information on these surveys. 82 JOBS FOR SHARED PROSPERITY FIGURE 2.21 Education level of unskilled rural workers as a FIGURE 2.22 Prevalence of nonwage percentage of the working population in selected economies in employment among rural unskilled workers in MENA, 2006–10 selected economies in MENA, 2006–10 60 80 70 50 % of nonwage workers % of working population 60 40 50 30 40 30 20 20 10 10 0 0 an Ga nd t, A Iraq p. sia p. co Re Re oc rd an q sia co p. . za ka ni za ep Ira Re Tu Jo or Ga b en oc rd ni an ,R ra M Tu Jo or m ab tB nd en M Ye Ar m es ka yp Ye W t, an Eg yp tB Eg es W All other workers Rural unskilled Primary Below primary (unpaid/family worker/other) Rural unskilled Rural unskilled (employer) (self-employed) Source: Based on the Arab Republic of Egypt’s LFS 2010, Iraq’s HSES 2006, Jordan’s LMPS 2010, Morocco’s LFS 2009, Tunisia’s LFS 2010, the West Bank and Gaza’s LFS 2008, and the Republic of Yemen’s HBS 2006. Source: Based on the Arab Republic of Egypt’s LMPS 2006, Iraq’s HSES Note: MENA = Middle East and North Africa. 2006, Jordan’s LMPS 2010, Morocco’s LFS 2009, Tunisia’s LFS 2010, the West Bank and Gaza’s LFS 2008, and the Republic of Yemen’s HBS 2005–06. Note: MENA = Middle East and North Africa. unskilled workers, compared to 40 and 22 percent among adults. An additional dimension of working pov- social security mechanisms and remain erty, besides low earnings associated with highly vulnerable to economic shocks. low consumption, is that workers oper- The high propensity of self-employment ate largely unprotected from risk due to the among rural unskilled workers is also an informal nature of their employment. Gatti et indication of the diffi culty that this group al. (2012) have showed how in MENA infor- faces in finding a wage job. In MENA, entre- mality decreases as wealth increases, even preneurship continues to be the employment if in some countries informality is so wide- of last resort in the absence of wage jobs, spread that it also affects wealthier segments and for the low skilled, entry into entrepre- of the population. neurship is not as difficult as entering wage Compared to the overall working popula- employment. A recent Gallup survey among tion, rural unskilled workers are more likely existing and aspiring entrepreneurs found to be entrepreneurs (mostly of the subsistence that more than half of entrepreneurs said level) (figure 2.22). Most rural unskilled that they had become entrepreneurs because workers are unpaid family workers, employ- they could not find a suitable wage job ers, and especially self-employed rather than (figure 2.23). Yet, the same survey shows wage employees. Unpaid family work is far that on average less than 10 percent of those more prevalent among women than among “aspiring” to be entrepreneurs reported the men, while the opposite is true for entrepre- inability to find wage employment as the rea- neurship. As their fi rms are almost entirely son for their choice. This body of evidence informal, these workers are not eligible for suggests that entry into self-employment WOMEN, YOUTH, AND THE WORKING POOR 83 FIGURE 2.23 Percentage of those entering entrepreneurship in selected economies in MENA because of no suitable job in the wage labor market, by educational level, 2011 100 % answering "no suitable jobs" 90 80 70 60 50 40 30 20 10 0 ic an co q n p. a a . ti ep si z no Ira ou bl Re Ga oc rd ni ,R pu ba ib Tu Jo or b nd en Dj Re ra Le M m ka t, A b Ye ra an yp nA tB Eq ria es W Sy Primary Secondary Tertiary Source: Based on Gallup World Poll. See the appendix for more information on the poll. Note: Question: “Why did you become an entrepreneur?” Answer: “No suitable job.” MENA = Middle East and North Africa. (of the low-productivity type) is not as con- 3. According to the available microdata, infor- strained as fi nding a wage job, especially for mality rates among low-skilled women (defined unskilled individuals. However, as discussed as those with at most a secondary school edu- further in chapter 7, this is not necessar- cation) are 69 percent in Egypt, 43 percent in Iraq, 39 percent in Jordan, 67 percent in ily the case for higher-skilled workers who urban Morocco (estimates from the Morocco would need to forgo higher wage income to Household and Youth Survey), and 42 percent prefer entrepreneurship. in Tunisia. 4. Estimates using available microdata for Egypt, Notes Iraq, Jordan, Morocco, and Tunisia indicate that having additional women or seniors in 1. This term is used in labor economics to the household helps women increase their define the lowest wage rate at which a labor force participation, probably because worker would be willing to accept a particu- these household members can share chores lar kind of job. and child care. However, this effect is small, 2. Standard Mincerian regressions are used to suggesting that the lack of child care is not provide an estimate of the gender wage gap. as important a determinant of female partici- To avoid a strong selection bias, the analysis pation as marriage (results are available on that follows is restricted to those individuals request). living in urban areas who are wage earners 5. Women, Business and the Law Database and work full time (between 30 and 60 hours (2012), http://wbl.worldbank.org. a week). In their basic form, Mincerian regres- 6. Apart from gifts or inheritance specifically sions estimate the natural logarithm of hourly designated to one spouse, or property clearly wages as a function of a number of key observ- intended to be under sole ownership. able characteristics, such as gender, age (non- 7. This definition of youth joblessness was linear), years of education, and years of expe- prominent in the World Development Report rience. A dummy variable for male workers is 2007 (World Bank 2006), and has been then added to this standard specification. The widely used for MENA by La Cava et al. exponential of the coefficient on this variable (2012a, b), O’Higgins (2005), and Kabbani captures an estimate of the gender wage gap. and Kothari (2005). 84 JOBS FOR SHARED PROSPERITY 8. Such analysis is restricted to young males, who Educated Youth.” IZA Discussion Paper 6139, can be assumed (based on the sociological Institute for Study of Labor, Bonn, Germany. literature) to be all willing and unconstrained Bodor, A., and D. Robalino. 2009. “On The to enter into employment after completing their Financial Sustainability of Earnings-Related education. This measure assumes that second- Pension S chemes with ‘Pay-as-You- Go’ ary-educated students enter the labor market Financing and the Role of Government- around the age of 19 and that university gradu- I nd exe d B ond s .” Jo u r n a l of Pe n s i o n ates enter the labor market around age 23. Economics and Finance 8 (2) (April): 153–87. 9. To ensure comparability between self- Boudarbat, B., and A. Ajbilou. 2007. “Youth reported reservation wages and actual wages, Exclusion in Morocco: Context, Consequences, wages are normalized as hourly earnings and and Policies.” Youth Initiative Working Paper then transformed into monthly wages using a 5, Wolfensohn Center for Development, typical workweek. Brookings Institution, Washington, DC. 10. Even accounting for the forgone income due Bourqia, R. 2010. Values and Social Change to unemployment, lifelong returns to a public i n M o r o c c o [ Va l e u r s e t c h a n g e m e nt job in Morocco continue to be better than in social au Maroc] . Instituto Europeo del the private sector (Bodor and Robalino 2009). Mediterráneo (I E Med) Quaderns de la Mediterrània=Cuadernos del Mediterráneo, Nº. 13: 105–15. Brodmann, S., E. Cuadra, M. I. Allouche, and S. References A. Hillis. 2012. “From Education-to-Work: Angel-Urdinola, D., and A. Semlali. 2010. “Labor Opportunities and Challenges in the West Markets and School-to-Work Transition in Bank and Gaza.” Quick Notes Series 55, Egypt: Diagnostics, Constraints, and Policy World Bank, Washington, DC. Framework.” MPRA paper 27674, University C ha m lou , N ., S . Mu zi , a nd H . A h med. Library of Munich, Munich, Germany. 2010. “Female Home-Based Work and Angel-Urdinola, D., and Q. Wodon. 2010. Entrepreneurship in MENA: Evidence from “Income Generation and Intra-Household Amman, Cairo, and Sana.” Working paper, Decision Making: A Gender Analysis for World Bank, Washington, DC. Nigeria.” MPRA paper 27738, University Clancy-Smith, J. 1999. “A Woman without Library of Munich, Munich, Germany. Her Distaff: Gender, Work, and Handicraft A s s a ad , R . 2 0 0 6 . “ W hy D id E conom ic Production in Colonial North Africa.” In Liberalization Lead to Feminization of the Social History of Women and Gender in Labor Force in Morocco and De-Feminization the Modern Middle East , ed. J Tucker, 25. in Egypt?” In Gender Impact of Trade Boulder, CO: Westview Press. Liberalization in the MENA Region, 12–22. F a r r é , L . , a n d F. Ve l l a . 2 0 0 7. “ T h e Tunis: Center of Arab Women for Training Intergenerational Transmission of Gender and Research. Role Attitudes and Its Implications for Female Assaad, R., and Barsoum, G. 2007. “Youth Labor Force Participation.” IZA Discussion Exclusion in Egypt: In Search of ‘Second Paper 2802, Institute for the Study of Labor, Chances.’” Working paper 2, Middle East Bonn. Youth Initiative, Wolfensohn Center for Fernández, R., A. Fogli, and C. Olivetti. 2004. Development, Dubai School of Government, “Mothers and Sons: Preference Formation and Dubai, United Arab Emirates. Female Labor Force Dynamics.” Quarterly Assaad, R. A., C. Binzei, and M. Gadallah. 2010. Journal of Economics 119 (4): 1249–99. “Transitions to Employment and Marriage Frankel, J. 2010. “The Natural Resource Curse: among Young Men in Egypt.” Wolfensohn A Survey.” Harvard Kennedy School Faculty Center for Development, Brookings Institution, Resea rch Work i ng Paper RW P10 - 0 05, Washington, DC. Harvard Kennedy School, Cambridge, MA. Bardhan, K., and S. Klasen. 2000. “On UNDP’s Furstenberg, F. F., and S. B. Kaplan. 2004. Revisions to the Gender-Related Development “Social Capital and the Family.” In The Index.” Journal of Human Development and Blackwell Companion to the Sociology of Capabilities 1 (2): 191–95. Families, ed. J. L. Scott, J. K. Treas, and M. Binzel, C. 2011. “Decline in Social Mobility: P. Richards, 218–32. Malden, MA: Blackwell Unfulfilled Aspirations among Eg ypt’s Publishers. WOMEN, YOUTH, AND THE WORKING POOR 85 Gatti, R., D. F. Angel-Urdinola, J. Silva, and Countries.” Social Protection Discussion Paper A. Bodor. 2012. The Challenge of Informality 0321, World Bank, Washington, DC. in the MENA Region . Washington, DC: Paterno, A., G. Gabrielli, and A. V. D’Addato. World Bank. 2008. “Women’s Work, Family and Social ILO-KILM (International Labour Organization– Characteristics: Morocco’s Case” [Travail Key I ndicators of the Labour Market) des femmes, caractéristiques familiales et (database) (data for various years). 7th ed. sociales: le cas du Maroc]. Cahiers québécois International Labour Organization, Geneva, de démographie 37 (2): 263–89. Switzerland. http://kilm.ilo.org/kilmnet/. Rachik, H. 2006. “Synthesis report of the Jaumotte, F. 2003. “Female Labour Force National Su r vey on Values” [ R appor t Pa r t ic ipat ion : Pa s t Trend s a nd M a i n de synthèse de l’enquête nationale sur les Determinants in OECD Countries.” OECD valeurs]. Cinquantenaire de l’Indépendance du Economics Department Working Paper 376, Royaume du Maroc, Royaume du Maroc. Organisation for Economic Co-operation and Rauch, J., and S. Kostyshak. 2009. “The Development (OECD) Publishing, Paris. Three Arab Worlds.” Journal of Economic Kabbani, N., and N. Kamel. 2007. “Youth Perspectives 23 (3): 165–88. Exclusion in Syria: Social, Economic, and Rodary, M. 2007. “Female Labor in Pre-Colonial Institutional Dimensions.” Wolfensohn Center Morocco, between Oppression and Resistance: for Development at the Brookings Institution, Right to Work or Access to Benefits?” [Le Washington, DC. travail des femmes dans le Maroc précolonial, Kabbani, N., and E. Kothari. 2005. “Youth entre oppression et résistance. Droit au travail E mploy ment i n t he M E NA Reg ion: A ou accès aux bénéfices?]. Cahiers d’études Situational Assessment.” Social Protection africaines 47: 753–80. D i s c u s s i o n Pap e r 53 4 , Wo rl d B a n k , Ross, M. 2008. “Oil, Islam and Women.” Washington, DC. American Political Science Review 102 (1): Keddie, N. R. 2007. Women in the Middle East: 107–23. Past and Present. Princeton, NJ: Princeton Silver, H. 2007. “Social Exclusion: Comparative University Press. Analysis of Europe and Middle East Youth.” La Cava, G., T. Vishvanath, G. Morgandi, Working paper 1, Wolfensohn Center for U. S erajudd i n , a nd I . Ji l s on . 2 012 a . Development , T he M idd le E ast Yout h “Kingdom of Morocco: Promoting Youth Initiative, Dubai School of Government, Opportunities and Participation.” World Dubai, United Arab Emirates. Bank Knowledge Repository, Washington, U N Women. 2011. A nnu al Re por t . New DC. https://openknowledge.worldbank.org/ York: UN Entity for Gender Equality and handle/10986/11909. the Empowerment of Women. http://www ———. 2012b. “Reclaim ing T heir Voice: .unwomen.org/wp-content/uploads/2011/06/ New Perspectives from Young Women and UNwomen_AnnualReport_2010-2011_en.pdf. Men in Upper Egypt.” Views and News. Women, Business and the Law (database). 2012. http://documents.worldbank.org /curated/ I nternational Finance Corporation and en/2012/06/16575100/reclaiming-voice-new- World Bank, Washington, DC. http://wbl perspectives-young-women-men-upper-egypt. .worldbank.org. Moghadam, V. M. 1998. Women,Work, and World Bank. 2003. Gender and Development in Economic Reform in the Middle East and the Middle East and North Africa: Women North Africa. Boulder, CO: Lynne Rienner and the Public Sphere . Washington, DC: Publishers. World Bank. ———. 2004. “Women’s Economic Participation ———. 2004. The Road Not Traveled: Education in the Middle East: What Difference Has the Reform in the Middle East and North Africa. Neoliberal Policy Turn Made?” Journal of Washington, DC: World Bank. Middle East Women’s Studies 1 (1): 110–46. ———. 2006. World Development Report 2007: Offenhauer, Patricia. 2005. Women in Islamic Development and the Next Generation . Societies: A Selected Re view of Social Washington, DC: World Bank. Scientific Literature. Washington, DC: Library ———. 2007. “Youth–An Undervalued Asset: of Congress. Towards a New Agenda in the Middle East O’Higgins, S. N. 2005. “Trends in the Youth and North Africa.” World Bank, Washington, Labour Market in Developing and Transition DC. 86 JOBS FOR SHARED PROSPERITY ———. 2010. Arab Republic of Egypt Gender ———. 2012b. Reclaiming Their Voice: New Assessment 2010: Narrowing the Gap, Perspectives from Young Women and Men in Improving Labor Market Opportunities for Upper Egypt. Washington, DC: World Bank. Women in Egypt . Washington, DC: World ———. 2013. Opening Doors: Gender Equality Bank. in the Middle E ast and Nor th Afric a . ———. 2011. World Development Report Washington, DC: World Bank. 2012: Gender Equality and Development. Youssef, N. H. 1972. “Differential Labor Force Washington, DC: World Bank. Participation of Women in Latin American ———. 2012a. Kingdom of Morocco: Promoting and Middle Eastern Countries: The Influence Youth Opportunities and Participation . of Family Characteristics.” Social Forces Washington, DC: World Bank. 51 (2): 135–53. Where Are the Barriers? PART 2 From a complaint letter: Mr President, I went to great trouble and made great efforts in order to obtain a score of 98 (Humanities) and I was able, through God’s bounty, to pass all the tests for admission to the Police Academy. Is it then just, Mr President, that I should be denied admission to the police force for no better reason than that my father is a decent but poor man who works as a property guard? From the answer to the letter: With reference to your complaint presented to the Presidency of the Republic concerning your rejection by the accep- tance examination at the Police Academy. We have to inform you that the matter has been reviewed with the director of the Police Academy and it is evident to us that the complaint is unfounded. We wish you success. — Alaa Al Aswani, The Yacoubian Building Introduction T he second part of this report presents evidence of the distortions and the incentives that lead to the unfavorable labor market outcomes described ear- lier. To set the stage, chapter 3 reviews the evolution of the Middle East and North Africa’s economies, taking into account their development model, sector composition, economic growth, and demography in past decades. This long-term look provides a macroeconomic context for the more specific analysis of the labor supply, labor demand, and labor intermediation process. In particular, chapter 4 assesses the dynamics of the private sector by examining those aspects of the invest- ment climate that most stifle productivity and employment growth among fi rms. These barriers include the discretionary enforcement of regulations, barriers to competition, distortions to input prices (including capital), and the ensuing lack of innovation. Chapter 5 focuses on the current labor market regulations and social insurance design, which affect the formality and distribution of jobs, and looks at their associated benefits. It identifies the mechanisms that make some groups 87 88 JOBS FOR SHARED PROSPERITY “insiders” and discusses why these may no longer be sustainable. Finally, chapter 6 discusses the supply and intermediation of skills in the labor market. First, it shows how the incentives built into the educational system are not designed to equip students with the skills demanded by the economy; next, it analyzes the extent to which candidates are able to signal their skills and access job opportunities on a meritocratic basis. Finally, it discusses why jobs are seldom attained by the most qualified candidates. A Long-Term View of MENA’s Economies and Labor Markets 3 Main findings • In the past two decades, MENA’s economies progressively shifted their economic policies away from government-led strategies. However, growth has been volatile, driven largely by natural resources, foreign aid, or investment in low- productivity sectors, and it has not led to a structural transformation of the economy. • The generation of predominantly low-productivity jobs (such as those in agriculture and construction), and the low share of manufacturing and high-value-added services, reflect the incomplete structural transformation of the regional economies. • Economic growth will require additional investments (with a stronger role for private investment and foreign direct investment) and policies that support an enabling environment for the private sector. • While employment grew at a relatively fast pace compared to other regions, the demographic pressure outpaced job creation and diluted the effect of GDP growth • Yet, the youth bulge is a tremendous opportunity, as it will create a larger tax base and a smaller need for public spending on the dependent population, enabling countries to prepare for those decades when the dependency ratio will begin to rise again. Introduction economies to fully absorb the large num- bers of young labor market entrants (what U neven and uncertain growth pat- is known as the youth bulge). This chapter terns, together with the lack of sound aims to describe the economic and demo- policies to ensure macrostability, still graphic developments of the past 50 years in represent a barrier to effective job creation in the MENA region, which have contributed the Middle East and North Africa (MENA) to shaping the macroeconomic and labor by hurting long-term investment in the more market situation observed today. dynamic sectors of the economy. In addition, The chapter provides a long-term over- demography challenges the ability of many view of growth and economic trends in the 89 90 JOBS FOR SHARED PROSPERITY region, describes how shifts in economic pol- and features common to subgroups that icies over time have contributed to employ- are distinct from the economic, social, and ment quality and patterns, analyzes the role demographic constraints and challenges of demographic trends, and explores the faced by the region as a whole. macroeconomic constraints faced by the region, particularly the barriers to structural transformation and accelerated growth and A long-term view of growth and employment creation given MENA’s unique economic trends demographic challenges. MENA’s growth performance in the past While this report covers the region as a 50 years has been mixed and highly volatile whole, it is important to recognize MENA’s (see figures 3.1 and 3.2). After two decades extraordinary diversity and heterogeneity. of solid growth (1960s and 1970s), most Countries differ in their social and eco- countries in MENA suffered a sharp drop in nomic characteristics, such as their natural output and a period of economic stagnation resource endowments, and also in their insti- in the 1980s, led by the oil crisis. Countries tutional and other development. An effort to in the region started to recover in the mid- take this diversity into consideration is made 1990s, albeit at different paces. These periods throughout the report by referring to coun- coincide with three interrelated policy shifts: try groupings (see table 3.1). In particular, (1) a shift from public sector–dominated to subregional trends provide detailed context private sector–led economies; (2) a move and background, beyond the regional anal- from closed to globally integrated econo- ysis and evidence presented in the report. mies; and (3) a transition from oil-dominated These country typologies highlight issues to more diversified production structures. TABLE 3.1 Heterogeneity in economies in MENA Labor availability Labor abundanta Resource richb Algeria, Islamic Republic of Iran, Iraq, Syrian Arab Republic, Republic of Yemen Resource poor Djibouti, Arab Republic of Egypt, Jordan, Lebanon, Morocco, Tunisia, West Bank and Gaza Labor importingc and resource rich Bahrain, Kuwait, Libya, Oman, Qatar, Saudi Arabia, United Arab Emirates Population size Larged Algeria, Arab Republic of Egypt, Islamic Republic of Iran, Iraq, Morocco, Saudi Arabia Small Bahrain, Djibouti, Jordan, Kuwait, Lebanon, Libya, Oman, Qatar, Tunisia, United Arab Emirates, West Bank and Gaza, Republic of Yemen Income Low income Djibouti, Republic of Yemen Lower-middle income Arab Republic of Egypt, Iraq, Morocco, Syrian Arab Republic, Tunisia Upper-middle income Algeria, Islamic Republic of Iran High Kuwait, Oman, Saudi Arabia, United Arab Emirates Geography or colonial heritage Maghreb Algeria, Libya, Morocco, Tunisia Mashreq Arab Republic of Egypt, Iraq, Jordan, Lebanon, Syrian Arab Republic, West Bank and Gaza Members of the Gulf Cooperation Council Bahrain, Kuwait, Oman, Qatar, United Arab Emirates Note: MENA = Middle East and North Africa. a. Refers to net inflows of workers’ remittances. b. As defined by the World Bank (2004), “resource-rich countries” are those in which extractive industries account for, or are expected to soon account for, more than 50 percent of government revenue. c. Refers to net outflows of workers’ remittances. d. Population over 20 million. A LONG-TERM VIEW OF MENA’S ECONOMIES AND LABOR MARKETS 91 Each phase provides useful insights into FIGURE 3.1 Five-year moving average of real GDP growth rates in today’s economies and labor markets. GCC and non-GCC MENA countries, 1960–2010 25 Early post-independence period, 1960s–80s: High growth fueled 20 by oil revenues Smoothed growth rate, % The region’s growth model from the 1960s 15 to the mid-1980s was government led, fea- turing high public spending and protected 10 national markets. In the 1960s, MENA’s growth performance was the highest in the 5 world, averaging 7.0 percent (or 4.6 percent per worker), reflecting not only acceler- 0 1960 1970 1980 1990 2000 2010 ated accumulation of production factors but Year also increased productivity linked to invest- GCC countries Non-GCC MENA countries ment in physical capital, starting from a low Lower-middle-income Upper-middle-income base. These trends are reflected in the high countries countries rates of measured growth in physical capital accumulation and total factor productivity Source: World Development Indicators. Note: GCC = Gulf Cooperation Council; MENA = Middle East and North Africa; GDP = gross domestic (TFP), which captures the efficiency with product. which inputs (or factors) are used (see table 3.2). The 1970s continued to feature strong FIGURE 3.2 Growth volatility in MENA and other regions, growth, driven by high rates of capital accu- 1970–2008 mulation, which were in turn made possible by the availability of financial resources dur- 9 ing the oil boom. 8 The growth strategy during these decades 7 was mainly fi nanced by high oil revenues in 6 resource-rich countries—mainly the Gulf 5 Index Cooperation Council (GCC)—and by work- 4 ers’ remittances1 and public borrowing in the 3 resource-poor countries. Public capital invest- 2 ment in infrastructure, education, and health 1 largely drove growth.2 Highly procyclical fis- 0 cal policies reflected the windfall of oil rev- es Re ntr rich As tr r ic id e d Af and sia ca un oo M ribb an cif tri fri hA enues and intraregional financing flows, with so ies ia ies dl an So a Ea co ce-p co ce- Pa nA ric un a a rth st c ut No e Ea e C ri ur d co ra ur th me an large outlays going to universal food and fuel u so ha C GC A Re Sa tin subsidies. 3 Productivity growth eventually b- La st Su slowed down as the region remained depen- dent on oil windfalls and largely isolated GDP growth Standard deviation from the global economy. Source: World Development Indicators. Note: MENA = Middle East and North Africa. Standard deviation of per capita growth in gross domestic product (GDP) 1970–2008. Falling oil prices and structural changes: 1980s–90s growth from the earlier decades collapsed. The government-led development model of With eroding macroeconomic balances and the early years delivered dividends, but could growing debt burdens, and despite heavy not be sustained. As international oil prices external assistance (which permitted spend- plummeted in the 1980s, the foundations of ing for several more years), investments 92 JOBS FOR SHARED PROSPERITY TABLE 3.2 Decomposition of growth in MENA, 1960s–90s 2003 and to 5.8 percent between 2004 and Percentage points 2008. The individual country results show Gross domestic Physical Labor Total factor that most of the changes in GDP growth product capital force productivity were associated with changes in TFP (see box 3.1 for the case of Tunisia), reflecting a 1960s 7.0 7.3 0.8 3.4 1970s 6.1 9.6 2.8 0.6 more efficient use of production factors and 1980s 3.8 6.1 3.3 –0.6 increases in employment (proxied by the 1990s 3.7 2.3 3.3 0.8 labor force), while human and capital stocks exhibited much lower variation over this Source: Keller and Nabli 2002. Note: MENA = Middle East and North Africa. period (figure 3.3). While these initial reforms brought tan- gible benefits to the macroeconomic envi- declined dramatically, with the rate of ronment, they did not lead to a change growth of the physical capital stock almost in the structure of production (see figure halved4 compared to the previous decade. 3.4). The mining sector still represents Most countries experienced a dramatic roughly 50 percent of value added in the decline in capital accumulation and in TFP GCC countries (figure 3.5). Agriculture between the 1970s and the 1980s. 5 By the contracted as elsewhere in the world but late 1980s, it was clear that the past devel- did not give way to vibrant and innova- opment model had failed and that a para- tive manufacturing and services sectors, digm shift was needed. In the late 1980s, a particularly in the case of resource-rich handful of countries in the region—Jordan, countries, where the literature has often Morocco, and Tunisia—embarked on pro- referred to the “Dutch disease” (that is, the grams of macroeconomic stabilization and apparent relationship between an increase policy reforms, followed in the 1990s by the in exploitation of natural resources and a GCC countries. Reforms varied markedly in decline in the manufacturing sector) as a timing and intensity,6 but all included reduc- potential explanation. The services sector, ing subsidies and public spending to some whose share of GDP has increased only extent, liberalizing trade and encouraging modestly in resource-poor countries in the exports and investment, and strengthening past 30 years and has actually decreased in the institutional foundations of a market-led resource-rich countries. economy, including consolidation of the rule of law. The structural reforms initiated in the Contribution of growth to mid-1990s allowed the region to catch up employment generation on several fronts, even if they remain largely During the era of state-led industrializa- incomplete (as explained in the following tion in the 1960s and the oil boom in the section). Thanks to the increased global 1970s, unemployment in MENA countries integration of the region,7 export growth declined steadily (World Bank 2004). The (excluding minerals and fuels) was higher region had modest unemployment rates, than the average of developing countries, similar to those in the advanced industrial leading to an increase in world market share economies. Up to the early 1980s, some despite strong competition from Asia. 8 A labor-abundant countries even reported progressive shift away from procyclical fiscal shortages of workers, as rural areas were policy and the government-led growth model being depopulated by rural-urban and also helped reduce macroimbalances,9 par- international migration. The GCC coun- ticularly in the GCC (see annex 3A). Average tries became increasingly reliant on mil- real gross domestic product (GDP) in MENA lions of foreign workers, especially from rose from 3.6 percent per year between 1996 other MENA countries, to compensate for and 1999 to 4.6 percent between 2000 and acute national labor shortages. A LONG-TERM VIEW OF MENA’S ECONOMIES AND LABOR MARKETS 93 BOX 3.1 Productivity gains and contribution to growth from early reforms in Tunisia In Tunisia, trade integration has been a key driver bution to growth in gross domestic product jumped of technical progress and productivity growth to 43 percent in 2007. Tunisia’s TFP growth, how- since the mid-1990s (World Bank 2008). Through ever, remains below that of comparable (and suc- easier access to technology-intensive equipment cessful) countries, such as the Republic of Korea and machinery, greater penetration of new mar- (1.9 percent) and Malaysia (1.5 percent). A recent kets, and enhanced competitive pressure, trade World Bank publication (see World Bank 2010) integration has led to higher growth in total factor argues for a second generation of reforms and a productivity (TFP) and a higher contribution of the change in the growth model of the country, aimed latter to growth (figure B3.1.1).a Annual growth in at facilitating the transition to a more technology- TFP ranged between 1.24 percent in the 1990s and intensive economy to support productivity and thus 1.40 percent in 2000–06. As a result, TFP’s contri- competitiveness and growth. FIGURE B3.1.1 Contribution of total factor productivity to growth in Tunisia, 1961–2008 Growth decomposition, 1961–2008 25 Contribution US$ (billions, constant 2000) 20 of total factor productivity: 45% 15 Contribution 10 of capital and labor: 55% 5 0 61 65 69 73 77 81 85 89 93 97 01 05 08 19 19 19 19 19 19 19 19 19 19 20 20 20 Real GDP: average annual Real GDP if total factor growth in 1961–2008: 5.04% productivity was zero Source: World Bank 2010. Note: GDP = gross domestic product. a. In 1995, trade protection vis-à-vis the European Union (EU) was gradually unraveled as part of the Tunisia-EU Association Agreement, and by January 2008, the free trade area with the EU was completed for industrial goods. These reforms, supported by earlier investments in human and physical capital, have effectively led to a sharp increase in productivity (World Bank 2008). By the early 1990s, the collapse of oil prices First, employment was created in low-quality and the emergence of serious macroeconomic jobs,10 skewed toward the informal sector11 imbalances, together with continued rapid (particularly construction) or special employ- labor growth, propelled unemployment rates ment programs to support job creation to their highest in decades (figure 3.6). This (mainly in resource-rich countries). Second, well-known unemployment challenge remains the number of jobs created, while compara- entrenched in the region today. ble to other countries in the world, was not Two main factors appear to be behind enough to absorb the increasing number of these disappointing employment outcomes. new entrants into the labor market. 94 JOBS FOR SHARED PROSPERITY FIGURE 3.3 Decomposition of growth in GDP adjusted for human capital in selected countries in MENA, 1990s and 2000s a. Growth decomposition, 1990s b. Growth decomposition, 2000s 7 7 6 6 5 5 4 4 Percent 3 Percent 3 2 2 1 1 0 0 –1 –1 –2 –2 it a p. ria n bl b sia co ria it a p. an sia co ria bl b bi bi a pu ra wa wa pu ra Re Re oc oc rd rd ge ge ni ni ic ra ra Tu c Re n A Re n A Ku Ku i Tu Jo Jo or or b b iA iA Al Al ra ra M M ria ud ud t, A t, A Sy Sy Sa Sa yp yp Eg Eg Total factor productivity Labor Human capital Total factor productivity Labor Human capital Physical capital GDP growth, 1990s Physical capital GDP growth, 2000s Source: Based on World Development Indicators for data on GDP growth data, the International Labour Organization (ILO)-Key Indicators of the Labour Market (KILM) for data on labor, the Nehru and Dhareshwar database (1993) for data on physical capital; and Barro and Lee (2010), for data on education and human capital. Note: MENA = Middle East and North Africa; GDP = gross domestic product. FIGURE 3.4 Changes in the composition of GDP in selected world Effect of slow structural regions, 1980–83 and 2007–10 transformation 80 The typical MENA country differs in sig- nificant ways from other middle-income 60 countries in the composition of employment 40 generation.12 A first difference is the rela- tively large contribution of the agricultural 20 sector to employment growth (figure 3.7). A second significant difference is the relatively Percent 0 low contribution to job creation of manu- –20 facturing and private services, especially the trade, tourism, logistics, and communica- –40 tion sectors (figure 3.8), which were the main engines of growth and job creation during –60 the 2000s in rapidly growing middle-income –80 comparators such as Indonesia or Malaysia. A third difference is the significance of the cif d l A nd sia be d Sa rica d ca Pa a an ib an Su th A st an ra a hA fri sia ic an nt pe ar ica nA public sector (as shown in chapter 1). i ut As r a Ce uro e C er f No e E ra So st th Am ha dl E Ea International evidence indicates that id tin M b- La employment outcomes in the long term are Change in share of services driven mainly by structural changes, which Change in share of industry (including mining reallocate labor between sectors with dif- and manufacturing) ferent productivities. MENA countries are, Change in share of agriculture like other developing economies, character- Source: Diop, Marotta, and DeMelo 2012. ized by large productivity gaps between dif- Note: GDP = gross domestic product. ferent economic sectors (such as agriculture, A LONG-TERM VIEW OF MENA’S ECONOMIES AND LABOR MARKETS 95 government services, information and FIGURE 3.5 Changes in the composition of GDP, excluding communication technology, and manufac- mining, in GCC Countries, 1980–83 to 2007–10 turing). Therefore, moving labor between sectors is likely to be associated with sig- Oil exporters nificant aggregate productivity changes. A 100 decomposition based on the methodology 90 developed by McMillan and Rodrik (2011) breaks down growth in labor productivity 80 into two components: within-sector pro- 70 ductivity growth and productivity growth due to changes in relative labor shares for 60 Percent the Arab Republic of Egypt, Jordan, and 50 Morocco (see box 3.2). The three countries have reached different degrees of struc- 40 tural transformation. Jordan’s structural 30 transformation is the most advanced, as refl ected in its low share of labor in agri- 20 culture in 2008 (12 percent). Egypt is at an intermediate stage (with a 27 percent 10 share of labor in agriculture in 2007), 0 while Morocco is at an earlier stage (with r a s n t an am ria p. Re q ic en a . ep ta ai te bi y ai Ira bl Re ib Om e w Qa ira hr ,R ra pu g L a 40 percent share of labor in agriculture Ku ic iA Al Ba Em ud m ab ab in 2010). Isl Ye Sa Ar Ar n, Ira d The evidence suggests that the devel- n ite ria Un Sy opment process in Jordan resulted in an Services Construction Manufacturing increased share of labor in lower-productivity Mining and utilities Agriculture sectors (that is, retail trade or construction) that absorbed the large number of new labor Source: World Bank 2011. market entrants. In contrast, the reduction Note: GDP = gross domestic product; GCC = Gulf Cooperation Council. in the share of labor working in agricul- ture drove labor reallocations, on average, to higher-productivity sectors in Egypt and Morocco (see box 3.2). However, half of the 2000s, MENA’s employment it appears that, even when labor moved elasticity of 0.65 compared favorably with out of agriculture toward relatively more employment elasticity of nearly all regions productive sectors, in many cases it went and countries in the world (World Bank toward low-value-added sectors in absolute 2011), and its annual employment growth terms (construction and tourism) and not was the highest in the world between 1998 toward the higher end of the value-added and 2008 (figure 3.9). This elasticity was chain (fi nancial or telecommunication ser- especially high in the resource-rich coun- vices), which continue to generate too few tries (on average 0.91), despite a production jobs. structure skewed toward capital-intensive industries. By contrast, resource-poor coun- tries such as Egypt, Jordan, Lebanon, and Mismatch of the pace of job creation Morocco displayed a propensity to create and growth in the labor force employment that was substantially lower The high unemployment rates for the region than the MENA average. However, the level mask the fact that MENA countries created of employment creation in the whole region jobs at a higher pace than other parts of the did not keep pace with population growth world during the past decade. In the second (figure 3.10). 96 JOBS FOR SHARED PROSPERITY FIGURE 3.6 Unemployment rates in selected world regions, in the demographics of M ENA. As in 1980s–2000s most of the developing world, M ENA countries have experienced demographic Middle East and changes of unprecedented and extraor- North Africa d i n a r y prop or t ion s du r i ng t he pa s t GCC 50 years, with important implications for countries Non-GCC the economy and the labor market. In this countries section, the demographic changes in MENA are examined in an effort to pinpoint their Africa main drivers and their implications for the Latin America labor market and the economy at large in and the Caribbean the years to come. Sub-Saharan Africa East Asia and Pacific Fertility and mortality trends Europe and Central Asia Population trends offer a summary snapshot 0 2 4 6 8 10 12 14 16 of demographic changes. Population growth Percentage has been declining overall, but subregional Mean, 2000s Mean, 2000s Mean, 2000s patterns are highly diverse. Figure 3.11a Mean, 1990s Mean, 1990s Mean, 1990s shows long-term trends in population growth Mean, 1980s Mean, 1980s Mean, 1980s between 1950 and 2010. On average, popu- lation growth has decreased by half since Source: Based on data from ILO-KILM. the 1950s, from about 4 percent to about Note: GCC = Gulf Cooperation Council. 2 percent on an annual basis. However, trends within regions are very different. The FIGURE 3.7 Employment creation in MENA by sector, 1998–2008 low-income countries such as Djibouti and the Republic of Yemen and the GCC coun- Sector contribution to annual employment growth rates tries show remarkable swings during short Sub-Saharan Africa periods. For example, in the 1970s and dur- Latin America and the ing the most recent period, annual population Caribbean South Asia growth in GCC countries ranged between 2 Southeast Asia & Pacific and 8 percent (if migrants are included). These East Asia gross trends are the result of a complex web Central/SE Europe & CIS of factors, including fertility, mortality, and Developed migration. Dissecting population growth into World these components gives important insights into economic and labor market trends. Non-GCC Fertility rates started to fall as growth GCC accelerated during the past two decades Middle East and North Africa (figure 3.11b). This remarkable fall in fertil- –40 –20 0 20 40 60 80 100 ity rates coincides with the period when most Percent countries had started to grow at sustained lev- Agriculture Industry Services els, suggesting that the well-known relation- Source: Based on World Development Indicators. ship between fertility and growth observed in Note: SE = Southeast; CIS = Commonwealth of Independent States; GCC = Gulf Cooperation postwar Europe and the United States would Council; MENA = Middle East and North Africa. also hold for the MENA region. As countries become richer and education improves, fer- Demographic challenges tility declines. All MENA countries seem to Against the backdrop of these economic follow this pattern, with intraregional differ- changes, less visible changes have occurred ences explained by differences in well-being, A LONG-TERM VIEW OF MENA’S ECONOMIES AND LABOR MARKETS 97 as suggested by the later decline in fertility in FIGURE 3.8 Sectoral contribution to annual employment growth Djibouti, Iraq, and the Republic of Yemen. in a typical country in MENA and in Brazil, Indonesia, and Malaysia, Infant mortality (figure 3.11c) also declined average 2000s for all countries. From over 200 deaths per 1,000 children in the 1950s, rates are now 140 well below 50 per 1,000, converging across 120 groups of countries toward one-digit deaths 100 per 1,000. This trend is evident throughout 80 the post- independence period, suggesting Percent 60 that health and sanitation improved even 40 during periods of economic stagnation. The 20 combined effect of fertility and mortality 0 changes is that the natural rate of increase of –20 the population has slowed significantly since –40 the 1950s, although it remains positive in all Typical MENA Malaysia Indonesia Brazil country countries. However, to explain recent changes in Agriculture Manufacturing Trade, tourism, logistics, Government, administration, population growth, we must also look at and communication and social services net migration (figure 3.11d). Two different Mining and utilities Construction patterns emerge. The GCC, and to a minor Financial and real estate extent the low-income countries, shows services much sharper fluctuations in net migration rates. These countries have been largely Source: World Bank 2011. Note: MENA = Middle East and North Africa. receivers of immigrants, and immigration flows have played a very important role in population trends over relatively short spells of time. The Maghreb and Mashreq coun- tries show much less variation over time, important short- and long-term changes in and for long periods their net migration was population size, GDP and GDP per capita are negative. two very different measures. The most important migration flows Figure 3.12 illustrates this phenomenon: within the MENA region are explained the difference between GDP growth and partly by differentials in economic condi- GDP per capita growth is plotted for Egypt tions across countries (GCC and neighboring and Morocco. During the past decade, Egypt countries) and partly by the displacement of and Morocco had economic growth rates of people due to conflicts (Djibouti, Iraq, Leba- around 4–5 percent per year, but between non, and the West Bank and Gaza). Maghreb one-third and one-half of that growth was countries experienced periods of large emi- dampened by population increase. gration flows, mostly to southern European countries such as France, Italy, and Spain, especially in the 1970s and 1980s. Demography and the labor market From a labor market perspective, the bal- ance between the working-age population Demography and growth (WAP) and the non-working-age popula- These historical demographic changes have tion is crucial to an understanding of the had important implications for macroeco- changes in the main labor market indi- nomic indicators. In high-income countries, cators. It is important to discriminate the choice of whether to look at GDP growth between purely demographic factors and or GDP per capita growth has become almost economic factors. Consider, for example, irrelevant, but in countries that experience the three main labor market indicators: the 98 JOBS FOR SHARED PROSPERITY BOX 3.2 Structural change and employment outcomes in the Arab Republic of Egypt, Jordan, and Morocco Morocco also increased between 1999 and 2010, but they Figure B3.2.1 plots the average labor productivity started from low levels, and the total labor share (relative to the average across all sectors) and the in both sectors still accounted for only 6.7 percent change in the employment share for 12 sectors in in 2010. Morocco between 1999 and 2010. The size of the circle refl ects the employment share in 2010. The plot shows that structural change contributed to Egypt aggregate labor productivity growth in Morocco In Egypt, the reduction in the share of labor working in t h is period. Specifically, ag g regate labor in the large and low-productivity government ser- productivity growth was 3.5 percent, of which vice sector supported aggregate productivity growth within-sector growth accounted for 2.3 percent over the past decade. Overall, however, growth due and structural change for 1.2 percent. The struc- to reallocations of labor (structural change) should tural change component was positive because of have been even higher, given both Morocco’s and the declining labor share of the large, low-produc- Egypt’s stage of development. If the labor share of tivity agricultural sector. The abundant labor from higher-productivity sectors does not increase, both agriculture was absorbed mainly by construction, countries are likely to have negative or stagnant manufacturing, and wholesale and retail trade. The growth effects due to labor reallocations in the near labor shares of transport and telecommunication future, once agricultural employment shrinks to a as well as fi nance, real estate, and business services more sustainable level. FIGURE B3.2.1 Structural change in Morocco, 1999–2010 Financial, estate, and Labor share in 2010 Sectoral productivity / average productivity (log) business service Electricity and water Agriculture 40.3% Trade 12.8% Mining Education, health, and Manufacturing 11.4% social services Transportation 0 and telephone Construction 9.9% Government administration Manufacturing Other services 6.1% Hotels and Education, health and 5.0% restaurants Trade social services Construction Government administration 4.6% Agriculture Transportation and telephone 4.5% Hotels and restaurants 2.4% Other services 0 Financial, estate, and 2.2% Change in employment share 1999−2010, % business service Beta=9.8; t-test=1.49 Electricity and water 0.4% Mining 0.4% Source: World Bank 2012b. (continued next page) A LONG-TERM VIEW OF MENA’S ECONOMIES AND LABOR MARKETS 99 BOX 3.2 Structural change and employment outcomes in the Arab Republic of Egypt, Jordan, and Morocco (continued) FIGURE B3.2.2 Structural change in the Arab Republic of Egypt, 2000–07 Labor share in 2007 Suez Canal Sectoral productivity / total productivity (log) Agriculture 27.0% Petroleum products Government services 24.6% Education, health, and social services Trade and finance Industry 13.1% 0 Telecommunication Electricity Trade and finance 10.0% Real estate Industry Education, health, and 8.2% social services Agriculture Construction 7.9% Transportation Hotels and restaurants Transportation 4.0% Hotels and restaurants 1.7% Construction Real estate 1.4% Government services Telecommunication 0.8% 0 Change in employment share 2000−07, % Electricity 0.8% Beta=1.05; t-test=1.99 Petroleum products 0.6% Suez Canal 0.1% Source: World Bank 2012b. FIGURE B3.2.3 Structural change in Jordan, 2000–08 Labor share in 2008 Telecommunication Government administration 26.7% Sectoral productivity / total productivity (log) Manufacturing 13.9% Estate and business service Retail 11.9% Mining Electricity Agriculture 11.8% Finance Construction 7.3% Manufacturing Transportation 6.8% 0 Transportation Wholesale goods Social 4.8% Estate and business service 3.4% Hotels and restaurants Education Construction Education 3.1% Health Wholesale goods 2.6% Government Hotels and restaurants 2.3% administration Retail Finance 1.7% Agriculture Water Social Health 1.7% 0 Change in employment share 2000−08, % Electricity 0.6% Beta=1.73; t-test=0.06 Telecommunication 0.5% Mining 0.5% Water 0.4% Source: World Bank 2012b. (continued next page) 100 JOBS FOR SHARED PROSPERITY BOX 3.2 Structural change and employment outcomes in the Arab Republic of Egypt, Jordan, and Morocco (continued) Jordan construction, and retail trade. In contrast, apart Jordan initiated major economic policy reforms in from wholesale trade and real estate and busi- 2000–01 that led to privatizations, inflows of for- ness services, sectors with the highest productiv- eign direct investment, increased trade openness, ity (transport, telecommunication, electricity, and and higher rates of private investment. Aggregate mining) saw their shares reduced. Government growth in labor productivity in Jordan was 3 percent administration, which is a large, low-productivity from 2000 to 2008, of which within-sector labor sector, also reduced its employment share, mitigat- productivity growth accounted for 3.5 percent and ing the negative aggregate productivity impact. structural change for –0.5 percent. Figure B3.2.3 These fi ndings suggest that the reforms led to an illustrates that the employment shares of the high- increase in fi rm productivity within modern sec- est-productivity sectors declined somewhat, while tors but that it came at least partly at the expense the employment shares increased in several low- of lower relative employment shares in these productivity sectors: social and personal services, sectors. FIGURE 3.9 Total annual employment growth in selected regions unemployment rate (UR), the employment of the world, 1998–2008 rate (ER), and the labor force participation rate (LFPR): Middle East and U E North Africa UR = ; ER = ; GCC countries E+U E+U+I E+U Non-GCC , countries LFPR = E+U+I where E = employed; U = unemployed; and Developed countries I = inactive; working age population (WAP) = East Asia E + U + I; labor force (LF) = E + U. Changes in the WAP arise from changes Europe and Central Asia in demographics, whereas changes in the World rates of E, U, or I could be induced by a demographic factor, an economic factor, Southeast Asia and Pacific or both. For example, if the employment Latin America and the rate decreases, that decrease could be due Caribbean to an increase in the WAP with no changes South Asia in employment (the new WAP entrants fall Sub-Saharan Africa into the pools of the unemployed or the inactive) or to a decrease in employment. 0 1 2 3 4 5 In the first case, the economy has not Percent been able to create new jobs for the new Source: ILO-KILM. entrants, whereas in the second case, the Note: GCC = Gulf Cooperation Council. economy has destroyed jobs. A LONG-TERM VIEW OF MENA’S ECONOMIES AND LABOR MARKETS 101 This phenomenon is illustrated further stabilized at around 10 percent in 1995. But by looking at the Egyptian labor market after normalizing the labor market stocks between 1970 and 2010. Labor market by the population (in other words, steril- stocks, labor market indicators, and labor izing the demographic effect), we can see market population shares are shown in figure 3.13. Labor market stocks clearly FIGURE 3.10 Growth in employed and working-age population show positive and steep trends over the in non-GCC countries in MENA, 1998–2009 period (figure 3.13a). The pools of inac- tive and unemployed steadily grew, as did 55% not employment. During this period, the WAP 250 working 56% not rapidly increased (explained by the entry 200 working of the baby-boom generation into the labor market), and increases in the inactive and 150 Millions unemployed populations were greater than increases in the employed population. Over- 100 all, Egypt has created an increasing num- 50 ber of jobs every year for the past 40 years, clearly not an economy that destroys jobs. 0 98 99 00 01 02 03 04 05 06 07 08 09 The standard labor market indicators 19 19 20 20 20 20 20 20 20 20 20 20 allow observation of the labor market from Year a different angle (fi gure 3.13b). The labor Working-age employed Working-age population force participation rate and the employment rate in Egypt both peaked around 1990 and then declined slowly; the unemploy- Source: Based on the ILO’s EAPEP (Economically Active Population, Estimates and Projections) database. ment rate rose throughout 1970 –95 and Note: GCC = Gulf Cooperation Council; MENA = Middle East and North Africa. FIGURE 3.11 Demographic trends in MENA, 1950–2010 a. Population growth b. Fertility 9 9 8 8 7 7 Growth rate, % Growth rate, % 6 6 5 5 4 4 3 3 2 2 1 1 0 0 1950–55 1960–65 1970–75 1980–85 1990–95 2000–05 2005–10 1950–55 1960–65 1970–75 1980–85 1990–95 2000–05 2005–10 c. Infant mortality d. Net migration 250 70 60 200 50 Growth rate, % Growth rate, % 150 40 30 100 20 10 50 0 0 –10 1950–55 1960–65 1970–75 1980–85 1990–95 2000–05 2005–10 1950–55 1960–65 1970–75 1980–85 1990–95 2000–05 2005–10 Maghreb Mashreq GCC LIC Source: UN Population Information Network 2007. Note: MENA = Middle East and North Africa; GCC = Gulf Cooperation Council; LIC = low-income country. 102 JOBS FOR SHARED PROSPERITY FIGURE 3.12 Difference between GDP growth and GDP per capita growth, Arab Republic of Egypt and Morocco, 1960–2010 percent a. Arab Republic of Egypt b. Morocco 2.8 3.0 GDP gap – per capita GDP growth 2.6 GDP gap (%)—growth rate 2.5 2.4 2.0 2.2 2.0 1.5 1.8 1.0 60 70 80 90 00 10 60 70 80 90 00 10 19 19 19 19 20 20 19 19 19 19 20 20 Year Year Source: Based on World Development Indicators. Note: GDP = gross domestic product. FIGUR E 3.13 Labor market developments in the Arab Republic of Egypt, 1970–2010 a. Labor market stocks b. Labor market indicators c. Labor market population 30,000 60 100 shares 90 25,000 50 80 % per population 20,000 40 70 Thousands 60 15,000 30 50 40 10,000 20 30 5,000 10 20 10 0 0 0 19 0 75 19 0 19 5 90 20 5 20 0 05 10 19 0 19 5 19 0 85 19 0 20 5 00 20 5 10 19 0 19 5 19 0 85 19 0 20 5 00 20 5 10 7 8 8 9 0 7 7 8 9 9 0 7 7 8 9 9 0 19 19 19 20 19 19 20 19 19 20 Employment Employment rate Employment Unemployment Unemployment rate Unemployment Inactive Labor force participation Inactive Source: ILO-Laborsta. that the inactive population slowly declined of job creation, once the baby-boom effect between 1970 and 2010, thanks to mod- is taken into account. A demographic tran- erate increases in both employment and sition similar to the one observed in Egypt unemployment (figure 3.13c). Therefore, requires policies tailored for the new and Egypt has not performed so poorly in terms young entrants. A LONG-TERM VIEW OF MENA’S ECONOMIES AND LABOR MARKETS 103 Population changes: Opportunities the cost of society. A window of opportunity and challenges appears for countries to introduce contribu- tory pension systems, which we will discuss An additional impact of the demographic in chapter 5. transition in MENA is that the median Tunisia is an example of this phenome- age is expected to increase (figure 3.14). non. Its baby-boom generation, born 15–20 Between the 1960s and the 1980s, high fer- years ago, is entering the labor market, and tility rates, combined with improvements the largest concentration of its population in infant mortality, created a baby boom is 15–29 years old. This generation will be that reduced the median age, but this trend of pensionable age around 2050, when the has inverted since the 1980s because of population just entering the labor market the subsequent drastic reduction in fertil- will have stabilized at much lower levels. ity. The implication is that the dependency That is, the population pyramid typical of ratio (the ratio of the population ages 0–19 poor countries by then will have converted and ages 65+ relative to the population to the inverse pyramid typical of richer aged 20–64) will continue to decline until countries. around 2040 and then may stabilize or rise In essence, the MENA region is caught in further. an economic and demographic conundrum. The U-shaped median age curve has well- On the one hand, the youth bulge now enter- known implications for the economy. In ing the working-age population places a the fi rst phase, countries experience a baby heavy burden on job creation. On the other boom, in which both the number of children hand, the youth bulge is a tremendous eco- and the number of pensioners increase rela- nomic opportunity for countries, as it cre- tive to the labor force. In the second phase, ates a larger tax base and a smaller need for as the baby-boom generation enters the labor public spending on the dependent population market, the larger labor force is in a better (children and pensioners). position than the previous generation to bear FIGURE 3.14 Median age and dependency ratio in selected subregions in MENA, 1960–2050 a. Median age b. Dependency ratio 50 120 45 100 40 35 Dependency ratio 80 Years of age 30 25 60 20 40 15 10 20 5 0 0 60 70 80 90 00 10 20 30 40 50 60 70 80 90 00 10 20 30 40 50 19 19 19 19 20 20 20 20 20 20 19 19 19 19 20 20 20 20 20 20 Maghreb Mashreq GCC Low-income countries Source: UN Population Information Network 2007. Note: MENA = Middle East and North Africa; GCC = Gulf Cooperation Council. 104 JOBS FOR SHARED PROSPERITY An unfinished reform agenda for greater participation in the economy and for an inclusive and shared growth process.13 The preceding sections described what can be defined as an unfinished reform agenda. Because of this, many of the constraints to sus- Characteristics of successful growth tained growth typical of the old development lacking in MENA model are still present today, to some extent. This section will present a brief overview of Freund and Rijkers (2012) show that “employ- the main barriers that need to be addressed in ment miracles”14 are much less frequent in order to reach sustained growth, a necessary MENA than in other regions. They find that precondition for continued employment cre- employment miracles tend to coincide with an ation as well as for an accelerated pace. They acceleration of growth, an overall improve- require a host of policy measures that include ment in macroeconomic conditions (mani- removing obstacles to structural transforma- fested, for example, in higher trade flows, high tion, diversifying away from natural resources, investment, and lower government spend- deepening trade integration, and strengthen- ing), and improvements in the regulatory ing macroeconomic management. They also framework (see box 3.3). Overall, MENA relate to other factors, such as governance and seems to lack these macroeconomic condi- accountability, which create the preconditions tions and most of the common characteristics BOX 3.3 Employment miracles How can policymakers engineer enduring reduc- FIGURE B3.3.1 Incidence of “employment tions in unemployment? To address this question, miracles” by region, 1980–2008 Freund and Rijkers (2012) document the incidence and determinants of “employment miracles,” defined Middle East and as substantial and sustained reductions in unemploy- North Africa ment that have occurred throughout the world over Developed the past three decades. More specifically, an employ- countries ment miracle is a period of unemployment decline Latin America and of (1) at least three percentage points and (2) at least the Caribbean one-quarter of its initial level over a four-year period East Asia and Pacific that (3) persists for at least another three years. In Europe and addition, (4) employment miracles start with a strict Central Asia decline in unemployment, and (5) contiguity is ruled 0 2 4 6 8 out to avoid double-counting the same employment Percent reduction spell as two separate episodes. The frequency with which such miracles occur Source: Freund and Rijkers 2012. is encouraging; each year approximately 1 in every 20 countries embarks on one. Unfortunately, they are less prevalent in MENA (see figure B3.3.1). More- example, in higher trade flows, high investment, and over, the associated decline in unemployment is large: lower government spending, as well as improvements average unemployment, seven years after the onset of in the regulatory framework (see figure B3.3.2). the decline, stood at less than half its initial level. Except for suffering higher initial unemployment, Employment miracles tend to coincide with an countries that embark on miracles do not differ, acceleration of growth and an overall improve- on average, from ones that do not in their macro- ment in macroeconomic conditions manifested, for economic preconditions, their governance, or their (continued next page) A LONG-TERM VIEW OF MENA’S ECONOMIES AND LABOR MARKETS 105 BOX 3.3 Employment miracles (continued) FIGURE B3.3.2 Predictors of the incidence of “employment miracles” Predicted increase in 1 unit change of the explanatory Government effectiveness Control of corruption variable at the sample mean Rule of law Political stability Starting a business Time to export Time to import Time to enforce a contract –10 –8 –6 –4 –2 0 2 4 6 8 Percent Source: Freund and Rijkers 2012. Note: NB Times are in logs. Red, orange, and blue denote significance at the 1%, 5%, 10% level, respectively. a. The Index of Economic Freedom aims to measure 10 components of the right of every human to control his or her own labor and property. Such index assigns a grade in each component, ranging from 0 to 100, where 100 represents the maximum freedom. The 10 economic freedoms are grouped into four broad categories or pillars of economic freedom: (1) rule of law (property rights, freedom from corruption); (2) limited government (fiscal freedom, government spending); (3) regulatory efficiency (business freedom, labor freedom, monetary freedom); and (4) open markets (trade freedom, investment freedom, financial freedom). A country’s economic freedom score is a simple average of its scores on the 10 individual freedoms (see http://www.heritage.org/index/). regulatory frameworks at the time of onset. Busi- A one-percentage-point improvement in economic ness regulation and good governance are especially regulation (as proxied by the Economic Freedom strongly correlated with the incidence of employ- Index)a is associated with an increase in the prob- ment miracles, which perhaps should not come as ability of the onset of an employment miracle of a surprise, since sound regulation and good gov- approximately 3.6–5.0 percent, all things being ernance often go hand in hand (see figure B3.3.2). equal . associated with successful growth and com- • High rates of savings and investment. petitiveness. The Commission on Growth • Full exploitation of the world economy. and Development (2008) has identified five of Knowledge acquired in the global econ- these characteristics: omy and exploitation of global demand are the foundation of economic “catch • Committed, credible, capable govern- up” and sustained growth, and the pro- ment. It is important that government has motion of foreign direct investment and the capacity to devise, and the institutions foreign higher education can support to implement, a growth strategy. knowledge transfer. • Macroeconomic stability. Modest infl a- • Letting markets allocate resources. Poli- tion and sustainable public fi nances are cies need to ensure that product and labor necessary to spur growth. 106 JOBS FOR SHARED PROSPERITY markets are flexible enough to allow a and technical progress, and increased flows structural transformation from agricul- of ideas, knowledge, and innovation from ture to manufacturing and to ensure that abroad. Increased trade and investment flows there is, at a minimum, no bias against not only promote faster economic growth exports. and job creation but also help countries move up the global value chain, diversify, and Each of these characteristics is discussed improve their resilience to external shocks.19 next in the context of MENA.15 Despite recent liberalization, MENA’s trade regimes remain more restrictive than Lack of macroeconomic stability those of comparator countries. MENA coun- tries have streamlined and lowered tariffs The fundamentals of sound fiscal and mon- over the past two decades, often in the con- etary policies—prerequisites for sustained text of trade agreements with the European economic performance—have been largely Union and the United States. However, tariffs missing in the MENA region over the past remain high (averaging 12 percent), and sev- 30 years. Their absence is reflected in the eral countries rank at the higher end among region’s high output volatility, the highest 139 countries surveyed on a measure of over- in the world (see table 3A.2 in annex 3A). all trade restrictiveness (see figure 3.15 and Volatility of the real effective exchange rate table 3A.3 in annex 3A).20 As a consequence, (REER)16 hampered the development of non- MENA trade remains below the potential of resource-based activities (Gelb 1988), result- countries at the same level of development. ing in a so-called volatility-induced inefficient Cross-sectional results (Behar and Freund specialization pattern and the low diversifi- 2011) show that despite some evidence of cation outcome presented earlier.17 A volatile convergence, MENA’s exports to the outside REER increases uncertainty and reduces the world were only one-third of their potential incentive to invest in nonresource tradables. even in recent years, after controlling for the It dims the prospects for development in more standard determinants of trade. At histori- dynamic sectors of the economy that are not cal growth rates, it would take 20 years for capital intensive and have higher potential to MENA countries to reach their full potential. create employment (see Hausmann, Panizza, If natural resources are excluded, exports are and Rigobon 2004). Procyclical fiscal policy has also fueled macroeconomic instability, accompanied by FIGURE 3.15 Trade restrictions in selected the lack of high and sustained investment. countries in MENA, 2009 In many MENA countries, public finances are tied up in large subsidies and short-term 150 consumption expenditures.18 Food and fuel Algeria Iran, Islamic Rep. Overall trade barrier ranking subsidies alone amount to about 6 percent Egypt, Arab Rep. Lebanon of GDP on average for the region as a whole 100 MENA oil importers average (World Bank forthcoming), with wide varia- Yemen, Rep. tion by country. Chapter 4 will discuss how Average Jordan energy subsidies distort relative prices and 50 Morocco Kuwait incentives, in addition to their fiscal costs. Oman Saudi Arabia Bahrain United Arab Emirates 0 Incomplete global integration 0 5 10 15 20 25 Trade and investment raise productiv- Trade-weighted average tariff (%, 2009) ity through more efficient allocation of Source: Based on World Economic Forum 2012. resources, economies of scale, increased com- Note: MENA = Middle East and North Africa. Overall trade barriers are petition, faster rates of capital accumulation ranked from 1 to 139, with 1 being the least restrictive. A LONG-TERM VIEW OF MENA’S ECONOMIES AND LABOR MARKETS 107 also at only one-third of the benchmark, but growth are moderate, mainly because of a the improved export performance over time lack of technological progress and productiv- is much slower, implying that it could take ity growth. In an analysis of the correlates of twice as long to reach potential. Further- TFP (see box 3A.1), it appears that the effect more, the level and composition of foreign of globalization on TFP depends positively on direct investment (FDI) make only limited the quality of human capital. Thus, despite contributions to the region’s development the progress it has made, MENA still needs and employment generation, having been to deepen its global integration and invest biased toward either capital-intensive or in human capital and knowledge-driven low-skilled, labor-intensive sectors (such as absorptive capacity. By itself, global integra- oil, construction, and tourism) rather than tion might not bring the expected benefits toward more dynamic, employment-creating in increased growth and productivity if not sectors (figure 3.16 and chapter 4). accompanied by reforms in other important Moreover, the quality of human capital in areas (see chapters 4 and 5). MENA is not strong enough to fuel innova- tion and increase productivity. Assuming con- Unfulfilled potential of the private stant TFP growth and constant gross capital sector to spur economic growth formation in the next decade (and projecting increases in the labor force in line with demo- While the private sector has a larger role in graphic assumptions), the growth potential MENA today than before, it remains a far- of most MENA countries appears to be con- from-strong engine of growth (World Bank siderably constrained. Overall prospects for 2009). Governments have, by and large, FIGURE 3.16 Structure of foreign direct investment in selected countries in MENA, in China, and in the European Union, 2000–07 90 80 70 60 % of total FDI 50 40 30 20 10 0 ria p. co a an on ic a on ey isi in bl Re oc rk rd ge n ni Ch n pu ba Tu nU Tu Jo or b Al ra Re Le M ea t, A b p ra yp ro nA Eu Eg ria Sy Country and region Manufacturing Telecoms Finance Tourism and construction Energy High-tech services Source: United Nations Conference on Trade and Development (UNCTAD) database; World Development Indicators 2000–07. Note: MENA = Middle East and North Africa; FDI = foreign direct investment. 108 JOBS FOR SHARED PROSPERITY failed to establish rule-based modes of inter- on growth and job creation (see chapter action with the private sector. Except for 4). Freund and Bolaky (2008) find that all some of the resource-poor countries (oil 12 MENA countries included in their sample importers in the Maghreb), where the ratio of 126 countries are above the median in reg- of private investment has actually increased ulation of their economies. In the same vein, due to a substantial increase in FDI, the drawing on Doing Business data for a sample composition of investment in most of the of 133 countries, Amin and Djankov (2009) resource-rich countries favors public invest- fi nd that the proclivity to undertake micro- ment. MENA lags behind all other regions, reforms that reduce unjustified regulatory with the exception of Sub-Saharan Africa, in restrictions is much lower in countries whose the share of private sector gross fixed-capital exports are concentrated in abundant natu- formation in GDP (figure 3.17). This is only ral resources. The lack of a level playing field one part of the story, however. The compo- that would allow the private sector to play a sition and efficiency of investment also con- pivotal role in achieving structural change is tribute to the private sector’s limited impact explored in more detail in chapter 4. FIGURE 3.17 Private sector gross fixed-capital formation as a percentage of GDP in selected countries and regions, average 2004–09 25 20 15 Percent 10 5 0 Th sia Sy gyp ala a n A Ara sia Sa epu . n c nd e ca Af . A a As arib ia an ean Ch c a an am ey Dj p. d Tun i nt sia ut d Le sia n a co p rth ep t i i M by ric in di no So an ou bl cif r Su rab Re Re t a Yem Afri oc Isl Turk rd Ea the lge lA y i hA No n, R In Li b ail Pa ba b Jo or b ic ra i M d R ra C Ce ha ria t, ia n, an Ira b- d E st an pe as ica ro eE Eu er dl Am id M tin La Country or region MENA countries Comparator countries or regions Source: World Development Indicators. Note: GDP = gross domestic product; MENA = Middle East and North Africa. A LONG-TERM VIEW OF MENA’S ECONOMIES AND LABOR MARKETS 109 Annex 3A Macroeconomics TABLE 3A.1 Decomposition of average GDP growth rates in selected countries in MENA, 1960–2000, by decade Growth of Physical Human TFP GDP per capital per capital per growth Decade laborer (%) laborer (%) laborer (%) (%) Kuwait 1980s –5.80 –1.20 0.40 –5.00 1990s –0.90 –0.10 0.10 –0.80 2000s 3.60 0.50 0.10 3.00 Saudi Arabia 1960s 5.30 2.70 0.00 4.20 1970s 6.40 17.00 0.00 –0.40 1980s –9.10 –2.60 0.50 –7.00 1990s –0.10 –0.60 0.40 0.10 2000s –0.20 –0.90 0.50 0.30 United Arab 1980s –7.50 — 0.50 — Emirates 1990s –1.30 –2.00 0.80 0.00 2000s 0.50 –1.10 0.70 0.90 Egypt, Arab 1960s 3.00 3.30 0.30 1.50 Rep. 1970s 4.40 5.70 1.00 1.50 1980s 2.90 2.00 0.60 0.40 1990s 1.70 0.00 0.50 1.20 2000s 2.10 0.30 0.40 1.40 Jordan 1960s –1.30 –6.50 11.40 –6.70 1970s 6.40 10.30 8.10 –2.60 1980s –2.40 0.70 0.60 –3.80 1990s –2.20 –1.00 0.50 –1.70 2000s 2.90 0.10 0.40 2.40 Syrian Arab 1960s 1.00 1.30 0.80 0.00 Republic 1970s 6.30 5.50 1.50 3.20 1980s –2.70 –2.90 0.40 –0.10 1990s 1.10 –3.00 0.00 4.00 2000s 0.50 –2.70 0.20 3.00 Yemen, Rep. 1980s — — 0.30 — 1990s 0.60 –0.30 0.40 0.50 2000s –0.20 –0.10 0.50 –0.60 Tunisia 1960s 3.40 4.60 0.90 1.00 1970s 2.80 2.40 1.50 0.90 1980s 0.40 0.70 0.40 –0.70 1990s 2.00 0.30 0.50 1.20 2000s 2.50 0.60 0.50 1.50 Morocco 1960s 2.20 1.80 –3.20 3.50 1970s 1.00 4.30 0.80 –1.20 1980s 0.80 0.40 0.40 0.00 1990s –0.30 0.20 0.30 –0.90 2000s 3.00 0.80 0.30 1.80 Algeria 1960s 3.00 1.40 0.60 2.10 1970s 2.70 5.30 1.10 –0.10 1980s –1.30 0.30 0.70 –2.30 1990s –3.10 –1.10 0.60 –2.50 2000s 0.20 –0.30 0.40 0.20 Source: Data from the 1960s to 1980s are taken from Keller and Nabli (2002). The authors followed the same structure and data set to update the calculation to recent years. Note: — = not available; GDP = gross domestic product; MENA = Middle East and North Africa; TFP = total factor productivity. 110 JOBS FOR SHARED PROSPERITY TABLE 3A.2 Economic growth and its volatility, 1982–2010 Coefficient Mean Mean of change in REER growtha variationb growthc volatilityd (1) (2) (3) (4) GNI per capita 1982 Low (20) 2.03 2.99 4.55 49.46 Lower middle (52) 0.96 5.00 3.31 35.34 Upper middle (32) 1.90 2.36 3.63 21.44 High (33) 1.86 1.52 1.93 8.71 GNI per capita 2010 Low (24) 0.76 8.11 4.18 45.15 Lower middle (38) 1.72 3.62 3.72 33.96 Upper middle (32) 1.96 2.81 3.99 22.15 High (43) 2.24 1.87 2.44 11.12 MENA (17) 1.40 4.76 4.78 52.20 GCC (6) 1.37 5.58 5.29 110.41 Non-GCC (5) 2.28 2.81 4.50 23.66 Source: Diop, Marotta, and De Melo 2012. Note: Number of countries in parenthesis. GCC = Gulf Cooperation Council; GNI = gross national income; REER = real effective exchange rate; MENA = Middle East and North Africa. a. Mean growth is the average growth rate over the period 1982–2010, that is, approximately 38 observations per country resulting in sufficiently large samples to give significantly different mean growth rates in each sample. b. Standard deviation divided by the mean. c. Mean change is the absolute mean change in growth rates observed between two years. d. The standard deviation of the monthly real effective exchange rate is computed over the period 1980–2010. The sample is not exactly the same as those for columns 1 to 3. A LONG-TERM VIEW OF MENA’S ECONOMIES AND LABOR MARKETS 111 TABLE 3A.3 Policy indicators affecting trade in selected countries in MENA Tr. across TTRI Value OTRI Value LPI value IE (DB) borders Rule of law TAR-AGR TAR-MAN (rank: 125) (rank: 102) (rank: 164) (rank: 183) (rank: 183) (rank: 213) (1) (2) (3) (4) (5) (6) (7) (8) 1990–95 1990–95 Year 2006–09 2006–09 2006–09 2008 2006–09 2006–09 2010 2009 RPLA — — 58.4 — 79.6 102 60.6 102.6 Egypt, Arab Rep. 35.8/54.6 23.6/9.2 3.3 (68) 10.0 (59) 94 106 21 97 Jordan a–/16.7 a–/10.0 4.6 (108) 11.3 (66) 80 100 77 81 Lebanon –/11.4 –/5.1 1.9 (50) — 33 107 95 145 Morocco 66.5/26.7 63.9/10.8 1.8 (48) 14.1 (75) 131 128 80 106 Tunisia 29.6/38.6 28.0/21.0 0.9 (18) 11.7 (69) 60 69 30 84 RRLA — — — — 119.2 133.6 135.4 169.2 Algeria 25.4/21.5 21.3/15.9 0.7 (9) 1.5 (4) 135 136 124 156 Iran, Islamic Rep. –/28.5 –/24.6 1.9 (49) 2.7 (14) 104 137 131 171 Iraq –/– –/– — — 156 153 179 210 Libya –/0 –/0 — — 137 — — 161 Syrian Arab Republic –/15.8 –/12.8 — — 80 143 120 132 Yemen, Rep. –/7.0 –/5.3 — — 103 99 123 185 GCC — — 49.8 — 41 38.5 50 77 United Arab Emirates –/4.7 –/4.2 3.6 (71) 3.5 (20) 24 33 3 76 Bahrain –/7.7 –/3.9 2.6 (58) 3.3 (19) 32 20 33 77 Kuwait –/3.1 –/4.2 — — 36 61 113 73 Oman 8.2/4.9 5.1/3.7 1.4 (31) 3.1 (16) 60 65 87 66 Qatar –/5.9 –/4.1 1.8 (47) — 55 39 46 81 Saudi Arabia 11.8/3.0 12.4/4.1 1.7 (42) 2.6 (12) 39 13 18 89 Sources: Columns 1–2, World Integrated Trade Solution (WITS), columns 3–5, World Trade Indicators; column 6, Doing Business; columns 7 and 8, World Governance Indicators. Note: For all ranks, a higher value means a worse rank. — = not available. MENA = Middle East and North Africa; GCC = Gulf Cooperation Council; TAR-AGR = applied tariffs in agriculture; TAR-MAN = applied tariffs in manufacturing; RRLA = resource rich, labor abundant; RPLA = resource poor, labor abundant; OTRI= Overall Trade Restrictiveness Index; TTRI= Tariff-Only Trade Restrictiveness Index; LPI = Logistics Performance Index; DB = Doing Business. 112 JOBS FOR SHARED PROSPERITY FIGURE 3A.1 Intensive and extensive margins of exports, excluding oil products, in various world regions, 1997–2007 a. Exports to/from b. Type of exports to/from resource-poor countries and regions resource-poor countries and regions 16 25 14 20 12 10 15 Millions Millions 8 6 10 4 5 2 0 0 ia d n a d ter ea d pe As e es LA LA CD y d y d ne nd ds s C ite nio ric ro al op an Eas ribb a an th Am As od ar ar se se at GC a ia Ce n n RP RR oo po s a OE Ca nts Af im im Eu ntr Eur es es St go Un n U lg a ic m rt oc oc pr pr e C er er co Pa ta pr pr d ial um pi o d ial str Fo o ns tin str Fo du Co La du In In c. Exports to/from d. Type of exports to/from resource-rich countries and regions resource-rich countries and regions 3.0 14 2.5 12 2.0 10 8 Millions Millions 1.5 6 1.0 4 0.5 2 0 0 –0.5 –2 ia d n ca d ter ea d pe As e es LA LA CD y d y d ne nd ds s C ite nio ro al op an Eas ribb a an th Am As od ar ar se se at GC ri a ia Ce n n RP RR oo po s a OE Ca nts Af im im Eu ntr Eur es es St go Un n U lg a ic m rt oc oc pr pr e C er er co Pa ta pr pr d ial um pi o d ial str Fo o ns tin str Fo du Co La du In In e. Exports to/from f. Type of exports to/from GCC countries GCC countries 20 35 30 15 25 10 Millions Millions 20 5 15 10 0 5 –5 0 ia d n ca d ter ea d pe As e es LA LA CD y d y d ne nd ds s C ite nio ro al op an Eas ribb a an th Am As od ar ar se se GC at ri a ia Ce n n RP RR oo po s a OE Ca nts Af im im Eu ntr Eur es es St go Un n U lg a ic m rt oc oc pr pr e C er er co Pa ta pr pr d ial um pi o d ial str Fo o ns tin str Fo du Co La du In In Exports of new products to exisiting markets Exports of new products to new markets Exports of existing products to new markets Intensive margins Source: Based on Comtrade. Note: OECD = Organisation for Economic Co-operation and Development; GCC = Gulf Cooperation Council; RRLA = resource rich, labor abundant; RPLA = resource poor, labor abundant. A LONG-TERM VIEW OF MENA’S ECONOMIES AND LABOR MARKETS 113 FIGURE 3A.2 Main trends in selected macroeconomic indicators for MENA since the 1990s a. Overall fiscal balance b. Total government debt 15 0 120 100 10 –10 Percentage of GDP Percentage of GDP 80 5 –20 60 0 –30 40 –5 –40 20 –10 –50 0 90 92 94 96 98 00 02 04 06 08 10 90 92 94 96 98 00 02 04 06 08 10 19 19 19 19 19 20 20 20 20 20 20 19 19 19 19 19 20 20 20 20 20 20 MENA oil exporters MENA oil importers MENA oil exporters MENA oil importers MENA oil exporters, non-oil balance as % of non-oil GDP (right axis) c. Current account balance d. External debt 25 0 90 80 15 –0.5 Percentage of GDP Percentage of GDP 70 5 –1.5 60 0 –5 –2.5 50 40 –15 –3.5 30 –25 –4.5 20 90 92 94 96 98 00 02 04 06 08 10 90 92 94 96 98 00 02 04 06 08 10 19 19 19 19 19 20 20 20 20 20 20 19 19 19 19 19 20 20 20 20 20 20 MENA oil exporters (left axis) MENA oil exporters MENA oil importers MENA oil importers (right axis) e. Reserves f. CPI 16 16 14 14 Average annual inflation, % 12 Months of imports 12 10 10 8 6 8 4 6 2 4 0 90 92 94 96 98 00 02 04 06 08 10 90 92 94 96 98 00 02 04 06 08 10 19 19 19 19 19 20 20 20 20 20 20 19 19 19 19 19 20 20 20 20 20 20 MENA oil exporters MENA oil importers MENA oil exporters MENA oil importers Source: IMF 2011. Note: MENA = Middle East and North Africa; GDP = gross domestic product; CPI = consumer price index. 114 JOBS FOR SHARED PROSPERITY BOX 3A.1 Factors contributing to higher total factor productivity The results of a growth accounting exercise for the middle-income countries. Only in high-income MENA region show that physical and human capi- countries are domestic innovations found to contrib- tal has contributed at steady rates to output growth ute significantly to productivity. Benhabib and Spie- over the past three decades for most countries in the gel (1994) argue that for technologically advanced region. Output growth rates, however, varied signifi- economies, domestic innovation is more effective cantly over time. As a result (and by construction), than technology adoption from abroad. Likewise, TFP, also known as the Solow residual, is highly less developed countries do not have the prerequi- volatile as it accounts for all the other unexplained sites for effective domestic innovation, and it is more factors that might lead to an increase in output efficient for them to adapt technologies from abroad. growth. For a better understanding of what drives Allowing for a nonlinear relationship between TFP these changes in output growth rates, some of the and the share of the working-age population with factors that might contribute to a more productive secondary or tertiary education shows that only use of inputs are pinpointed using the framework after reaching a certain threshold of education suggested by Benhabib and Spiegel (1994), which sees (found at 64 percent of secondary and tertiary edu- human capital as the main driver for technological cation) does its contribution through domestic inno- progress (educational attainments are considered to vations turn positive. The specification of column promote domestic innovations on the one hand and 5 includes an interaction term between the level of the adoption of foreign technologies on the other). a country’s development (approximated by GDP The empirical results of this model a in MENA per capita) and the human capital variable, which (table B3A.1.1) support this assumption. Human tests whether human capital speeds up technology capital does not have a signifi cant impact on TFP, adoption. The results show that, holding the level of suggesting that domestic innovations are not the development constant, countries with a higher level driving force in productivity (column 4). Benhabib of education can adapt technology faster and thus and Spiegel (1994) fi nd similar results for low- and are better able to catch up with leading countries. TABLE B3A.1.1 Correlates of total factor productivity growth Log (TFP) (1) (2) (3) (4) (5) (6) (7) L.log(TFP) 0.845*** 0.900*** 0.864*** 0.820*** 0.660*** 0.810*** 0.704*** (0.000) (0.000) (0.000) (0.000) (0.002) (0.000) (0.000) L.log(GDP per capita) –0.0362 –0.0137 0.0271 0.125 0.0288 0.0884 (0.577) (0.867) (0.661) (0.121) (0.715) (0.133) L.Secondary+ –0.00107 –0.00561* –0.00382*** –0.000450 –0.00406** (0.444) (0.065) (0.007) (0.807) (0.038) L.Secondary+ sq. 0.0000434* 0.0000903** (0.067) (0.028) L.log(GDP per capita)* –0.00111* –0.000241 –0.000859** Secondary+ (0.057) (0.593) (0.032) L.Globalization 0.00204 –0.00288 (0.208) (0.107) L.Global* Secondary+ 0.000110** (0.015) Constant 0.956** 0.946*** 0.902*** 0.923** 1.141** 0.811** 1.217** (0.032) (0.000) (0.004) (0.014) (0.021) (0.034) (0.022) Observations 288 285 269 269 269 257 257 R2 0.841 0.861 0.857 0.862 0.872 0.847 0.861 Note: All specifications include fixed time and country effects. P-values in parentheses. GDP = gross domestic product; TFP = total factor productivity. * p < 0.1, ** p< 0.05, *** p< 0.01. (continued next page) A LONG-TERM VIEW OF MENA’S ECONOMIES AND LABOR MARKETS 115 BOX 3A.1 Factors contributing to higher total factor productivity (continued) The insignifi cant coeffi cient on the measure of more technology from abroad and thus, on aver- globalization (column 6) suggests no relationship age, has a higher TFP. Applying the same logic between deeper globalization and a country’s TFP as in column 4, it is estimated that whenever the growth. When the globalization variable interacts share of the working-age population with second- with the human capital variable, the effect of glo- ary or tertiary education exceeds approximately balization intensity depends positively on the level 26 percent (in the case of MENA countries),b glo- of human capital (column 7). Given two countries balization speeds up the process of technology with the same level of human capital, the one with adoption and thus increases TFP and ultimately the higher index of globalization is able to adopt output growth. a. Data on GDP per capita are taken from the World Development Indicators. The share of secondary- and tertiary-educated individuals in the total working-age population (15–64) is obtained from the Barro and Lee (2010) dataset on educational attainment. The IIASA/VID Educational Attainment Model (Lutz et al. 2007) would be preferable, because the back projections are undertaken in a more sensitive manner, but the coverage of the MENA countries is slightly better in the Barro and Lee (2010) dataset. To measure the dimension of globalization, we use the KOF Index of Globalization (Dreher 2006) obtained from the Quality of Governance dataset (Teorell et al. 2011). This index accounts for the three main dimensions of globalization: (1) economic globalization, measured by flows of capital, goods and services; (2) political globalization, characterized by the diffusion of government policies; and (3) social globalization, which captures the spread of ideas, information, images, and people (Dreher 2006). b. The GCC countries (in our sample, Bahrain, Kuwait, Saudi Arabia, and the United Arab Emirates) show the highest average level of educational attainment in the working-age population since the 1980s. However, while Bahrain had already reached the critical level of education in the mid-1990s, Saudi Arabia and the United Arab Emirates did so only in the past several years. Educational attainment in Kuwait is below the level necessary to positively contribute to productivity through the channel of domestic innovation. Since the 1980s, the average educational attainment in the Mashreq region (in our sample, Egypt, Jordan, Syria, and the Republic of Yemen) has clearly been below the GCC average; however, the sharp increase in educational attainment in Egypt and, in particular, Jordan enables the two countries to effectively adopt and implement new technologies. Syria and the Republic of Yemen both have educational levels that are insufficient for a positive contribution of domestic innovation to TFP. Concerning the Maghreb region, which is represented by Algeria, Morocco, and Tunisia in our sample, our model predicts that the level of human capital in those countries is not high enough to effectively innovate domestically. Annex 3B Demographics TABLE 3B.1 Population increase, net migration, and rate of population change in selected economies in MENA, mid-1980s–mid-2000s Rate of natural increase in Average annual rate of population (%) Net migration rate (%) population change (%) Economy group Economy 1985–90 1995–2000 2005–10 1985–90 1995–2000 2005–10 1985–90 1995–2000 2005–10 GCC Bahrain 4.1 2.9 2.6 6 7 90 3.3 2.6 11.1 GCC Kuwait 3.3 2.9 2.3 12 14 22 3.6 3.5 3.8 GCC Oman 7.9 4.5 2.5 1 –20 12 3.9 0.3 2.7 GCC Qatar 4.5 3.3 2.4 27 16 133 5.1 3.3 15.2 GCC Saudi Arabia 6.2 4.5 3.0 7 –9 8 4.0 1.6 2.7 GCC United Arab Emirates 4.8 3.0 1.9 33 35 106 5.9 5.1 12.3 GCC economy average 5.1 3.5 2.5 14.3 7.2 61.8 4.3 2.7 8.0 Low- Djibouti income 6.4 5.1 4.0 37 9 0 6.7 3.1 1.9 Low- Yemen, Arab Rep. income 8.9 7.0 5.5 –1 –1 –1 4.0 3.1 3.1 Low-income economy average 7.7 6.0 4.7 18.0 4.0 –0.5 5.4 3.1 2.5 Iraq Iraq 6.2 5.4 4.9 –8 0 –1 2.3 3.2 2.9 Mashreq Jordan 6.4 4.3 3.3 8 –8 7 4.0 1.9 2.9 Mashreq Lebanon 3.3 2.7 1.9 –15 0 –1 0.4 1.6 0.8 Mashreq Syrian Arab Republic 5.9 4.0 3.1 –3 –2 –1 3.1 2.4 2.0 (continued next page) 116 JOBS FOR SHARED PROSPERITY TABLE 3B.1 Population increase, net migration, and rate of population change in selected economies in MENA, mid-1980s–mid-2000s (continued) Rate of natural increase in Average annual rate of population (%) Net migration rate (%) population change (%) Economy group Economy 1985–90 1995–2000 2005–10 1985–90 1995–2000 2005–10 1985–90 1995–2000 2005–10 Mashreq economy average 5.2 3.7 2.7 –3.3 –3.3 1.7 2.5 2.0 1.9 Maghreb Algeria 5.3 2.9 2.4 –1 –1 –1 2.7 1.5 1.5 Maghreb Libyan Arab Jamahiriya 5.7 3.3 2.7 –1 –1 –1 2.4 1.8 1.9 Maghreb Morocco 4.5 3.0 2.4 –2 –4 –4 2.1 1.3 1.0 Maghreb Tunisia 4.1 2.3 2.0 –1 –1 0 2.3 1.1 1.1 Maghreb economy average 4.9 2.9 2.4 –1.3 –1.8 –1.5 2.4 1.5 1.4 Egypt, Arab Egypt, Arab Rep. 4.8 3.5 2.9 –2 –3 –1 2.3 1.7 1.8 Rep. Israel 3.1 2.9 2.9 3 9 8 1.9 2.4 2.3 Malta 2.1 1.8 1.3 –4 5 –5 1.0 0.6 0.4 West Bank and Gaza 6.4 5.8 4.7 1 2 1 3.4 4.2 2.6 Source: UN Population Information Network . Note: Variables are defined in table 3A.2.2. GCC = Gulf Cooperation Council; MENA = Middle East and North Africa. TABLE 3B.2 Demographic variables defined Variable Definition Population De facto population in a country, area, or region as of 1 July of the year indicated. Figures are presented in thousands. Population by sex De facto population as of 1 July of the year indicated classified by sex (male, female, both sexes combined). Data are presented in thousands. Population sex ratio Number of males per 100 females in the population. Median age Age that divides the population in two parts of equal size; that is, there are as many persons with ages above the median as there are with ages below the median. Population change Population increment over a period; that is, the difference between the population at the end of the period and that at the beginning of the period. Refers to five-year periods running from 1 July to 30 June of the initial and final years. Data are presented in thousands. Population growth Average exponential rate of growth of the population over a given period. It is calculated as ln(Pt/P0)/t, where t is the rate length of the period. It is expressed as a percentage. Rate of natural Crude birth rate minus the crude death rate. Represents the portion of population growth (or decline) determined increase exclusively by births and deaths. Crude birth rate Number of births over a given period divided by the person-years lived by the population over that period. It is expressed as number of births per 1,000 population. Crude death rate Number of deaths over a given period divided by the person-years lived by the population over that period. It is expressed as number of deaths per 1,000 population. Net reproduction The average number of daughters a hypothetical cohort of women would have at the end of their reproductive period if rate they were subject during their whole lives to the fertility rates and the mortality rates of a given period. It is expressed as number of daughters per woman. Total fertility The average number of children a hypothetical cohort of women would have at the end of their reproductive period if they were subject during their whole lives to the fertility rates of a given period and if they were not subject to mortality. It is expressed as children per woman. Mean age at The mean age at childbearing is the mean age of mothers at the birth of their children if women were subject throughout childbearing their lives to the age-specific fertility rates observed in a given year. Life expectancy by The average number of years of life expected by a hypothetical cohort of individuals who would be subject during all their sex lives to the mortality rates of a given period. It is expressed as years. Net migration rate The number of immigrants minus the number of emigrants over a period, divided by the person-years lived by the population of the receiving country over that period. It is expressed as net number of migrants per 1,000 population. Net migration Net number of migrants; that is, the number of immigrants minus the number of emigrants. It is expressed in thousands. Sex ratio at birth Number of male births per one female birth. Source: World Bank. A LONG-TERM VIEW OF MENA’S ECONOMIES AND LABOR MARKETS 117 TABLE 3B.3 Reproduction rates and average annual population growth in selected economies in MENA, mid-1980s–mid-2000s Average annual population Total fertility rate (%) Adult mortality rate (%) Net reproduction rate (%) growth (%) Economy group Economy 1985–90 1995–2000 2005–10 1985–90 1995–2000 2005–10 1985–90 1995–2000 2005–10 1985–90 1995–2000 2005–10 GCC Bahrain 4.1 2.9 2.6 106.6 90.7 1.9 1.4 1.3 3.3 2.6 11.1 GCC Kuwait 3.3 2.9 2.3 96.8 89.7 1.6 1.4 1.1 3.6 3.5 3.8 GCC Oman 7.9 4.5 2.5 128.5 120.4 3.5 2.1 1.2 3.9 0.3 2.7 GCC Qatar 4.5 3.3 2.4 82.5 67.6 2.1 1.6 1.2 5.1 3.3 15.2 GCC Saudi Arabia 6.2 4.5 3.0 153.4 123.4 2.8 2.1 1.4 4.0 1.6 2.7 GCC United Arab Emirates 4.8 3.0 1.9 121.7 85.7 2.3 1.4 0.9 5.9 5.1 12.3 GCC economy average 5.1 3.5 2.5 114.9 96.3 2.4 1.7 1.2 4.3 2.7 8.0 Low- income Djibouti 6.4 5.1 4.0 347.6 315.0 2.3 1.9 1.6 6.7 3.1 1.9 Low- income Yemen, Rep. 8.9 7.0 5.5 283.9 222.1 3.4 2.8 2.4 4.0 3.1 3.1 Low-income economy average 7.7 6.0 4.7 315.8 268.6 2.9 2.4 2.0 5.4 3.1 2.5 Mashreq Jordan 6.4 4.3 3.3 142.2 126.2 2.9 2.0 1.5 4.0 1.9 2.9 Mashreq Lebanon 3.3 2.7 1.9 149.9 130.6 1.5 1.3 0.9 0.4 1.6 0.8 Mashreq Syrian Arab Republic 5.9 4.0 3.1 119.8 97.9 2.7 1.9 1.5 3.1 2.4 2.0 Mashreq economy average 5.2 3.7 2.7 137.3 118.3 2.4 1.7 1.3 2.5 2.0 1.9 Maghreb Algeria 5.3 2.9 2.4 151.6 120.6 2.3 1.3 1.1 2.7 1.5 1.5 Maghreb Libyan Arab Jamahiriya 5.7 3.3 2.7 151.0 120.1 2.6 1.5 1.3 2.4 1.8 1.9 Maghreb Morocco 4.5 3.0 2.4 155.4 125.3 1.9 1.3 1.1 2.1 1.3 1.0 Maghreb Tunisia 4.1 2.3 2.0 119.5 101.6 1.9 1.1 1.0 2.3 1.1 1.1 Maghreb economy average 4.9 2.9 2.4 144.4 116.9 2.1 1.3 1.1 2.4 1.5 1.4 Iraq Iraq 6.2 5.4 4.9 128.5 177.6 2.7 2.4 2.2 2.3 3.2 2.9 Egypt, Egypt, Arab Arab Rep. Rep. 4.8 3.5 2.9 169.2 119.6 2.0 1.6 1.3 2.3 1.7 1.8 Israel 3.1 2.9 2.9 80.5 65.5 1.5 1.4 1.4 1.9 2.4 2.3 Malta 2.1 1.8 1.3 81.7 68.4 2.9 2.7 2.2 1.0 0.6 0.4 West Bank and Gaza 6.4 5.8 4.7 150.2 129.5 1.0 0.9 0.6 3.4 4.2 2.6 Source: UN Population Information Network. Note: Variables are defined in table 3A.2.2. MENA = Middle East and North Africa; IDA = International Development Association; GCC = Gulf Cooperation Council. Notes in using these inputs declined. Algeria’s rate of physical capital accumulation almost qua- 1. Intraregional migration flows are very drupled between the 1960s and the 1970s, important in the region. According to the but TFP growth went from 1.9 percent per International Organization for Migration, out year to –0.7 percent. Jordan tripled its rate of 13.0 million Arab migrants in the world, of physical capital accumulation, but TFP 5.8 million reside in Arab countries. Migration growth declined by 40.0 percent (from contributes to the circulation of financial and 2.8 percent per year to 1.7 percent). Morocco human capital within the region. Remittances doubled its rate of accumulation, but TFP sent to Egypt, Jordan, and Lebanon from growth fell from 1.8 percent per year to other Arab countries are 40–190 percent –0.3 percent per year. In the Gulf, Saudi higher than trade revenues between these and Arabia’s doubling of its physical capital other Arab countries. accumulation was accompanied by a decline 2. During that decade, MENA realized the in average annual TFP growth from 4.7 per- highest rates of growth in the world in both cent in the 1960s to zero in the 1970s (Keller physical and human capital, but the efficiency and Nabli 2002; rates are per laborer). 118 JOBS FOR SHARED PROSPERITY 3. The largest increase in intraregional aid applied most-favored nation tariff rate falling occurred in the 1970s, followed by a sharp by a third over 2000–07, to 15 percent. decline in the following decade. Between 8. MENA countries also expanded the reach of 1973 and 1986, Arab countries were the larg- export markets during the 2000s, as shown est donors to the Arab region. In 1980, inter- by the sharp increase in existing exports to Arab overseas development aid flows reached new markets (see figure 3A.1 in annex 3.1). a maximum of some US$9 billion, accounting 9. See Diop, Marotta, and DeMelo (2012). for almost 60 percent of total overseas devel- 10. This trend is captured by the relatively opment aid flows to Arab countries at that strong, positive correlation between the pace time. In absolute terms, during the 1970s, the of employment creation, as given by the Arab countries provided almost 3.5 times as employment growth elasticities, and infor- much bilateral aid as the next highest donor mality rates in MENA in the 2000s. group and 1.5 times more aid than all other 11. According to a recent report by the World donors combined. The strong rise in oil prices Bank (2012b), the average country in the during the 1970s led to increased investment region produces about one-third of its GDP flows from the countries of the GCC to the and employs two-thirds of its labor force rest of the region and to a surge in workers’ informally. remittances attributed to lower oil revenues 12. As recently argued in World Bank (2011). and an increased tendency of oil-rich coun- 13. The latter are, however, not reviewed in this tries to invest in their own infrastructure chapter. projects (see UN 2007). 14. Defined as a period of unemployment decline 4. In per capita terms, the growth rate of capital of at least three percentage points and at least stock was cut by almost three quarters. a quarter of its initial level over a four-year 5. See table B3A.1.1 in annex 3A. period that persists for at least another three 6. A group of resource-poor countries, includ- years. ing Morocco and Tunisia, followed by Egypt 15. The observations reported below refer to and Jordan, implemented earlier and more well-known and widely accepted policy rec- intensive reforms toward more open and ommendations for the region, and it is by private sector–led economies than the rest no means an exhaustive list. As there is no of the countries of the region. Other reform- “one-size-fits-all” policy option, policies need ers, including Algeria, the Islamic Republic to be tailored to individual countries’ needs; of Iran, and Syria (all resource-rich coun- yet common issues across the region can be tries) also pursued reforms but later on, more raised and addressed. gradually, and more sporadically than the 16. Defined as the standard deviation of the REER, early reformers. Algeria, for instance, with this is about four times higher in MENA than macroeconomic imbalances stemming from in countries belonging to the high-income the collapse in oil prices, aggressively pursued group (see table 3A.2 in annex 3.1). macroeconomic stabilization, but structural 17. For a more detailed discussion of the role reforms have been far more limited. Early of REER in MENA, see Diop, Marotta and on, the GCC managed to achieve an open DeMelo (2012). trade system with free movement of capital 18. This is due to a lack of political incentives or and advanced financial systems. Some of the institutions that impose constraints on pol- smaller GCC countries encouraged growth icy makers (Elbadawi and Soto 2011; Ross in selected sectors such as financial services 2012). There is no fiscal rule in any country of and tourism (Bahrain and the United Arab MENA. The political economy of the MENA Emirates). Oman made substantial efforts countries could be a major reason for the lack to broaden private sector participation and of a fiscal rule, owing to the limited political improve the foreign investment climate, with accountability in most MENA countries. privatizations and changes in its foreign capi- 19. See World Economic Forum and OECD tal investment law. In Saudi Arabia, reforms (2005) and World Bank (2012c). began later and progressed more slowly. 20. Chapter 2 of Diop, Marotta, and DeMelo 7. MENA became more open as a region, as (2012) presents a series of policy indicators average protection, which started from higher affecting trade that show the region is lagging in averages than competitors, fell more rapidly global integration. For further information on than elsewhere, with the region’s average how, despite recent trade integration reforms, A LONG-TERM VIEW OF MENA’S ECONOMIES AND LABOR MARKETS 119 MENA did not succeed in fully exploiting the Research Conference,” Washington, DC, benefits of participating in global production November 8–9. networks, increasing global trade in services, Gelb, A. 1988. Oil Windfalls: Blessing or Curse? and regional integration, see Lopez-Calix, New York: Oxford University Press. Walkenhorst, and Diop (2010) and World Hausmann, R., U. Panizza, R. Rigobon. 2004. Bank (2003). “The Long-Run Volatility Puzzle of the Real Exchange Rate.” NBER Working Paper 10751, National Bureau of Economic Research, References Cambridge, MA. Amin, M., and S. Djankov. 2009. “Democracy Hausmann, R., L. Pritchett, and D. Rodrik. 2005. and Reforms.” Policy Research Working Paper “Growth Accelerations.” Journal of Economic 4835, World Bank, Washington, DC. Growth 10: 303–29. Barro, R., and J. Lee. 2010. “A New Data Set ILO-KILM (International Labour Organization– of Educational Attainment in the World, Key I ndicators of the Labour Market) 1950–2010.” NBER Working Paper 15902, (database). 7th ed. International Labour National Bureau of Economic Research, Organization, Geneva, Switzerland. http:// Cambridge, MA. kilm.ilo.org/kilmnet/. Behar, A., and C. Freund. 2011. “The Trade ILO -LA BORSTA (database). International Performance of Middle East and North Africa.” Labour Organization, Geneva, Switzerland. Middle East and North Africa Working Paper http://laborsta.ilo.org/. 53, World Bank, Washington, DC. IMF (International Monetary Fund). 2011. Benhabib, J., and M. M. Spiegel. 1994. “The Role “Economic Transformation in MENA: Delivering of Human Capital in Economic Development: the Promise of Shared Prosperity.” Paper Evidence from Aggregate Cross-Country Data.” prepared for the G-8 Summit, “Transformation Journal of Monetary Economics 34 (2): 143–73. in MENA,” Deauville, France, May 27. Commission on Growth and Development. 2008. The K e l l e r, J . , a n d M . N ab l i . 2 0 0 2 . “ T h e Growth Report: Strategies for Sustained Growth Macroeconomics of Labor Market Outcomes and Inclusive Development. Washington, DC: in MENA over the 1990s: How Growth Has World Bank. https://openknowledge.worldbank Failed to Keep Pace with a Burgeoning Labor .org/handle/10986/6507. Market.” Paper presented at the Egyptian Diop, N., D. Marotta, and J. DeMelo. 2012. Center for Economic Studies Conference, Natural Resource Abundance, Growth and Cairo, January. Diversification in M ENA . Directions in Lopez-Calix, J., P. Walkenhorst, and N. Diop, eds. Development Series. Washington, DC: World 2010. Trade Competitiveness of the Middle East Bank. and North Africa. Directions in Development Dreher, A. 2006. “Does Globalization Affect Series. Washington, DC: World Bank. Growth? Evidence from a New Index of Lutz, W., A. Goujon, K. C. Samir, and W. Globalization.” Applied Economics 38 (10): S a nderson. 20 07. “Re con st r uc t ion of 1091–1110. Popu lat ion by Age , S ex a nd L evel of EAPEP (Economic Alive Population, Estimates Educational Attainment of 120 Countries for and Projections) database. 2011. 6th ed. 1970–2000.” In Vienna Yearbook of Population International Labour Organization, Geneva, Research, vol. 5, 193–235. Vienna: Vienna Switzerland. http://laborsta.ilo.org/applv8/ Institute of Demography. data/EAPEP/eapep_E.html. M c M i l l a n , M . , a n d D. R o d r i k . 2 011. Elbadawi, I. A., and R. Soto. 2011. “Fiscal “Globalization, Structural Change and Regimes in and outside the MENA Region.” Productivity Growth,” NBER Working Paper Documentos de Trabajo 398, Instituto de 17143, National Bureau of Economic Research, Economia. Pontificia Universidad Católica de Cambridge, MA. Chile. Santiago de Chile, Chile. Nehru, V., and A. Dhareshwar. 1993. “A New Freund, C., and B. Bolaky. 2008. “Trade, Database on Physical Capital Stock: Sources, Reg ulations, and I ncome.” Jour n al of Methodology and Results.” Revista de Development Economics 87 (2): 309–21. Analisis Economica 8 (1): 37–59. Freund, C., and B. Rijkers. 2012. “Employment Ross, M. 2012. The Oil Curse: How Petroleum Miracles.” Paper presented at the International Wealth Shapes the Development of Nations. Monetary Fund, “13th Jacques Polak Annual Princeton, NJ: Princeton University Press. 120 JOBS FOR SHARED PROSPERITY Teorell, Jan, Nicholas Charron, Marcus Samanni, ———. 2012b. “Applying the Limited Access Sören Holmberg, and Bo Rothstein. 2011. Order and Rentier State Models to MENA “The Quality of Government Dataset.” The Countries.” World Bank, Washington, DC. Quality of Government Institute, University ———. 2012c. “From Political to Economic of Gothenburg, Gothenburg, Sweden. http:// Awakening in the Arab World: The Path www.qog.pol.gu.se. of E c o nom i c I nt e g r at io n .” D e auv i l l e UN (United Nations). 2007. Economic Trends Partnership Report on Trade and Foreign and Impacts: Foreign Aid and Development Di re c t I nve st ment (i n Two Volu me s). in the Arab Region . New York: United Report No. 68832-M NA , World Bank, Nations. Washington, DC. https://openknowledge UN Commodity Trade Statistics (database). United .worldbank.org /handle/10986/11886 and Nations, New York, NY. http://comtrade ht t ps: //op en k nowledge.worldba n k.org / .un.org/db/. handle/10986/11887. UN Population Information Network (database). ——— . Forthcoming. Inclusion and Resilience: United Nations, New York, NY. The Way Forward for Social Safety Nets in the U NC TA D (U N Conference on Trade and Middle East and North Africa. Washington, Development) (database). United Nations, DC: World Bank. New York, NY. http://unctad.org/en/Pages/ World Development Indicators (database). Statistics.aspx/. World B a n k , Wash i ng ton , DC . ht t p: // World Bank. 2003. Trade, Investment and d a t a .w o r l d b a n k . o r g /d a t a - c a t a l o g / Development in MENA . Washington, DC: world-development-indicators. World Bank. World Economic Forum. 2012. The Global ———. 2004. Jobs, Growth, and Governance Competitiveness Report 2012–2013. Geneva: in M EN A: Unlocking the Potential for World Economic Forum. Prosperity. Washington, DC: World Bank. World Economic Forum and OECD (Organisation ———. 2009. From Privilege to Competition : for Economic Co-operation and Development). Unlocking Private-Led Growth in the Middle 2005. Arab World Competitiveness Report. East and North Africa . Washington, DC: Geneva: World Economic Forum and OECD. World Bank. World Governance Indicators (database). World ———. 2010. “Republic of Tunisia Development Bank, Washington, DC. http://info.worldbank Policy Review: Towards Innovation-Driven .org/governance/wgi/. Growth.” Report No. 50847-TN, World Bank, World Integrated Trade Solution (database). Washington, DC. World Bank, Washington, DC. http://wits ———. 2011. Middle East and North Africa .worldbank.org/wits/. Economic Developments and Prospects, World Trade Indicators (database). World Bank, September 2011: Investing for Growth and Washington, DC. http://web.worldbank.org/ Jobs. Washington, DC: World Bank. WBSITE/EXTERNAL/TOPICS/TRADE/0 ———. 2012a. Jordan Development Policy contentMDK:22421950~pagePK:148956~piPK Review. Washington, DC: World Bank. :216618~theSitePK:239071,00.html. Privileges, not Competition: The Dynamics of the Private Sector in 4 the Middle East and North Africa Main findings • The process of creative destruction is attenuated in MENA, limiting productivity growth. This is reflected in a suboptimally small private sector, skewed toward unproductive small-scale activities. • The lack of dynamism, which is also manifested in low rates of firm entry and exit as well as in lower growth of incumbent firms, comes at a significant cost in terms of employment. • Distortive policies, such as subsidies on fuel and high taxes on labor, depress the demand for labor and simultane- ously stifle incentives to innovate. • Burdensome business regulations are often used as a means to redistribute rents rather than to catalyze creative destruction. Their implementation is discretionary and inconsistent, resulting in an uneven access to key inputs, such as credit, licenses, permits, customs clearing, and the like. • The ensuing uncertainty in policy implementation hampers investment and job creation, especially of those high-value-added jobs to which the increasingly educated youth in MENA aspire. MENA’s lack of economic Lackluster performance in dynamism higher-value-added activities T wo features of MENA economies While on average, private investment rates in seem to be at the root of the private the Middle East and North Africa (MENA) sector’s lackluster performance in job lag behind comparator economies, up to the creation: (1) MENA’s inability to diversify time of the economic crisis a few countries— into higher productivity activities, and (2) including the Islamic Republic of Iran, the lack of creative destruction; that is, the Jordan, Lebanon, Morocco, and Tunisia— dynamic process of reallocation from lower had achieved investment rates comparable to to higher productivity activities. China and Turkey. Yet, even these countries 121 122 JOBS FOR SHARED PROSPERITY failed to create sufficient productive employ- The evidence also shows that, compared ment for skilled labor in manufacturing and to firms in fast-growing East Asian coun- services, attesting to the region’s lack of suc- tries, fi rms in MENA are slow in diversify- cess in diversifying into higher-productivity ing their exports into higher-productivity activities. goods.1 Figure 4.1 illustrates the evolution of Rather than the quantity, the quality of the productivity content of exports over time private investments appears to be lacking in Jordan, Lebanon, Morocco, and Tunisia in these countries. For example, the invest- relative to selected East Asian countries. The ment inflows into MENA countries are often productivity content in all four countries has neither technology nor skills intensive. Fig- increased somewhat since the 1990s. Never- ure 3.16 in chapter 3 disaggregates foreign theless, the region’s evolution appears stag- direct investment (FDI) stocks into the main nant compared to the increases observed in economic sectors. Despite some of the high- China, Thailand, and Vietnam. For instance, est FDI inflows in the world in some coun- Thailand’s export sophistication index was tries (Jordan and Lebanon), FDI inflows are comparable to Morocco’s in the early 1980s biased toward low-value-added service sec- but exceeded those of Jordan, Morocco, and tors (often reflecting real estate investments Tunisia by almost 50 percent in 2008.2 from countries in the Gulf Cooperation These dynamics are also reflected in the low Council, or GCC), which typically have very generation and use of knowledge in the pro- low potential for technology spillovers. In duction processes across the MENA region. contrast, FDI inflows into China or the Euro- Overall, MENA countries innovate less than pean Union are concentrated in manufac- other countries at similar levels of develop- turing or higher-technology services, which ment. This is illustrated in figure 4.2, which potentially facilitate foreign technology and plots the World Bank knowledge economy knowledge transfers. This process of integra- score on innovation; this score proxies the tion into the global economy through for- degree of production and use of knowledge eign technology adoption has been essential against gross domestic product (GDP) per to high and sustained growth in total factor capita. 3 Moreover, even among the highest productivity (TFP) and facilitated moving up the value chain. FIGURE 4.2 Knowledge economy innovation score and GDP per capita for selected world FIGURE 4.1 Evolution of the productivity content of exports in regions and countries in MENA, 2009 selected regions and countries in Asia and MENA, 1980–2009 High-income 17,500 countries 10 KE innovation pillar score Productivity content of exports Upper-middle- 15,500 income countries 8 (normalized) Jordan ECA United 13,500 EAP Qatar 6 Iran, Islamic Rep. Arab Emirates Saudi Arabia Tunisia Oman 11,500 Lebanon Bahrain 4 Morocco Egypt, Arab Rep. Algeria Syrian Arab Republic Middle-income countries 9,500 2 Yemen, Rep. Djibouti 7,500 0 10,000 20,000 30,000 40,000 GDP per capita 5,500 1980 1984 1988 1992 1996 2000 2004 2008 World MENA Fitted values Tunisia Jordan China Thailand Vietnam Lebanon Morocco Source: Knowledge Economy Innovation Score and World Development Indicators. Note: GDP = gross domestic product; EAP = East Asia and the Pacific; Source: Based on Comtrade data. ECA = Europe and Central Asia; GDP = gross domestic product; Note: MENA = Middle East and North Africa. KE = knowledge economy; MENA = Middle East and North Africa. PRIVILEGES, NOT COMPETITION 123 performers in the region (Jordan, Qatar, and some of the lowest densities of firm entry the United Arab Emirates), most research among emerging economies. Entry densities and development are publicly financed rather are measured by the average yearly number than driven by the private sector (figure 4.2). of newly registered limited liability firms per 1,000 working-age people (between ages 15 and 64) in the period 2004–09. In this sam- Lack of creative destruction ple, almost two-thirds of the 80 emerging A dynamic process of creative destruction countries for which data are available had a underpinned technological upgrading in fast- higher entry density than the two best per- growing East Asian and Eastern European formers in non-GCC countries in MENA: economies.4 In these countries, resources Morocco and Tunisia. For instance, with a were reallocated toward more productive working-age population comparable to Jor- uses among existing firms as well as from dan’s, Croatia’s number of newly registered exiting to entering firms. In contrast, firm firms was five times higher than Jordan’s. dynamics in the MENA region are stagnant. A lack of turnover is also manifested in The region experiences lower fi rm entry and lower fi rm exit. Figure 4.4 illustrates entry exit as well as lower growth of incumbent and exit rates among Moroccan manufactur- fi rms than relevant comparator countries in ing fi rms. The average annual exit rate for other parts of the world. This low turnover manufacturing fi rms is 5.1 percent over the significantly limits the scope for creative sample period, which is substantially lower destruction. than exit rates documented in other devel- Figure 4.3 shows that the Arab Republic oping countries; for instance, in Chile and of Egypt, Jordan, Morocco, and Tunisia have Colombia, exit rates have been measured as FIGURE 4.3 Average entry density for selected emerging economies, 2004–09 4.00 3.50 3.00 Entry density 2.50 2.00 1.50 1.00 0.50 0 s ia tia ile lic il a ey co an sia p. a rie az isi di ar Re ub oc rk Ch rd oa ne In n Br nt lg Tu Tu Jo or b ep Cr do Bu ou ra M nR In t, A 0c ica e8 yp in Eg ag m er Do Av MENA countries Comparator countries Source: Klapper and Love 2010. Note: Entry density = the number of newly registered limited liability firms per 1,000 working-age people (those between ages 15 and 64). The average of 80 countries represents the average entry density in the 80 developing countries for which data are available. MENA = Middle East and North Africa. 124 JOBS FOR SHARED PROSPERITY amounting to about 11.0 and 12.0 percent, Furthermore, table 4.1 shows that the small- respectively. est firms are also the most likely to exit. As a Moreover, the results from a census of result, the two countries have relatively many firms in Morocco and Tunisia show that small fi rms despite comparatively low entry employment growth stagnates in incumbent rates because fi rms tend to enter small and firms: few firms grow so that small firms not grow. This lack of economic dynamism is stay small. In particular, table 4.1 documents at the root of MENA’s unemployment prob- transitions of firms among crude size cat- lem: fi rms fail to create enough new jobs to egories in Morocco and Tunisia. Even after absorb a rapidly growing labor force. 10 years, one-person, microfi rms, and small In contrast, fi rm dynamics in competitive firms are extremely unlikely to have become economies are typically characterized by an large firms. In addition, very few micro- “up-or-out” pattern: fi rms that are success- fi rms graduate into the small size category. ful enough to stay in business grow, while the least productive ones exit. A comparison of average firm size by age in Brazil and Jordan FIGURE 4.4 Firm turnover and sector growth in Moroccan manufacturing, 1986–2001 illustrates the lack of that dynamic in MENA countries; firms start out larger in Jordan but grow more slowly over time, such that 15 8 firms in Brazil are about twice as large after 6 10 years in operation (figure 4.5).5 Competition is a catalyst in the process 4 10 of creative destruction, which, in turn, is Percent Percent 2 crucial for productivity growth and hence 0 employment creation. In particular, indus- 5 –2 tries that are contestable— that is, in which –4 fi rms can enter and challenge incumbents— –6 tend to exhibit much more rapid productiv- 0 ity improvements, not only because more 86 89 92 95 98 01 19 19 19 19 19 20 productive firms are entering the market but Entry rate (left axis) Exit rate (left axis) also because the threat of entry itself serves Manufacturing sector growth (right axis) as a disciplining device, forcing incumbents Source: Bartelsman, Haltiwanger, and Scarpetta 2004. to innovate more rapidly. Figures 4.6 and Note: The total firm turnover is the entry rate plus exit rate. The entry rate is defined as the number 4.7 show that across industries in a sample of new firms divided by the total number of incumbents and entrant firms producing in a given year. The exit rate is defined as the number of firms exiting the market in a given year divided by the of 24 industrial and developing countries, population of origin; i.e., the incumbents in the previous year. the contribution of entrants to aggregate TABLE 4.1 Mobility among size categories for firms in Morocco and Tunisia, 1996–2010 Long-term transition matrix, percentages Tunisia: all private firms 2000–10 (excluding cooperatives) Morocco: manufacturing only, 1996–2006 Status in year t + 10 Status in year t + 10 Status in year t 1-person Micro Small Large Status in year t Exited Micro Small Large (baseline) Exited firm firm firm firm (baseline) firm firm firm firm 1-person firm 30.8 65.5 3.4 0.3 0.0 Micro firm 19.0 41.2 37.0 2.8 0.1 Micro firm 52.1 36.5 11.3 0.1 Small firm 14.8 28.4 14.0 39.5 3.3 Small firm 44.6 9.5 41.2 4.8 Large firm 15.8 23.3 2.9 15.2 42.7 Large firm 40.7 0.6 12.9 45.9 Source: Rijkers and Arouri 2012. Note: Microfirms = 2–9 employees; small firms = 10–99 employees; large firms = 100 or more employees. In Morocco, the microcategory also contains one-person firms, but there are only a few (the data include only firms whose turnover exceeds a specific threshold). PRIVILEGES, NOT COMPETITION 125 productivity growth is strongly correlated FIGURE 4.5 Relationship between firm size and age in Brazil, 2009, with the contribution of incumbents to and Jordan, 2006 aggregate productivity growth (figure 4.6). Moreover, in industries with more firm turn- over, both (surviving) incumbents (figure 50 4.7a) and (surviving) entrants (figure 4.7b) tend to increase productivity faster. This 40 process of creative destruction is lacking in Number of firms MENA. 30 Although the average private sector per- formance in MENA is not rosy, aggregate statistics obscure some significant success 20 stories. Many of the region’s successful entre- preneurs, however, come from the highest 10 political and country leadership spheres. 6 In some countries, this factor is transpar- 0 5 10 15 20 Age of the firm (years) ent and highly visible. In others, networks Brazil Jordan of businessmen allied with the ruling elites of society, the military, or politicians domi- nate the private sector. Although such ties Source: Based on Enterprise Surveys in Jordan (2006) and Brazil (2009). are not always sufficient for success, they can Note: Data apply to private domestic firms only. often get things started in a heavily regulated environment. High entry barriers protecting FIGURE 4.6 Relationship between the monopolies, privileged positions in highly contribution of net entry and incumbents to regulated sectors, preferential access to large productivity growth public procurement contracts, and other non- competitive practices have made some of the Labor productivity—pooled manufacturing region’s most publicized and spectacular suc- Five-year differencing, real gross output, Incumbent’s productivity growth cesses possible. country and industry time average Overall, the lack of firm turnover is a 1.5 symptom of a discriminatory business envi- 1.0 ronment that prevents unproductive incum- bents from going out of business and limits 0.5 incentives for innovation, thereby undermin- 0 ing the process of creative destruction. The following section shows that this lack of –0.5 –1.0 –0.5 0 0.5 1.0 1.5 economic dynamism in MENA takes a sig- Net entry productivity growth nificant toll on employment, in particular on skilled labor. Source: Bartelsman, Haltiwanger, and Scarpetta 2004. Note: Data exclude Brazil, República Bolivariana de Venezuela, and outliers. Firm dynamics and employment existing large firms tend to account for a dis- creation proportionate share of employment. The private sector in MENA countries is sub- While micro and small firms typically optimally small and skewed toward unpro- grow very little if at all in MENA countries, ductive small-scale activities. Figure 4.8 the literature on fi rm growth suggests that, depicts the fi rm-size distribution of formal in more competitive economies, younger enterprises in Tunisia, which is overwhelm- and smaller fi rms have higher employment ingly populated by one-person businesses.7 growth rates than older and larger firms The figure also illustrates that the few (see, for example, Ayyagari, Demirguc-Kunt, 126 JOBS FOR SHARED PROSPERITY FIGURE 4.7 Firm turnover and productivity growth in 24 industrial and developing countries, various years, 1979–2000 a. Labor productivity—pooled manufacturing b. Labor productivity—pooled manufacturing Five-year differencing, real gross output, Five-year differencing, real gross output, country and industry time average country and industry time average 1.5 1.5 Mean difference in entry and Incumbent’s productivity 1.0 exit productivity 1.0 0.5 growth 0 0.5 –0.5 –1.0 0 0 0.5 1.0 1.5 –1.0 –0.5 0 0.5 1.0 1.5 Firm turnover rate (f ) Firm turnover rate (f ) Source: Bartelsman, Haltiwanger, and Scarpetta 2004. Note: Data exclude Brazil, R. B. de Venezuela, and outliers. FIGURE 4.8 Distribution of firms in Tunisia by size and share of age rather than firm size in the United States. employment, 1997–2009 Young fi rms appear to be the driver of job creation; in particular, the authors show that All formal private firms except for cooperatives fi rm startups contribute substantially to net 100 job creation. However, this engine stalls if 90 86 fi rm entry is obstructed or if the privileges 80 70 of mature incumbents inhibit the growth 60 opportunities of young firms by excluding Percent 50 their access to competitive regulatory ser- 40 37 29 vices, land, or credit. 30 20 The importance of enabling firm growth is 20 13 12 10 2 illustrated in figure 4.9, showing that larger 0 0 firms both have higher labor productiv- One-person Micro firm Small firm Large firm ity and pay higher wages in Morocco. They firm (2–10 (11–100 (>100 employees) employees) employees) also tend to hire more skilled workers. These % of employment % of firms findings highlight the importance of creating an enabling environment in which firms can Source: Repertoire National des Entreprises Tunisien (statistics provided by the Institut National de grow, exploit economies of scale, and create la Statistique, Tunisia). well-paid jobs. Conversely, economic struc- Note: Data cover all formal private firms except cooperatives. tures or policies discriminating against young (small) fi rms, owing to their lack of influen- tial connections, reduce their opportunities Maksimovic 2011; Hall 1987; Hart and to grow and hence reduce the scope for pro- Oulton 1996; Mansfield 1962). Haltiwanger, ductive private sector job creation. Jarmin, and Miranda (2010) have given Most microeconomic studies also find nuance to these findings, showing that net a positive relation between innovation and employment growth is associated with fi rm employment creation. 8 In this regard, it is PRIVILEGES, NOT COMPETITION 127 FIGURE 4.9 Labor productivity, wages, and firm A relatively recent finding in the litera- size in Moroccan manufacturing, 1996–2006 ture that links employment growth to fi rm dynamics extends these results by suggest- Moroccan manufacturing firms ing that a small group of fast-growing firms, 4.5 often referred to as gazelles, are the main drivers of employment creation. That is, a Value added (log) 4.0 handful of firms experience high employment growth while most other fi rms hardly grow 3.5 at all (see Bottazzi et al. 2007). This small group of high-growth firms experiences a 3.0 period of accelerated growth that levels off 2 4 6 8 over time, leading to a “tent-shaped” dis- Size (log) tribution of employment growth rates. This Log of value added per worker pattern has been confirmed in various coun- Log of wages per worker tries, industries, and years, making it a robust Source: Ministry of Industry, Trade, and New Technologies (Morocco). feature of the firm growth process (Coad and Hoelzl 2010).9 Thus, a few rapidly growing firms create most of the new jobs.10 Found in all industries, gazelles are usu- useful to distinguish between product and ally young firms that are more innovative process innovation. Product innovation is and take more risk. There is no evidence that generally found to increase labor demand gazelles cluster in specific industries, but, if and hence firm-level employment growth. anywhere, they tend to be overrepresented Process innovation is associated with pro- in knowledge-intensive service industries ductivity growth, which might reduce the (Henrekson and Johansson 2010), and they demand for labor in the short term. Indeed, tend to rely more on innovation for their the fi ndings for process innovation are less growth (Goedhuys and Sleuwaegen 2009). clear-cut and also indicate job destruction in The few existing high-growth firms in some cases (see, for example, Hall, Lotti, and MENA are more likely to innovate, consis- Mairesse 2008; Harrison et al. 2008). tent with the findings from other regions. Product and process innovation in devel- Stone and Badawy (2011) use World Bank oping countries takes the form of diversifi- Enterprise Surveys to identify firms with cation into new products and the adoption 10 percent or more employment growth in of foreign technologies (or organizational MENA countries. The authors find that these structures), respectively. Both processes firms are more likely to innovate (58 percent have been found to increase the demand for of fast-growing firms reported innovating, skilled labor. Conte and Vivarelli (2010), compared to only 39 percent of slower-grow- Hanson and Harrison (1999), and Fuen- ing fi rms); to offer formal training to their tes and Gilchrist (2005) fi nd that imported employees (32 percent versus 18 percent); to skill-biased technological change through use e-mail and have a website; to have an foreign technology adoption has been international quality certification (for exam- an important determinant of the recent ple, ISO 9000); to have a workforce in which increase in the relative demand for skilled 5 percent or more received a tertiary educa- labor in developing countries. As these fi nd- tion; and to be located in Lebanon, Libya, ings suggest, a dynamic private sector that the Syrian Arab Republic, or the Republic enables foreign technology adoption or of Yemen as opposed to Egypt or Morocco diversification into new products is essential (figure 4.10). to create new jobs that can absorb the large In sum, these fi ndings suggest that young number of university graduates in MENA firms have the highest potential to create new countries. jobs in MENA, in particular for skilled labor. 128 JOBS FOR SHARED PROSPERITY FIGURE 4.10 Employment growth in selected energy-intensive industries. In turn, this countries in MENA, 2007–10 misallocation has discouraged investments in labor-intensive industries and hence has 100 limited job creation. As a result, economies in 90 the Arab world are among the most energy- 80 70 intensive in the world (figure 4.11). 60 Energy-intensive production is associated Percent 50 with large investments in capital. Figure 4.12 40 30 compares the capital intensity in Egyptian 20 and Turkish firms in selected sectors and 10 0 shows how in some manufacturing (textiles and garment, food products, and chemicals), pu ian n . ya p. co ep no Re Lib oc ,R Re Syr ic ba firms in Egypt are more capital intensive than or bl b en ra Le M m t, A firms in Turkey,11 despite Turkey’s higher Ye yp ab Eg Ar ≥ 10% employment growth GDP per capita. 0–10% employment growth Fuel subsidies are disadvantageous not < 0% employment growth only because they repress the demand for labor but also because they suppress incen- Source: Stone and Badawy 2011. tives to innovate, thereby impeding pro- Note: MENA = Middle East and North Africa. ductivity growth, which is a crucial deter- minant of long-run labor demand. They During their expansion, these firms tend not tend to benefit older and publicly owned only to create employment but also to dem- fi rms disproportionately, which in turn are onstrate economic dynamism by adopting likely to use more outdated technologies new technologies and processes or by diver- and consequently to use more energy (see, sifying into new products. However, so far, for example, figure 4.13, which depicts the such firms are rare in MENA economies. kernel density of energy use in Egypt by fi rm age and ownership). In benefiting less efficient fi rms, subsidies not only distort the Explaining the lack of economic demand for labor but also repress market dynamism in MENA forces that would push older and public The following sections document how policy fi rms to innovate and improve energy effi - distortions—including burdensome business ciency to remain competitive. regulation and their inconsistent implementa- Moreover, subsidies are fiscally very costly. tion as well as energy subsidies that distort For the year 2010, they have been estimated relative input prices—depress the demand for to account on average for about 9 percent of labor. MENA’s performance is benchmarked GDP in MENA countries (Silva et al. 2013). against that of dynamic emerging economies Their removal could pay a double dividend in in other regions. employment creation if (some of) the freed-up fiscal space were used to reduce labor taxa- tion, which may affect the demand for labor. Energy subsidies depress the demand for labor A business environment that maintains One reason for MENA’s lackluster perfor- privilege mance in generating jobs is that labor taxes and subsidies on other inputs, such as energy, On average, MENA economies rank aver- increase the relative cost of labor and limit its age in overall business environment—on demand. High energy subsidies in Arab coun- par with China; worse than, say, Turkey tries continue to distort price signals, result- but better than Brazil or Indonesia. A com- ing in a misallocation of investments toward parison of the scores of MENA countries in PRIVILEGES, NOT COMPETITION 129 FIGURE 4.11 Energy use per US$1,000 of GDP in selected countries and regions, 2010 constant 2005 PPP Iraq Bahrain Russian Federation Saudi Arabia Oman Kuwait China Jordan Syrian Arab Republic Indonesia Middle East and North Africa Canada World United Arab Emirates India Qatar United States Latin America and the Caribbean Egypt, Arab Rep. Algeria Norway OECD France Netherlands Brazil Mexico Yemen, Rep. Lebanon Morocco Germany Tunisia United Kingdom Switzerland 0 50 100 150 200 250 300 350 400 Kg of oil equivalent Source: World Development Indicators (database). Note: GDP = gross domestic product; kg = kilogram; PPP = purchasing power parity; OECD = Organisation for Economic Co-operation and Development. Doing Business 2012 (World Bank 2012a) • A second group comprises economies with with those of the rest of the world results a legal business environment that is among in a natural grouping, based on the aver- the most hostile to and restrictive for pri- age ranking across all policy dimensions vate sector development in the world, (see fi gure 4.14, which reports the overall apart from a few bright spots in some rankings of MENA countries and selected policy areas in individual countries. These dynamic emerging economies in 2012):12 economies include Algeria, Djibouti, the Islamic Republic of Iran, Iraq, and, to • The legal business environment in the some degree, also Syria and the West Bank GCC countries (excluding Jordan and and Gaza. For instance, with the excep- the Republic of Yemen) compares well tion of the Islamic Republic of Iran, all of with the de jure regulations in the most these economies have some of the highest advanced countries in the Organisation for regulatory barriers to entry (costs of start- Economic Co-operation and Development ing a business) in the world. As such, they in many policy areas. Nevertheless, most require fi rst-generation legal and regula- GCC countries are among the worst per- tory reforms in most policy areas. formers worldwide in enforcing contracts. 130 JOBS FOR SHARED PROSPERITY FIGURE 4.12 Capital intensity in three manufacturing sectors in the Arab Republic of Egypt and Turkey, 2004–08 a. Textiles and garments b. Food products 0.8 1.0 0.6 0.8 Density Density 0.6 0.4 0.4 0.2 0.2 0 0 0 2 4 6 0 1 2 3 Capital intensity Capital intensity Egypt, Arab Rep. Turkey Egypt, Arab Rep. Turkey Kernel = epanechnikov, bandwidth = 0.2213 Kernel = epanechnikov, bandwidth = 0.1851 c. Chemicals 1.5 1.0 Density 0.5 0 0 1 2 3 Capital intensity Egypt, Arab Rep. Turkey Kernel = epanechnikov, bandwidth = 0.1884 Source: Based on Enterprise Surveys data. Note: Firms with a capital-value added ratio above 6 in textiles and garments are excluded and above 3 in the food and chemical sectors. The sample covers 714 Egyptian and 151 Turkish firms in textiles and garments, 282 Egyptian and 54 Turkish firms in chemicals, and 231 Egyptian and 84 Turkish firms in the food sector. FIGURE 4.13 Energy use of firms in the Arab Republic of Egypt by firm age and ownership, 2004, 2007, and 2008 a. Kernel density estimate b. Kernel density estimate 0.20 0.25 0.15 0.20 Density Density 0.15 0.10 0.10 0.05 0.05 0 0 –5 0 5 10 15 –5 0 5 10 15 Energy use Electricity use Young firms (0–10 years) Private domestic Private Arab Middle-aged firms (11–25 years) Private foreign Government Old firms (26+ years) Other Mixed Kernel = epanechnikov, bandwidth = 0.5094 Kernel = epanechnikov, bandwidth = 0.4465 Source: Abouleinein, El-Laithy, and Kheir-El-Din 2009. PRIVILEGES, NOT COMPETITION 131 FIGURE 4.14 Doing Business rankings for economies in MENA and for selected dynamic emerging economies, 2012 b. Doing Business ranking without access to a. Doing Business overall ranking finance and contract enforcement Average MENA Average MENA rank rank GCC GCC countries countries Chile Chile Intermediate MENA Bulgaria performers Turkey Bulgaria Croatia Turkey China Croatia Intermediate MENA China performers Costa Rica Indonesia Brazil Costa Rica Indonesia Brazil India India Worst MENA Worst MENA performers performers 0 20 40 60 80 100 120 140 0 20 40 60 80 100 120 c. Doing Business ranking d. Doing Business ranking access contract enforcement to bank finance Average MENA Average MENA rank rank China Bulgaria Croatia India Turkey Croatia Chile Chile Bulgaria China Intermediate MENA Turkey performers GCC GCC countries countries Brazil Brazil Worst MENA Costa Rica performers Intermediate MENA Costa Rica performers Indonesia Indonesia Worst MENA India performers 0 20 40 60 80 100 120 140 160 0 20 40 60 80 100 120 140 160 Source: World Bank 2012a. Note: MENA = Middle East and North Africa; GCC = Gulf Cooperation Council. The GCC includes Bahrain, Kuwait, Oman, Qatar, Saudi Arabia, and the United Arab Emirates; intermediate MENA performers include the Arab Republic of Egypt, Jordan, Lebanon, Morocco, Tunisia, and the Republic of Yemen; the worst MENA performers include Algeria, Djibouti, the Islamic Republic of Iran, Iraq, the Syrian Arab Republic, and the West Bank and Gaza. • In a third group of countries—such as emerging economies that have successfully Jordan, Morocco, and Tunisia and, to managed to diversify into higher-value- some extent, also Egypt, Lebanon, and the added activities. All of these countries Republic of Yemen—the business environ- have improved their legal business envi- ment ranks comparably to that of dynamic ronment over the past decade. While legal 132 JOBS FOR SHARED PROSPERITY policy gaps remain urgent in some areas the main shortcomings in MENA countries (for example, access to bank fi nance and with regard to these three factors. contract enforcement), the overall legal Many MENA countries do not have a business environment can only partially fully operational credit bureau. This restricts explain these countries’ disappointing per- banks’ information on potential borrowers, formance in higher-value-added activities. in particular the smaller firms for which the Thus, these countries appear to require a information asymmetry is most pronounced. second-generation type of reform in many Consequently, the collateral required is policy areas to unleash economic dyna- among the highest in the world, which often mism and creative destruction. bars the access of smaller fi rms to credit.15 Often, land is the only asset recognized as Against this backdrop of average overall appropriate collateral for bank loans, while rankings, MENA countries underperform in most MENA countries, large fractions of systematically in two dimensions: access to society (fi rm owners) are de facto excluded bank fi nance and enforcing contracts. This from (access to) land ownership. Keefer is the case even for GCC countries, which (2007) shows that banks in MENA often use otherwise rank relatively well in other policy collateral requirements as a credit-rationing dimensions. Access to long-term external tool instead of as a mechanism for allocating finance is generally considered an impor- credit based on risk analysis. tant factor in promoting riskier investments Figure 4.15 shows that smaller fi rms are in higher-value-added projects as it facili- more credit constrained than larger firms tates risk sharing. The Enterprise Surveys in almost all MENA countries. The gap is show that, on average, 39 percent of firms particularly severe in Egypt, where almost in MENA consider limited access to finance 60 percent of firms with less than 20 employ- a major constraint to their operations. This ees are estimated to be credit constrained (as is the highest share in the world’s regions, compared to 30 percent of larger firms). excluding Sub-Saharan Africa. The restrictive access to credit can to “Enforcing contracts” measures the num- some extent be explained by barriers to ber of official procedures, the time, and the (foreign) entry that reduce competition in costs involved in enforcing a sale-of-goods financial markets and exclude firms from dispute from the moment the plaintiff fi les the lawsuit until the actual payment. Hence, it indicates a problem of implementation rather than legislation. The following sec- FIGURE 4.15 Share of credit-constrained firms tions discuss these points in greater detail. in selected countries in MENA, various years, 2004–08 Restrictive access to credit 70 % of firms that are credit Even countries that have reformed other 60 aspects of their business environment often 50 constrained rank poorly when it comes to access to credit. 40 The World Bank (2011) report on fi nancial 30 access and stability in MENA shows that 20 despite fi nancial reforms in several MENA 10 0 countries in the past, financial systems remain an on co a ria p. undiversified, uncompetitive, and exclusive.13 bi Re oc rd ge an ra Jo or b iA Al b ra Le M The report identifies three major factors ud t, A Sa yp inhibiting access to finance: limited depth of Eg credit information, deficient collateral and 1–19 employees 100+ employees insolvency regimes, and weak bank competi- tion.14 The following discussion summarizes Source: World Bank 2009. PRIVILEGES, NOT COMPETITION 133 financial services. Barriers to entry arise FIGURE 4.17 Percentage of nonperforming loans from widespread state ownership of banks in banks in OECD countries and in various world and a lack of transparency in financial regions, 2007 systems. Moreover, apart from Lebanon, MENA countries have very low shares of 25 foreign ownership in banking compared to 20 other emerging countries (figure 4.16). % of total loans In addition, low transparency in the oper- 15 ations of both enterprises and banks results 10 in high numbers of nonperforming loans and exclusion from access to bank loans in gen- 5 eral. As figure 4.17 shows, MENA’s share 0 of nonperforming loans is the highest in sia D rn be nd l A nd As ca id c d Af and M Pa an th Am EC the world. This symptom is directly linked i hA So sia Ce uro n No le E ic a ste rib a ra a Ea Afr ric E a O d if Ea Ca rica nt pe ia rth ast ut to weak risk-management systems and the e e st prominence of state-owned banks. In almost tin all MENA countries, governments have used La state-owned banks to finance or cover losses Source: World Bank 2009. of state-owned enterprises (or private busi- Note: OECD = Organisation for Economic Co-operation and Development. nessmen with preferential access). Only a few countries have a fully operational, privatized banking sector (for example, Jordan, Leba- discretion over rules—which is reflected in non, and Oman). Thus, reforming public the discretion of state-owned banks in credit banks is an urgent policy priority in many allocation, less-than-independent supervisory MENA countries. institutions, a lack of access to land (collat- High collateral requirements, credit ration- eral), inefficient and unequal enforcement of ing, and other features of financial markets in laws, and a lack of transparency in the bank- MENA suggest that inadequate formal rules ing system—seems to be at the core of the alone cannot explain poor access to fi nance exclusion of many (often smaller) firms from in the region. Instead, the predominance of access to credit. The World Bank (2011) report on fi nan- cial access concludes that banks in MENA FIGURE 4.16 Foreign ownership of banks in have failed to provide access to large seg- selected countries in MENA, 2006 ments of the population and the enterprise sector because they have focused on large, 70 well-connected firms. % of foreign-owned banks 60 The concluding chapter of this report 50 provides an interpretation of the histori- 40 cally high credit rationing in MENA as an 30 important element of the political-economy 20 equilibrium that has undermined private sec- 10 tor development to sustain the interests of a 0 restricted elite. yp Tu on ra sia M ep. Al co Sa Jor a iA n ra Om a n, ep n ic c Lib . ya p am li ri bi ud da Ira b R a Re Isl ub oc ge an t, A ni R ra or b b Le Policy implementation uncertainty nA Eg ria This section provides fresh evidence that Sy Emerging countries, average the discriminatory and uncertain implemen- tation of policies and regulations, rather Source: World Bank 2009. than the legislation itself, distinguishes the 134 JOBS FOR SHARED PROSPERITY region from fast-growing emerging econo- surveys see the implementation of rules and mies. Uncertainty over policy implementa- regulations as inconsistent and unpredictable. tion reduces the expected profitability of Table 4.2 summarizes the averages and higher-productivity investments and, as a dispersion of the number of days that fi rms result, hinders technological upgrading. For had to wait for different regulatory services instance, inconsistent policy implementa- across countries. The survey results confi rm tion creates a de facto business environment that legal business regulations are relatively that is not the same for all firms in the same competitive in Jordan and Morocco along industry. This situation leads to an uneven these four dimensions of regulatory ser- playing field and undermines the competitive- vices. However, in the same countries large ness of firms. Both the lack of competition variations exist across fi rms: the coefficients and the higher uncertainty prevent firms from of variation in waiting times along the four investing in new, riskier, higher-productivity dimensions of regulatory services are higher products or foreign technologies, ultimately than in most other emerging economies, so resulting in a lack of economic dynamism and that, in this sample, Jordan ranks seventh employment creation. and Morocco ninth.16 The resulting discriminatory policy The de facto evidence suggests that waiting implementation of business regulations is time for different regulatory services varies a key concern for entrepreneurs in MENA. substantially across fi rms, and that this pat- Investors in MENA, especially managers of tern is very common throughout MENA.17 smaller enterprises, consistently lament pol- For instance, firms below the 25th percentile icy uncertainty and the favorable position of of the distribution of waiting time in Leba- some incumbent firms at the expense of new non or Syria wait on average between 1 and entrants and competitors. Corruption, anti- 30 days to obtain a construction permit; in competitive practices, and regulatory policy contrast, firms above the 75th percentile wait uncertainty all rank high among the concerns between 190 and 380 days. Similarly, the of business managers in Enterprise Surveys in median firm in the Republic of Yemen waits the region. Figure 4.18 illustrates the degree about 20 days for imports to clear customs, to which managers of firms included in these while the 10th and 90th percentile firms wait between 1 and 55 days.18 This uneven implementation provides fer- FIGURE 4.18 Percentage of survey respondents tile terrain for rent seeking and results in a in selected economies in MENA disagreeing that large number of firms striking deals to speed interpretations of regulations are consistent and up the implementation of legal procedures.19 predictable, 2005–08 For example, based on pooled fi rm sur- vey data, table 4.3 shows that the waiting time for regulatory services is positively % of respondents disagreeing 80 66.2 66.6 correlated with government officials’ 57.7 60.0 60 requests for bribes. 20 This is consistent with 41.2 42.3 a “hold-up” problem, whereby officials use 40 their discretion to delay regulatory services 21.5 so that they can demand bribes. 20 The discretion in implementing legisla- 0 tion originates from the nature of incen- a . an ria k a occo n ep za tives in the public administration. With bi no rd ge Ga ,R ra ba Jo r iA Al en t B Mo nd recruitment and promotion that have fre- Le ud m Ye Sa an quently been based on regional or sectar- es ian considerations rather than merit and a W weakly institutionalized public administra- Sources: World Bank 2009. Note: Those surveyed were asked whether interpretations of regulations tion, accountability to the broad citizenry is are consistent and predictable. limited. PRIVILEGES, NOT COMPETITION 135 Uncertainty and innovation TABLE 4.2 Averages and dispersion of firms’ waiting days for Finally, international evidence shows that regulatory services in selected developing countries, various years, discriminatory policy implementation 2004–08 reduces economic dynamism (competi- Construction Operating Import Clear customs tion or innovation) and fi rm growth. For permit license license imports example, empirical fi ndings for Jordan sug- Average number of days gest that uncertainty over policy implemen- Indonesia 32 21 11 3 tation reduces (perceived) pressure from Jordan 43 6 5 9 domestic competition (table 4.4). 21 Interest- Morocco 61 4 4 ingly, no correlation emerges in the same Croatia 182 26 12 2 data between implementation uncertainty India 28 29 15 14 and pressure from foreign competition Bulgaria 94 62 21 3 (column 4). Taken together, these fi ndings Turkey 42 37 21 10 suggest that it is unlikely that the observed Chile 143 84 17 17 correlation between policy uncertainty and Brazil 139 83 43 15 domestic competition is due to spurious Coefficient of variation (standard deviation divided by the mean) factors. Bulgaria 1.04 1.59 1.17 1.10 Moreover, firm-level data for both Brazil 1.31 1.14 1.25 1.10 Egypt and Jordan show that the lower the Indonesia 1.93 1.43 0.94 1.09 perceived predictability and implementa- India 1.33 1.40 1.82 1.02 tion of regulation within fi rms of the same Croatia 1.25 1.69 1.27 1.25 size, location, and sector, the lower is the Turkey 1.65 2.88 1.67 1.34 fi rms’ probability of innovation (table 4.4). Morocco 1.72 1.87 1.46 In these regressions, uneven policy imple- Chile 1.94 2.62 1.59 1.39 mentation is proxied by the coeffi cient of Jordan 1.75 2.33 2.14 1.50 variation in firms’ reporting that policy Source: World Bank 2012b; Enterprise Surveys in different years. Note: Survey question: “What was the wait, in days, to obtain a construction permit, operating is predictable. For robustness, results are license, import license, or clear customs for imports?” The dispersions across firms are measured by also tested using the difference between the coefficient of variation (standard deviation divided by the mean). TABLE 4.3 Relationship between delays in policy implementation and demands for bribes in MENA, various years, 2004–08 The determinants of policy implementation times — OLS Time it takes to get an Time it takes to get a operating license (log) construction permit (log) Time to export (log) Time to import (log) (1) (2) (3) (4) (5) (6) (7) (8) Dependent variable coef/se coef/se coef/se coef/se coef/se coef/se coef/se coef/se Gift requested by government official? (1=Yes, 0=No) 0.955*** 0.407*** 0.715*** 0.605*** 0.110** 0.243*** 0.390*** 0.312*** Firm size (ln) 0.270* – 0.157 – 0.034 – 0.158* Firm size^2 (ln) – 0.021 0.014 0.002 0.012 Country dummies No Yes No Yes No Yes No Yes Sector dummies No Yes No Yes No Yes No Yes R2 0.067 0.356 0.061 0.214 0.005 0.076 0.055 0.146 Number of observations 914 845 429 664 670 665 805 802 Source: Hallward-Driemeier, Freund, and Rijkers 2012. Note: MENA = Middle East and North Africa; coef/se = coefficient/standard error; OLS = ordinary least squares. Significance is based on robust standard errors that are clustered by firms’ location-industry-size. *, **, and *** indicate significance at the 10 percent, 5 percent, and 1 percent levels, respectively. Data are from pooled MENA Enterprise Surveys across countries with available data. 136 JOBS FOR SHARED PROSPERITY TABLE 4.4 Effect of policy implementation uncertainty on competition and firm growth in Jordan and the Arab Republic of Egypt, 2006–07 Pressure from Pressure Employment growth domestic competition from foreign (Jordan, 2003–06; Egypt, Dependent variables to reduce costs competition Arab Rep., 2006–07) Probability to innovate Jordan Coefficient of variation of firms’ −1.01* 0.822 −0.647** −0.807 reporting implementation as consistent and predictable Difference 75–25 percentile in firms’ −0.155** −0.072** −0.206* reporting implementation as consistent and predictable R2 0.120 0.122 0.336 0.199 0.197 0.238 0.241 Number of firms 467 467 419 436 436 487 487 Egypt, Arab Rep. Coefficient of variation of firms’ 0.025 0.060 −0.111** −0.787** reporting implementation as consistent and predictable Difference 75–25 percentile in firms’ 0.013 −0.002 −0.090 reporting implementation as consistent and predictable R2 0.031 0.031 0.043 0.420 0.418 0.109 0.106 Number of firms 902 902 899 878 878 905 905 Source: Based on Enterprise Survey data for 2006, World Bank 2012b (Jordan); based on Enterprise Survey data for manufacturing firms in 2007 (Egypt). Note: All regressions include sector dummies (apart from specification 1), heteroscedasticity robust standard errors that are clustered at the group level; * and ** indicate significance at the 10 percent and 5 percent levels, respectively. The average and standard deviations are computed over grouped firms in each location-sector-size group. Innovation is a binary variable equal to 1 if the firm introduced a new product or new process, or licensed a foreign technology in the past three years and 0 otherwise (roughly half the firms in the sample innovated). the 75th and the 25th percentile in fi rms’ fi rms in Egypt in 2007 are consistent with reports. The probability of innovating is the fi ndings for Jordan. 22 then estimated with a probit regression where the dependent variable takes a value of 1 if a fi rm either introduced a new prod- The employment price of privilege uct or a new process or licensed a new The process of creative destruction—which technology within the past three years, is crucial for productivity growth and job and equal to 0 otherwise. About 42 and creation, as evidenced by the recent experi- 50 percent, respectively, of firms in the ences of fast-growing East Asian and Eastern sample in Egypt and Jordan were innova- European economies—is weak in MENA. tors according to this defi nition. Similarly, As a result, the private sector in the region higher uncertainty in policy implementation is small, skewed toward unproductive small- is also negatively correlated with employ- scale activities, and stagnant. The lack ment growth in Jordan: that is, the higher of dynamism is manifested, among other the disagreement (coeffi cient of variation) ways, in lower entry and exit of firms, lower about whether government implementa- growth of incumbent firms, and lower inno- tion is “consistent and predictable” within vation than in relevant comparator countries a location-sector-size fi rm group, the lower in other parts of the world. the employment growth or probability to Entrants and young fi rms have the high- innovate for firms in these groups. The est potential for creating new jobs in MENA, results for the sample of manufacturing in particular for skilled labor. During their PRIVILEGES, NOT COMPETITION 137 expansion, these firms tend not only to create derive an index (EXPY) that ranks traded employment but also to contribute to produc- goods by their implied productivity content. tivity growth by adopting new technologies 3. See Knowledge Assessment Methodology and processes or diversifying into new prod- (2011), at http://www.worldbank.org/kam, on how an innovation score is built. ucts. Moreover, international evidence has 4. The process of creative destruction describes shown that incumbents innovate faster and firm entry, exit, and growth, whereby resources increase productivity more quickly when the are reallocated toward higher-productivity threat of entry is higher. firms. Therefore, the lack of economic dynamism 5. Hsieh and Klenow (forthcoming) observe a in MENA countries comes at a significant similar pattern between young, small manu- cost in employment, in particular for skilled facturing firms in the United States relative labor. At the same time, the lack of creative to India. That is, young manufacturing firms destruction and fi rm entry into higher pro- in the United States grow quickly or exit, ductivity activities constrains the scope for while in India, young firms do not grow. product and process innovations (foreign 6. This section draws substantially on World Bank (2009). technology adoptions), which are the main 7. Earlier profiling of the labor market has drivers of employment creation (for skilled shown that MENA has an extremely large labor) in other countries. It also prevents informal sector and a high prevalence of the exit of inefficient incumbent fi rms that microfirms constituted by self-employed indi- employ resources that could more produc- viduals. A caveat is that the firm census does tively be deployed elsewhere. not cover informal firms, but since these tend MENA’s lackluster record of job cre- to be small, it is arguable that the true firm- ation is due to distortions that depress the size distribution is even more skewed toward demand for labor and limit incentives to small-scale activities. innovate, such as fuel subsidies, a relatively 8. See, for example, Van Reenen (1997), burdensome business environment, and, Blanchflower and Burgess (1998), and Piva and Vivarelli (2004). Moreover, Coad and crucially, inconsistent and arbitrary imple- Hoelzl (2010) and Vivarelli (2012) provide mentation of regulation, which results in extensive literature surveys on the relation an uneven and unpredictable business envi- between innovation and job creation. ronment. Uncertain policy implementation 9. For instance, Geroski and Gugler (2004) dis- and the protection of privileged firms by tinguish between the within- and between- government authorities reduce the expected variance of employment growth and find that profitability of higher productivity invest- most of the variation (around 60 percent) is ments, hinder technology upgrading, and due to changes in employment growth within encourage informality. For many policy individual firms over time. areas, the discrimination in implementing 10. For instance, empirical studies for developed legislation, rather than the legislation itself, countries found that 5 to 10 percent of the firms deliver 50 to 80 percent of employment distinguishes the region from fast-growing creation (see, for example, Acs, Parsons, and emerging economies. Tracy 2008; Coad and Hoelzl 2010). 11. The results are based on Enterprise Surveys from Egypt and Turkey. The former contains Notes observations for manufacturing firms from 1. Several recent contributions provide empiri- 2004, 2007, and 2008, while the latter con- cal evidence that countries that export higher- tains observations from 2005 and 2008. The productivity goods will grow faster; see, for capital intensity is computed as the ratio of example, Hausman, Hwang, and Rodrik the net book value of machinery and equip- (2007); Krishna and Maloney (2011). ment relative to value added (sales divided by 2. This section follows the methodology of the total cost of raw material and intermedi- Hausman, Hwang, and Rodrik (2007), who ate goods). 138 JOBS FOR SHARED PROSPERITY 12. The Doing Business indicators measure the are grouped by their location, sector, and time and costs of official legal procedures for size, for 30 groups of firms (each containing a representative domestic firm based in the at least five firms). For each group, the coef- capital or the largest business center of the ficient of variation of the perceived consis- country. Thus, comparing the business envi- tency of policy implementation across firms is ronment between Brazil, India, and Jordan computed. The spread between the 75th and compares effectively regulations in São Paolo, the 25th percentiles of the perceived policy New Delhi, and Amman. implementation consistency is computed as 13. World Bank (2011) provides a detailed review an alternative measure of policy implemen- on the status of and constraints on financial tation uncertainty. In addition, control vari- development in MENA countries. ables measuring the (initial) size, location, 14. In fact, the Doing Business indicators focus age, exporting status, and the original level on similar de jure aspects of access to credit: of employment of firms are included in the credit information registries, collateral, and regressions. Results for the control variables bankruptcy laws. are consistent with findings in the literature 15. The following paragraphs are based on World on firm growth (not shown in table 4.3). Bank (2009). While significant analog impact on domestic 16. The analysis for MENA countries in table 4.2 competition among manufacturing firms is is restricted to Jordan and Morocco since the found in Jordan, the relationship is not sig- corresponding questions are formulated dif- nificant in the Egypt sample. ferently, and in another context in Enterprise 22. Employment growth is only observed over Surveys of Egypt, Lebanon, and Syria, and one year (2006–07) for most enterprises in the hence cannot be compared straightforwardly. Egypt sample. In fact, waiting times for construction per- mits and for operating and import licenses in Egypt and Syria were substantially and systematically longer and more dispersed References (outliers) than the waiting times observed Abouleinein, S., H. El-Laithy, and H. Kheir-El- in the other countries in table 4.2. In addi- Din. 2009. “Phasing Out Energy Subsidies in tion, only manufacturing firms are observed Egypt.” Working Paper 145, Egyptian Center in the Enterprise Surveys for Egypt, while the for Economic Studies, Cairo. number of responding firms is very small for Acs, Z., J. W. Parsons, and S. Tracy 2008. “High- three of these indicators in Lebanon (between Impact Firms: Gazelles Revisited.” Working 7 and 21 firms). Paper 328, SBA Office of Advocacy, Small 17. Enterprise Survey questions referring to the Business Administration, Washington, DC. waiting time for customs clearance are the Ay ya g a r i , M . , A . D e m i r g u c - Ku nt , a n d only questions referring to regulatory ser- V. Maksimovic. 2011. “Small vs. Young vices that are formulated identically and in Firms across the World: Contribution to the same context in MENA countries beyond Employment, Job Creation, and Growth.” Jordan and Morocco. Policy Research Working Paper 5631, World 18. A similar analysis has been conducted by Bank, Washington, DC. Hallward-Driemeier, Khun-Jush, and Pritchett Bartelsman, E., J. Haltiwanger, and S. Scarpetta. (2010) and Hallward-Driemeier and Pritchett 2004. “Microeconomic Evidence on Creative (2011) for Sub-Saharan African countries. Destruction in Industrial and Developing 19. See Hallward-Driemeier, Freund, and Rijkers Countries.” Discussion Paper TI 2004-114/3, (2012) for more details. Tinbergen Institute, Amsterdam. 20. The results for the control variables in table Berman, E., and S. Machin. 2004. “Skill-Biased 4.4 are not reported as their impact is not sta- Technology Transfer around the World.” tistically significant after the binary variable Oxford Review of Economic Policy 16: 12–22. on requested bribes is included. Note that the Blanchflower, D., and S. M. Burgess. 1998. “New coefficient on bribes remains significant even Technology and Jobs: Comparative Evidence after controlling for firm size, age, location, from a Two-Country Study.” Economics of and ownership. Innovation and New Technology 5: 109–38. 21. A variable proxying for “policy implementa- Bottazzi, G., E. Cefis, G. Dosi, and A. Secchi. tion uncertainty” is built as follows. Firms 2007. “Invariances and Diversities in the PRIVILEGES, NOT COMPETITION 139 Patterns of Industrial Evolution: Some Evidence vs. Young.” NBER Working Paper 1630, from Italian Manufacturing Industries.” Small National Bureau of Economic Research, Business Economics 29: 137–59. Cambridge, MA. Coad, A., and W. Hoelzl. 2010. “Firm Growth: Hanson, G., and A. Harrison. 1999. “Trade and Empirical Analysis.” Paper on Economics Wage Inequality in Mexico.” Industrial and and Evolution 1002, Max Planck Institute of Labor Relations Review 52: 271–88. Economics, Jena, Germany. Harrison, R., J. Jaumandreu, J. Mairesse, and Comtrade (UN Commodity Trade Statistics B. Peters. 2008. “Does Innovation Stimulate Database). United Nations, New York, NY. Employment? A Firm-level Analysis Using http://comtrade.un.org/db/. Comparable Microdata from Four European Conte, A., and M. Vivarelli. 2010. “Imported Countries.” NBER Working Paper 14216, Sk i l l B iased Tech nolog ic a l C ha nge i n National Bureau of Economic Research, D e v e l o p i n g C o u n t r i e s .” D e v e l o p i n g Cambridge, MA. Economies 49: 36–65. Hart, P. E., and N. Oulton. 1996. “The Growth Enterprise Surveys. Various years. World Bank, and Size of Firms.” Economic Journal 106 Washington, DC. http://data.worldbank.org/ (3): 1242–52. data-catalog/enterprise-surveys. Hausman, R., J. Hwang, and D. Rodrik. 2007. Fuentes, O., and S. Gilchrist. 2005. “Trade “What You Export Matters.” Journal of Orientation and Labor Market Evolution: Economic Growth 12: 1–25. Evidence from Chilean Plant-level Data.” In Henrekson, M., and D. Johansson. 2010. Labor Markets and Institutions, ed. J. Restrepo “Gazelles as Job Creators: A Survey and and A. Tokman, vol. 8, 411–35. Santiago: Interpretation of the Evidence.” Small Business Central Bank of Chile. Economics 35 (2): 227–44. Geroski, P. A., and K. P. Gugler. 2004. “Corporate Hsieh, C., and P. Klenow. Forthcoming. “The Growth Convergence in Europe.” Oxford Life Cycle of Plants in India and Mexico.” Economic Papers 56: 597–620. Quarterly Journal of Economics. Goedhuys, M., and L. Sleuwaegen. 2009. “High- Keefer, P. 2007. “Elections, Special Interests and Growth Entrepreneurial Firms in Africa: Financial Crises.” International Organization A Quantile Regression Approach.” Small 61: 607–41. Business Economics 34 (1): 31–51. Klapper, L., and I. Love. 2010. “The Impact of the Hall, B. H. 1987. “The Relationship between Financial Crisis on New Firm Registration.” Firm Size and Firm Growth in the U.S. Policy Research Working Paper 5444, World Manufacturing Sector.” Journal of Industrial Bank, Washington, DC. Economics 35: 583–600. Knowledge Economy Index (database). World Hall, B. H., F. Lotti, and J. Mairesse. 2008. B a n k , Wa s h i n g t o n , D C . h t t p : / /d a t a “Employment, Innovation, and Productivity: .worldbank.org/data-catalog/KEI. Evidence from Italian Microdata.” Industrial Krishna, P., and W. Maloney. 2011. “Export and Corporate Change 17: 813–39. Quality Dynamics.” Policy Research Working Hallward-Driemeier, M., C . Freund, and Paper 5701, World Bank, Washington, DC. B. Rijkers. 2012. “Predictably Protected.” Mansfield, E . 1962. “Entry, Gibrat’s Law, Manuscript. World Bank, Washington, DC. Innovation, and the Grow th of Firms.” Hallward-Driemeier, M., G. Khun-Jush, and L. A m e r i c a n E c o n o m i c R e v i e w 52 (5 ) : Pritchett. 2010. “Deals versus Rules: Policy 1023–51. Implementation Uncertainty and Why Firms Ministry of Industry, Trade, and New Technologies Hate It.” NBER Working Papers 16001, (Morocco), “Manufacturing Census Data National Bureau of Economic Research, 1996–2006.” Cambridge, MA. Piva, M., and M. Vivarelli. 2004. “The Skill Bias Hallward-Driemeier, M., and L. Pritchett. 2011. in Italy: A First Report.” Economics Bulletin “How Business Is Done and the ‘Doing 15: 1–8. Business’ Indicators: The Investment Climate Repertoire National des Entreprises Tunisien, When Firms Have Climate Control.” Policy Institut National de la Statistique, Tunisia. Research Working Paper 5563, World Bank, http://www.tunisieindex.com /entreprises/ Washington, DC. Institut-National-Statistique-778.html. Haltiwanger, J. C., R. S. Jarmin, and J. Miranda. Rijkers, B., and H. Arouri. 2012. “Job Creation: 2010. “Who Creates Jobs? Small vs. Large A Big Role for Big Firms.” Blog, World Bank, 140 JOBS FOR SHARED PROSPERITY Washington, DC. http://menablog.worldbank World Bank. 2009. From Privilege to Competition: .org/job-creation-big-role-big-firms. Unlocking Private-Led Growth in the Middle Stone, A., and L . T. Badawy. 2011. “SME East and North Africa. MENA Development Innovators and Gazelles in MENA: Educate, Report. Washington, DC: World Bank. Train, Certify, Compete!” MENA Quick Note ——— . 2011. Financial Access and Stability: A Series 43, World Bank, Washington, DC. Road Map for the Middle East and North Va n Reenen , J. 1997. “E mploy ment a nd Africa. Washington, DC: World Bank. Technological Innovation: Evidence from U.K. ——— . 2012a. Doing Business 2012: Doing Manufacturing Firms.” Journal of Labor Business in a More Transparent World. Economics 15: 255–84. Washington, DC: World Bank. Vivarelli, M. 2012. “Innovation, Employment and ———. 2012b. “Jordan Development Policy Review.” Skills in Advanced and Developing Countries: Manuscript, World Bank, Washington, DC. A Survey of the Literature.” IZA Discussion World Development Indicators (database). World Paper Series 6291, Institute for the Study of Bank, Washington, DC. http://data.worldbank Labour (IZA), Bonn, Germany. .org/data-catalog/world-development-indicators. MENA’s Institutional Framework: Reinforcing the Status Quo 5 Main findings • In MENA, institutional arrangements that regulate wages and contracts reinforce segmentation between labor market insiders (that is, the formally employed) and outsiders. • In some countries, collective wage agreements or minimum wages tend to exclude job seekers whose productivity does not meet the wage floor, while high taxes can exacerbate their effect. • On average, public sector employment offers more comfortable contractual terms than the private sector, often generating a de facto wage floor that induces eligible candidates to queue for the former. • The hiring options for private employers are relatively flexible, but dismissal is administratively more difficult in MENA than in other regions. • Firms can afford to hire along gender, age, or socioeconomic segments because, historically, competition has been low, while citizens’ limited “voice” has also placed little pressure on firms to perform. • Segmentation probably cannot be sustained at its current level, because global competition in product markets is increasing and because pension systems require a broader base of contributors to sustain themselves. Insiders and outsiders in MENA’s men) are stable in their positions and more labor markets likely to have access to social security, while newcomers (young, first-time job seekers) C hapters 1 and 2 described signs of and women usually fare worse. If they have segmentation in labor markets in a job at all, it is likely to be in the informal the Middle East and North Africa sector. (MENA) region: men participate more, are A situation in which incumbent work- employed more, and earn more than women; ers enjoy more favorable employment the same is true for prime-age adults com- opportunities than newcomers is called an pared to youth. In other words, workers who “insider-outsider” labor market.1 Chapters already have a job (typically middle-aged 1 and 2 pointed out the inefficiencies that 141 142 JOBS FOR SHARED PROSPERITY characterize labor markets in MENA, many In the labor market, fi rms’ quantitative of which are linked to segmentations: differ- and qualitative demands and people’s quan- ent segments—young men, young women, titative and qualitative supply meet. People prime-age men, prime-age women—emerge enter employment, which is defined by a as separate pools of workers whose mobility contract (specifying the terms of hire, dis- is limited and who may be excluded from the missal, and hours) and a wage. If no further labor market altogether. institutional or market failures interfere, This chapter identifies the mechanisms the wage resulting from that interaction that place some groups “inside” and others clears the market: in other words, exactly “outside” MENA’s labor markets. It inves- everyone who wants a job at that market tigates why this segmentation—and source wage gets one. of aggregate inefficiency—has persisted for a relatively long time in MENA, discusses Practice why it may no longer be sustainable, and reflects on alternatives that might remedy the In practice, the market can experience fail- situation. ures. People naturally know more about their own skills and effort than an employer does when fi rst meeting them. This “information How does segmentation asymmetry” makes it hard for an employer between insiders and outsiders to estimate the productivity of a prospective arise? employee. “Quality signals” like a good edu- cation or (better) previous similar work expe- Theory rience with good references can help bridge To understand how segmentation arises, the information gap. Obviously, the gap is we must understand the theory of how harder to bridge for young people without employment is generated when labor mar- work experience. kets are not segmented. 2 In such markets, Various institutions can intervene in mar- firms continuously demand labor and kets. Some are needed to reduce the risk to skills, either because of normal staff turn- individuals participating in the market. Social over or because of new investments. Firms insurance, for example, protects individuals try to hire the best people they can get, as against the risks of poverty, unemployment, long as the new employees’ productivity poor health, and old age, leaving them free to is above or equal to the market wage. In take more and more productive risks. Other a competitive market, the market wage is institutions, like taxes, are vital to financ- fi xed for the individual fi rm and cannot be ing the goods that the private sector will not influenced by it. Thus to make a profi t, a supply sufficiently (like roads and schools). fi rm tries to hire people who produce more Social insurance and taxes are often financed than they cost (the wage). through the payroll (wages), which effectively Labor and skills are supplied by people drives a wedge between the market-clearing who have acquired skills in their former jobs wage and the prevalent wage, creating a gap or through education. They seek a work-life between labor supply and demand. Some- mix in line with their acquired skills and thing similar happens if labor demand is inclinations. In their decisions, they trade the restricted through rigid regulation on con- “cost” of giving up an additional hour of lei- tracts. Regulations affecting contracts and sure against the reward from the wage earned the wage that is effectively paid will always for that hour of work. They give up leisure as affect the employment outcome of the labor long as the marginal utility lost is below the market. marginal utility gained from the wage. The W hen i nst it ut ions a f fec t d i f ferent wage is fi xed in a competitive market, and groups of the population differently, seg- the individual cannot influence it. mentation arises. Segmentation can occur MENA’S INSTITUTIONAL FRAME WORK : REINFORCING THE STATUS QUO 143 unintentionally, for example, as the regula- Wages and nonwage costs tory environment changes over time. In that If the prevailing labor law precludes adjust- case, some regulations affect only younger ing wages downward (that is, establishes people and newcomers to the labor market, wage floors), formal employment will be such as women. Segmentation can also occur unlikely for the people whose productivity when a regulation has a learning curve, in is below the floor. While minimum wages which case the first generation of people sub- that are carefully set have little consequence ject to the regulation are less affected by the for aggregate employment (see Machin, regulation than later generations. Manning, and Rahman 2003), higher wage Many of the above-mentioned institu- floors—for example, those resulting from tions are vital and cannot be done away high public sector wages or collective wage with. They must be designed carefully, agreements negotiated by insiders—can have however, to balance the intended benefits substantial negative impact. with the unintended effects on the employ- ment outcome of the labor market (World Bank 2013). The sections that follow exam- Effects of minimum wages ine in considerable detail how labor market Minimum wages in MENA cover a broad institutions in MENA relate to employment range and tend toward the higher end, from outcomes, especially segmentation. The dis- 10 percent of the average value added per cussion begins with the two basic elements worker in the Arab Republic of Egypt to of any formal employment—wages and con- nearly 70 percent in Morocco, with an aver- tracts. The next sections review the process age of 42 percent. Many MENA countries lie for establishing wages and contracts in the above typical comparator countries in Eastern private sector, including its possible inflexi- Europe and Latin America (figure 5.1). bilities, and compare it to the process in the However, while minimum wages appear public sector. The fi nal section explores the to influence the overall wage distribution role of social insurance regulation in labor in MENA, they are rarely enforced strictly. market outcomes. In Egypt and Jordan, a measurable share of FIGURE 5.1 Minimum wage as a share of value added per worker in selected countries in MENA and in selected comparator countries, 2010 a. Middle East and North Africa b. World Egypt, Arab Rep. 0.10 Oman 0.21 Tunisia 0.24 Iraq 0.28 Lebanon 0.29 Jordan 0.34 Chile 0.00 Algeria 0.38 Bulgaria 0.22 Average non-GCC 0.42 Spain 0.27 countries Yemen, Rep. 0.46 Greece 0.29 Iran, Islamic Rep. 0.57 Latvia 0.29 Syrian Arab 0.58 Colombia 0.37 Republic Morocco 0.69 Turkey 0.47 0 0.2 0.4 0.6 0.8 0 0.2 0.4 0.6 0.8 Source: World Bank 2012. Note: Preferred comparators for MENA usually include the Mediterranean rim and Eastern Europe. MENA = Middle East and North Africa; GCC = Gulf Cooperation Council. 144 JOBS FOR SHARED PROSPERITY FIGURE 5.2 Hourly wage distribution and minimum wage in the Arab Republic of Egypt and Jordan, 2010 a. Egypt, Arab Rep. b. Jordan 1.0 1.5 0.8 1.0 0.6 Density Density 0.4 0.5 0.2 0 0 2 4 6 8 10 2 4 6 8 10 12 Log, monthly wage Log, monthly wage Informal sector Formal sector Informal sector Formal sector Sources: The Arab Republic of Egypt’s Labor Force Survey (LFS) 2010; Jordan’s Labor Market Panel Survey (LMPS) 2010. See the appendix for more information on these surveys. Note: Dotted line represents the mandated minimum wage. wages falls below the current official mini- bargaining on wages between worker rep- mum wage (figure 5.2). This phenomenon is resentations (trade unions) and employer problematic, because employers who break representations. Often, a CWA is a matrix the minimum wage rule are bound to break with two dimensions: worker category further rules attached to formal employment, (competence, responsibility level, or educa- such as affiliation to social insurance cover- tional level, or a combination of these) and age and respect of work contracts. If mini- seniority in years. Each field for a particu- mum wages are set so high that they force lar type of worker of a particular senior- some employers to circumvent them, they ity contains a wage range from minimum also force people into informality.3 to maximum or only a minimum. In sev- Potentially, small firms could suffer the eral countries, these agreements have a sec- most harmful impact of minimum wages. torally or nationally binding character. In Larger firms, which are well represented other words, all employers in a given sector in employer associations, can influence the or in the entire country need to abide by the national or sectoral setting of minimum agreed wage ranges, and workers within the wages and thereby set them high enough to same category and the same seniority need eliminate smaller competitors who cannot to fall within the same wage range. The afford to pay the wage. An example of this countries in MENA for which there is evi- is Germany’s (liberalized) postal sector case dence create these matrices through CWAs in early 2008, when a minimum wage was (rather than setting them statutorily by introduced with strong support from the central government), with the CWA reach- incumbent Deutsche Post. The minimum ing the rank of labor legislation. Another wage was valid only for four months (after signifi cant difference between MENA and which a court ruling eliminated it), but dur- Northern and Eastern Europe, for example, ing that time 27, smaller providers left the is that the worker category in the wage market, eliminating 5,693 jobs. matrix depends strongly or solely on the educational degree, whereas in Europe, it Effects of collective wage agreements and depends on competency as refl ected in the bargaining on the labor market actual work and responsibilities carried The region also has several collective wage out, largely irrespective of formal educa- agreements (CWAs). A CWA is the result of tional level. MENA’S INSTITUTIONAL FRAME WORK : REINFORCING THE STATUS QUO 145 BOX 5.1 The conventions collectives in Tunisia In around 70 sectors in Tunisia, centralized employer The CWAs work by levels, where each level and employee representatives agree on a pay scale with ref lects the education, competence, and sof t wage floors for different professional levels. Country- skills in a mix that is ultimately the discretion of wide, the overarching Convention Collective Cadre the employer. A university education is usually between the UTICA (employer representation) and the required for the four or five highest levels. That UGTT (general trade union) determines the default. is, there are de facto wage floors for vocational Individual companies can deviate if needed, within graduates and higher graduates according to their a regulated negotiation ritual, but only in agreement degree. The differences are quite remarkable in with their worker representatives. The default situa- most sectors, ranging from 200- to 400-dinar dif- tion is always the convention, and if employer and ferences in starting salaries. “Downgrading” is not employees do not reach an outside agreement, the allowed. government arbitrates. The legislation establishing CWAs and the set wage floors by professional level in over enforcement of those laws vary widely across 70 sectors (see box 5.1) (Boughzala 2011). the region. The United Arab Emirates, for Centralized wage setting can be prob- example, has no trade unions (Chartouni lematic; international evidence shows that it 2011); Egypt has unions but no enforced often locks in an exogenous rise in unem- CWAs (Hendy and Selwaness 2011); and ployment, for example, through an oil-price in Algeria, only the public sector experi- or demographic shock. This is because, on ences enforced CWAs, while the private the employee side, wages are monopolisti- sector largely evades them (Musette and cally negotiated by unions whose voting Mohamed-Meziani 2011). Interviews with members are all employed. The unemployed, union officials and employer representa- who would profit from a wage reduction tives in Algeria confi rmed that “most larger against an increase in jobs, cannot vote. enterprises refuse to respect the CWA” and As a consequence, the negotiations rarely that “the CWA[s] are viewed as burdensome result in stagnating or lower wages, even if and therefore not applied …. The enterprises required by the labor market situation (see, are not ready to play along” (Musette and for example, Benassy 1995; Gaertner 1981). Mohamed-Meziani 2011). Jordan has CWAs This situation of insiders (employed, voting for a few private sectors, and wage rules are union members) and outsiders (unemployed, legally enforced; in the words of a local legal nonvoting) is particularly severe if the expert: “For example, banking and insur- unemployed are largely fi rst-time entrants in ance employees managed to get 14th and the job market. 15th salaries [that is, 14 or 15 monthly sala- As in the case of minimum wages, collec- ries in a year], in addition to other fi nancial tive bargaining can also exclude smaller firms incentives, while it is more difficult for the or new market entrants. If the bargaining textiles industry employees to obtain such process is dominated by larger firms, they benefits” (Mryyan 2011). Wage violations are can afford to set the wage floors at levels punishable in Jordan, which sees more than that exclude smaller competitors who achieve 3,000 cases every year. As all infringements fewer economies of scale. need to go to court (labor inspectors have no The organization of collective bargain- authority to resolve them), settlements take ing may matter for its impact. Nickell and time, and many cases are dropped (Mryyan Layard (1997) fi nd that unionization, both 2011). Last, Tunisia has enforced CWAs that in union density (share of union members 146 JOBS FOR SHARED PROSPERITY in the workforce) and collective bargaining 1997), the higher the unemployment rate. coverage (wages affected by the CWA), is The effect is not very large at the average associated with higher unemployment rates wage, but it can be more significant at lower when bargaining is at the sectoral level. levels of the wage distribution. The tax rate Higher (national) and lower (firm-level) can therefore figure in firms’ and workers’ centralization of bargaining do not signifi- decisions on whether to operate in the formal cantly increase unemployment. Later stud- sector, and it can even affect the very basic ies, while confi rming lower unemployment decision of whether to hire someone. with centralized bargaining, do not fi nd a Public sector wage packages can also con- confi rmation of this hump-shaped relation- stitute a wage floor, especially for graduates ship, however; see, for example, Aidt and of institutions of higher education. Fixed Tzannatos (2002) and OECD (2006). national pay scales that are out of sync with Collective agreements in which the profes- the private sector can generate a de facto sional category largely or solely depends on wage floor and induce eligible candidates to the formal educational level, like in various queue. Data from Tunisia show that the pub- MENA countries, set a de facto wage floor lic sector offers a substantial wage premium for university graduates and can contribute over the private sector for all educational lev- to high graduate unemployment. A CWA els but most strongly for secondary and ter- enforced de facto by educational level, as in tiary education (see figure 5.5). At the same Jordan and Tunisia (Mahjoubi 2003), for time, people working in the public sector do example, could explain a large part of gradu- so on much better contractual terms, with ate unemployment and inactivity. On the nearly 90 percent enjoying an open-ended one hand, employers do not absorb an ever- contract, compared to around 30 percent growing graduate population at the wages in the private sector. In Egypt, Jordan, the foreseen for graduates. On the other hand, United Arab Emirates, and the West Bank graduates who can afford to wait have an and Gaza, the public sector wage package for incentive to “queue” until a formal job, with a CWA salary, becomes available. Those who FIGURE 5.3 Labor taxes and contributions to cannot afford to wait will accept lower cat- social insurance as a percentage of the base salary egories and wages informally. in world regions and the OECD, 2011 Effect of taxes and public sector wage packages on unemployment OECD High taxes and contributions can exacerbate Eastern Europe and Central Asia wage floors. MENA countries, especially out- MENA (non-GCC side the Persian Gulf, have a relatively high countries) share of labor taxes, ranking only below the MENA (GCC countries) countries in the Organisation for Economic Latin America and the Caribbean Co-operation and Development (OECD) and Eastern Europe (see figures 5.3 and 5.4). Sub-Saharan Africa However, the social insurance and public East Asia and Pacific services provided in MENA countries are South Asia inferior to the average of the OECD, and 0 5 10 15 20 25 30 the effi ciency of spending these resources % of base salary warrants additional study. Empirically, taxes can significantly Source: World Bank 2012. Note: Calculations on labor taxes include social contributions and increase unemployment. The higher the labor taxes paid by the employer (see http://www.doingbusiness.org/ payroll tax level, whether in the tax wedge methodology/paying-taxes) and are not the classical “tax wedge,” which also takes into account taxes paid by the employee. OECD = Organisation (Elmeskov, Martin, and Scarpetta 1998) for Economic Co-operation and Development; MENA = Middle East and or the overall tax rate (Nickell and Layard North Africa; GCC = Gulf Cooperation Council. MENA’S INSTITUTIONAL FRAME WORK : REINFORCING THE STATUS QUO 147 the educated also outperforms average pri- Hiring and firing in MENA vate sector packages. The hiring options for private employers Unless public sector jobs are associated are relatively flexible in the region. With the with a much higher risk of being dismissed or exception of Algeria and Djibouti, all econo- downgraded, the public sector’s more generous mies in the region allow fixed-term contracts wage and work package gives job seekers an incentive to prefer and seek public sector jobs. FIGURE 5.4 Labor taxes and contributions to Qualitative and quantitative evidence from social insurance as a percentage of the base salary Egypt and Tunisia confirms that many young in selected economies in MENA, 2011 people queue for public sector jobs: in other words, they remain inactive or unemployed Egypt, Arab Rep. 27.1 until an opportunity to join the public sector Iran, Islamic Rep. 25.9 arises (Assaad 1992; Grun et al. 2008; Youssef Tunisia 25.2 2004). In this way, the public sector makes it Lebanon 24.1 difficult for the private sector to recruit. Morocco 22.7 As a large share of employees in MENA Syrian Arab Republic 19.3 work in the public sector (table 5.1), the Djibouti 17.7 incentives it sets are important. More than Bahrain 14.7 half of Iraqi nationals and nationals from the United Arab Emirates 14.1 United Arab Emirates work in the public sec- Iraq 13.5 tor, compared to about one-third in Jordan Jordan 12.4 and one-fourth in Egypt, Tunisia, and the Saudi Arabia 12.4 West Bank and Gaza. Oman 11.8 Yemen, Rep. 11.3 Qatar 11.3 Contracts Kuwait 10.7 West Bank and Gaza 0 The basic elements of a work contract, apart 0 5 10 15 20 25 30 from the wage, are its duration (temporary, % of base salary fi xed term, or open ended), the work hours (including annual leave), and the dismissal Source: World Bank 2012. procedures, including any severance pay. Note: MENA = Middle East and North Africa. FIGURE 5.5 Wages and job protection by employment sector in Tunisia, 2011 a. Public sector wage premium b. Type of contract by sector by education level 60 55 100 50 90 80 40 70 32 Percent 60 Percent 30 50 20 11 40 10 30 –5 20 0 10 –10 0 No education Primary Secondary Tertiary Private sector Public sector Open-ended contract Fixed-term contract No contract Source: World Bank and Institut National de la Statistique (forthcoming), using Tunisia’s LFS 2011. See the appendix for more information on the survey. 148 JOBS FOR SHARED PROSPERITY TABLE 5.1 Share of employment by sector in selected economies and have prescribed redundancy (as well as in MENA for the latest available year, 2005–10 reemployment priorities) for certain groups Percent of workers. All these rules also mean that Private Private challenging dismissal decisions legally is Economy Public sector formal sector informal sector comparatively easy for formal workers. Egypt, Arab Rep. 26.6 17.5 55.9 All things considered, the legislation in Iraq 54.0 0.8 45.2 MENA’s non-Gulf countries imposes more Jordan 33.5 22.0 44.6 procedural restrictions for dismissal than Morocco 11.4 13.4 75.2 anywhere else in the world. Within the Tunisia 22.0 15.2 62.8 region, Egpyt, Jordan, and Tunisia are the West Bank and Gaza 21.0 79.0 — most restrictive. Several empirical studies United Arab Emirates (all) 30.0 70.0 — show that difficult dismissal can discourage United Arab Emirates work effort and increase absenteeism (see, (nationals) 91.7 8.3 — for example, Jimeno and Toharia 1996). United Arab Emirates (expatriates) 21.6 78.4 — Egypt enacted a new labor law in 2003 giv- Yemen, Rep. 8.8 11.9 79.3 ing employers more flexibility in aligning the labor force with economic conditions. The law Source: The Arab Republic of Egypt’s LFS 2010, Iraq’s Household Socioeconomic Survey (HSES) 2006–07, Jordan’s LMPS 2010, Morocco’s LFS 2009, Tunisia’s LFS 2010, the United Arab Emirates’ LFS introduced fi xed-term contracts (that can be 2009, the West Bank and Gaza’s LFS 2008, and the Republic of Yemen’s Household Budget Survey extended indefinitely), relaxed dismissal laws 2005–06. See the appendix for more information on these surveys. Note: — = not available. to an extent, gave workers the right to strike, and passed rules to address collective bargain- ing and worker-employer disputes. Yet the cost for permanent tasks. A fi xed-term contract of firing a redundant worker in Egypt remains can last a maximum of 12 months (Morocco), one of the highest in the world (figure 5.6). 24 months (Djibouti, Lebanon, and the West Although labor regulation is not identified as Bank and Gaza), 48 months (Tunisia), or a leading constraint to business there (World 60 months (the Syrian Arab Republic) and has Bank 2009), labor regulations and mandatory no limit in all other countries in the region.4 contributions continue to prevent many enter- Morocco reformed its labor code in 2003, prises from expanding formal employment. allowing firms to hire under temporary con- According to reports from hotels, service tracts. Dismissal regulations remain restric- fi rms, and manufacturing firms, they would tive, however: Morocco has one of the most hire a net of 15, 9, and 21 percent more work- generous severance pay systems in the world. ers, respectively, if there were no restrictions In combination with flexible hiring, restric- on hiring and firing workers. tive firing provides incentives to game the Severance pay is rather low overall in system—for example, by hiring on temporary MENA, but the low average masks wide contracts only—which also leaves workers divergence across countries. In addition to rel- without access to adequate income protection. atively strong employment protection through Dismissal (from the formal sector) is more difficult dismissals, Egypt also requires a difficult in MENA than in other regions severance pay that exceeds the averages of of the world. Most MENA countries (all all other world regions. Legislation in the except most Gulf countries, Lebanon, and Islamic Republic of Iran, the West Bank and Morocco) require the employer to notify Gaza, and the Republic of Yemen also results a third party for the dismissal of just one in relatively high costs of redundancy. worker, and Egypt, the Islamic Republic of Iran, Jordan, Syria, and Tunisia also require Pervasive and distortionary public sector the approval of the third party—usually the privileges labor inspection or a committee chaired by Public sector employment offers more com- it. Morocco and Tunisia require employees to fortable terms than the private sector, on be retrained or reassigned before redundancy average (table 5.2). In Egypt, Jordan, Tunisia, MENA’S INSTITUTIONAL FRAME WORK : REINFORCING THE STATUS QUO 149 FIGURE 5.6 Cost of redundancy in selected regions in the world and in selected economies in MENA, 2011 a. World b. Middle East and North Africa Egypt, Arab Rep. 26.7 South Asia 19.4 West Bank and Gaza 23.1 Iran, Islamic Rep. 23.1 Sub-Saharan Africa 14.9 Yemen, Rep. 23.1 Latin America and United Arab Emirates 18.1 14.6 Qatar 16 the Caribbean Saudi Arabia 15.2 East Asia and Pacific 12.6 Kuwait 15.1 Morocco 13.5 Middle East and North Africa Algeria 13 10.9 (non-GCC countries) Tunisia 7.8 GCC countries 10.7 Syrian Arab Republic 0 Djibouti 0 Eastern Europe and Lebanon 0 10.7 Central Asia Jordan 0 Oman 0 OECD 6.8 Iraq 0 0 5 10 15 20 25 0 5 10 15 20 25 30 Weeks of salary Weeks of salary Source: World Bank 2012. Note: Severance pay for redundancy dismissal (average for workers with 1, 5, and 10 years of tenure, in salary weeks). MENA = Middle East and North Africa; GCC = Gulf Cooperation Council; OECD = Organisation for Economic Co-operation and Development. TABLE 5.2 Standard open-ended contract packages in the public and the private sectors in selected countries in MENA, 2011 Public sector Country Work hours/week Paid leave in days/year % of workers dismissed/year Severance pay Algeria 40 22 <1% Right to retire after 10 yrs Egypt, 36 From 14 for new hires to <1% Unused annual leave plus Arab Rep. 45 for over 50 yrs old compensation Jordan 38–42 21–30 <=3 cases/yr 0 Tunisia Officially as in Officially as in private <0.005% One month pay/yr of service, private max 12 yrs United Arab Emirates 35–40 15–30, plus leave for social Extremely rare for nationals Pension for nationals; expats: (nationals) activities, 2 mo health travel one month pay/yr for first 5 yrs, leave, 6 mo study leave 1.5 for next 5, 2 for next 5 Private sector Algeria 40 22 — 13 Egypt, 48 21; 30 for over 50 yrs 4% 26.7 Arab Rep. old/after 10 yrs Jordan 40–53 14–21 9% 0 Tunisia As in public; extra As in public 6% (only 1% if open-ended) 7.8 hrs expected United Arab Emirates 48 30 — 18.1 (nationals) Sources: Based on legislation analysis and expert interviews by Musette and Mohamed-Meziani (2011; Algeria), Hendy and Selwaness (2011; the Arab Republic of Egypt), Mryyan (2011; Jordan), Boughzala (2011; Tunisia), and Chartouni (2011; United Arab Emirates) World Bank 2012. Note that we have no further information on the specific severance pay for civil servants in Algeria and the Arab Republic of Egypt. Note: — = not available; MENA = Middle East and North Africa. a. Average for workers with 1, 5, and 10 years of tenure, in salary weeks (World Bank 2012). and the United Arab Emirates, the number of public sector in Egypt, Jordan, and the United official or expected work hours is higher in Arab Emirates. Dismissals in the public sec- the private sector than in the public sector. tor are extremely rare in all countries for Annual leave for the two sectors is the same which data are available, while the dismissal in Algeria and Tunisia, but it is greater in the rate reaches 9 percent for the private sector in 150 JOBS FOR SHARED PROSPERITY some countries. Severance arrangements are employment protection between France and invariably more generous in the public sector. Spain explain why France’s unemployment Some public staff “double dip” and pursue rate did not rise nearly as much as Spain’s employment on the side. There is evidence during the recession. Based on analysis of a from Egypt showing that public sector health panel of 60 developing and developed coun- staff hold private clinics during public sector tries, Caballero et al. (2004) show that strict shifts and on public premises. One-quarter labor regulations, if enforced, slow down the (25 percent) of all public personnel in public speed of adjustment to shocks. facilities are absent on an average day (Grun, Etter, and Jillson 2010). Effect of strict employment protection on productivity Employment legislation and job creation Restrictive legislation can sometimes in MENA hurt productivity by limiting the move- Legislation on contracts is important, ment of labor. Caballero and Hammour because, according to international evi- (2000) show that rigid dismissal legisla- dence, strict legislation to protect employ- tion reduces formal sector employment and ment, on balance, tends to reduce aggregate contributes to the growth of the informal employment and to increase unemployment. sector; depresses job creation and tech- For example, Heckman and Pages (2000) nological innovations; enables survival show that in Latin America, more strin- of ineffi cient, low-productivity units; and gent job security laws are associated with suppresses entry of potentially highly pro- lower employment and higher unemploy- ductive new fi rms. Several new empirical ment, particularly among young workers. studies support these findings. Bassanini Ahsan and Pages (2007) report that regula- and Venn (2007) fi nd that strict statutory tions concerned with labor disputes and job employment protection for regular con- security hurt covered workers. Other studies tracts reduces productivity growth, most using macroeconomic data have also found likely by restricting the movement of labor negative efficiency effects of high sever- into emerging high-productivity activities. ance pay (see Nicoletti et al. 2001). Nickell Eslava et al. (2010) examine the impact of and Layard (1997) fi nd that strong employ- labor and capital reforms on productivity ment protection slows turnover and there- in Colombia, showing that increased labor fore short-term unemployment but increases and capital turnover after the reforms in long-term unemployment. The OECD (1999) the early 1990s increased productivity. fi nds that stricter protection leads to lower Further studies show that restrictive legis- employment-to-population ratios through lation slows down the speed of labor real- higher inactivity rather than to higher unem- location from low- to high-productivity ployment. Micco and Pages (2006) also fi nd activities (Besley and Burgess 2004; Boeri a negative employment effect. The evidence and Jimeno 2005; Haltiwanger, Scarpetta, is not unequivocal, however. Some recent and Schweiger 2010; Messina and Vallanti studies (Baccaro and Rei 2005; Bassanini 2007; Micco and Pages 2006). However, and Duvall 2006) do not find significant rigid dismissal has also occasionally been employment impacts. One reason may be shown to increase productivity, through that enforcement matters. Ahsan and Pages higher training commitments by employ- (2009) and Almeida and Carneiro (2009) ers (who cannot get rid of people) or higher show that negative effects depend on the capital investments for the same reason strictness of enforcement. (Autor, Kerr, and Kugler 2007; Belot, Strict employment legislation also tends to Boone, and van Ours 2007). increase a country’s vulnerability to shocks. Labor turnover tends to be relatively Bentolila, Dolado, and Jimeno (2011) esti- low in the region (table 5.3) but is compa- mate that differences in the enforcement of rable to that in Eastern Europe. With the MENA’S INSTITUTIONAL FRAME WORK : REINFORCING THE STATUS QUO 151 TABLE 5.3 Job turnover in the past year as a share workers, because they increase the costs asso- of those currently working in the private sector in ciated with hiring a bad worker. selected economies in MENA, 2006–10 In other words, large dismissal costs con- Percent tribute to the segmentation of labor markets, People starting People losing in which well-protected workers in the for- Economy new jobs jobs mal sector (who tend to be predominantly Egypt, Arab Rep. 0.6 4.0 prime-age males) contrast with much less- Iraq 2.0 1.2 protected workers in the informal sector and Jordan 5.2 8.8 the unemployed. Evidence from the region, West Bank and Gaza 15.0 27.0 reviewed in the sections that follow, shows Tunisiaa 10.5 6.3a these effects among youth and women. Sources: The Arab Republic of Egypt’s LMPS 2006, Iraq’s HSES 2006–07, Consistent with the above observations, Jordan’s LMPS 2010, the West Bank and Gaza’s LFS 2008, and Tunisia’s LFS youth have less stable labor contracts than 2010 (see Appendix); expert interviews by Hendy and Selwaness (2011), Mryyan (2011), and Boughzala (2011). prime-age workers in MENA. In Egypt, Note: MENA = Middle East and North Africa. Tunisia, the United Arab Republic, and a. Formal layoff from open-ended contract only. the West Bank and Gaza, youth have sig- nifi cantly fewer open-ended contracts and more term or sometimes even noncontrac- extreme exception of the West Bank and tual work relationships (see figure 5.7) Gaza, job entry ranges from 2 to 10 percent (although in Jordan, as an exception, youth of all jobs per year, while layoffs range from have more stable contracts than prime-age 1 percent to 9 percent of total employment workers). Temporary jobs can be problem- per year. Comparable figures for OECD atic, because they can exacerbate any dual- countries are generally in the 10 percent ism in the labor market (Bentolila, Dolado, range, and available figures for Eastern and Jimeno 2011). The Spanish experience Europe range from 2 percent to 16 percent, shows how temporary contracts become a with most countries above 8 percent. “permanent” perpetuation of insecurity—a Effect of strict legislation on youth and first-time job seekers FIGURE 5.7 Types of work contracts in all sectors Strict employment protection legislation in the Arab Republic of Egypt, Jordan, Tunisia, and primarily hurts access to jobs by vulnerable the West Bank and Gaza for the last available year, groups such as youth and women, thereby 2006–10 leading to more exclusion. Blanchard (1998) shows that an increase in dismissal costs 100 leads to higher unemployment among mar- 90 80 46 37 44 46 49 ginal groups of workers because of their infe- 70 69 55 53 60 Percent rior access to jobs. Because the productivity 50 1 10 7 19 7 6 of young workers is not easily revealed before 40 3 30 3 hiring, their probability of being hired is 20 53 51 48 45 37 41 27 35 lower in the presence of increased firing costs. 10 0 OECD (1999) finds that stricter employment Youth Prime Youth Prime Youth Prime Youth Prime protection legislation reduces employment age age age age among prime-age women and youth, thus Egypt, Jordan Tunisia West Bank Arab Rep. and Gaza exacerbating exclusion effects, as difficult Open-ended contract Fixed-term contract firing leads to longer unemployment spells No contract and job duration, a trend that likely benefits prime-age males. Similarly, Kugler and Saint- Source: The Arab Republic of Egypt’s LMPS 2006, Jordan’s LMPS 2010, Paul (2000) show that greater firing costs Tunisia’s LFS 2010, and the West Bank and Gaza’s LFS 2008. See the increase discrimination against unemployed appendix for more information on these surveys. 152 JOBS FOR SHARED PROSPERITY trap rather than a bridge to an open-ended first-time job seekers in Tunisia search 9.0 job. Temporary work spells are unlikely to months longer than the average unemployed: end in permanent jobs regardless of work- in Jordan, nearly 4.0 months longer, in ers’ tenure. These results can probably be Egypt, 3.0 months longer, and in the West generalized to all dual labor markets, such Bank and Gaza, 2.6 months longer. Only in as those in MENA. Lebanon do first-timers find a job faster than Interestingly, the relative disadvantage of the average (by about a month). youth that is so pronounced in the public sec- tor is much milder in the private sector. In the private sector, youth have only marginally Regulation of social insurance fewer open-ended contracts than older work- The protection against key social risks (old ers, and they have more term contracts. In age, disability, death, unemployment, and most countries, however, older workers have health needs) is linked to labor market par- more noncontractual relationships. This situ- ticipation. While many argue that wide- ation may reflect the institutional differences spread access to protection (specifically, between both sectors, which restrict layoffs the extension of social insurance coverage) (and thereby job turnover) much more in the should be delinked from employment, the public sector. reality in MENA and the rest of the world In Egypt, Jordan, Tunisia, and the West is that access to social insurance tends to Bank and Gaza, fi rst-time job seekers have require formal employment status (or at a much longer unemployment spell than least formal employment for somebody in the average jobless. As figure 5.8 shows, the household). Low coverage of the poor, the young, and FIGURE 5.8 Duration of unemployment for the uneducated male and female first-time job seekers in selected The share of contributors to mandatory economies in MENA, 2006–11 social insurance schemes within the labor force is widely used as a coverage measure. 12 In MENA, social insurance coverage is low: 10 only 33 percent of the labor force contrib- utes to social insurance and thus has access 8 to effective tools of social risk management 6 (figure 5.9a). Rates of coverage in individual Months 4 countries vary a great deal, but even at the highest levels—that is, for countries not in 2 the Gulf Cooperation Council (GCC)—at 0 least a third of the labor force does not par- ticipate in a social insurance scheme (figure –2 5.9). Coverage is strongly related to the level –4 of economic development, as measured by n za n p. a gross domestic product (GDP) per capita (see is i da no Re Ga n r ba Jo Tu ab nd Forteza, Lucchetti, and Pallares-Miralles Le Ar ka t, an 2009), and the problem is not specific to yp tB Eg es MENA. At 6 percentage points below that W Men Women of Latin America and the Caribbean and 11 percentage points above that of East Source: Lebanon’s Employer-Employee Survey 2011, the West Bank and Asia and the Pacific, MENA’s coverage rate Gaza’s LFS 2008, Jordan’s LMPS 2010, the Arab Republic of Egypt’s LMPS reflects its composition of low- and middle- 2006, and Tunisia’s LFS 2010. See the appendix for more information on these surveys. income countries (figure 5.9b). MENA’S INSTITUTIONAL FRAME WORK : REINFORCING THE STATUS QUO 153 FIGURE 5.9 Percentage of the labor force contributing to social security in MENA and other regions of the world, 2000–07 a. Middle East and North Africa b. World regions Libya Egypt, Arab Rep. Tunisia MENA non-GCC Algeria countries Iran, Islamic Rep. Lebanon Developed Jordan Europe and Central Asia Morocco Latin America and West Bank and Gaza the Caribbean East Asia and Pacific Iraq Syrian Arab Republic South Asia Yemen, Rep. Sub-Saharan Africa 0 10 20 30 40 50 60 70 0 20 40 60 80 100 Percent Percent Source: Gatti et al. 2012, based on the International Labour Organization’s Key Indicators of the Labour Market (ILO-KILM). Coverage in GCC countries in the ILO-KILM is essentially limited to Bahrain. Note: MENA = Middle East and North Africa; GCC = Gulf Cooperation Council. The phenomenon of informal employment the Republic of Yemen compared to 72–98 (employment without social insurance) is percent among those with primary education heavily biased toward vulnerable groups: the or below (Gatti et al. 2012). poor, the young, and the uneducated. In gen- eral, the incidence of informal employment Covered insiders and uncovered outsiders decreases with wealth in MENA: informal- A comparison of the patterns of contribu- ity rates within the poorest income quin- tions to social insurance in Jordan with those tiles in Morocco, Lebanon, and Syria are in Chile provides further evidence of a divide 92, 83, and 83 percent, respectively, versus between insiders, who possess higher-quality 73, 59, and 35 percent in the richest quin- jobs with insurance coverage, and outsiders, tiles (Angel-Urdinola and Tanabe 2012). Not who are forced to accept lower-quality jobs surprisingly, the informality phenomenon is without coverage. In Chile, periods of contrib- associated with the notion of the working uting and not contributing to social insurance poor (Gatti et al. 2012) and the young. Youth have similar patterns: 21 percent of contribu- informality rates are especially high for indi- tory spells and 18 percent of noncontribu- viduals up to their mid-20s, reaching 40–70 tory spells last more than three years (figure percent in Egypt, Iraq, Jordan, Morocco, 5.10a and 5.10b, respectively). In the labor and the Republic of Yemen. For individuals market, contributory status signals the exis- older than 25, informality decreases steadily tence of a formal (covered) job, whereas non- in a manner consistent with the phenomenon contributory status may be associated with of queuing for a public sector job (Bodor, informal (uncovered) employment as well Robalino, and Rutkowski 2008). Employ- as with unemployment or (temporary) exit ment without social insurance coverage is also from the labor force. Long contributory and prevalent among those who enter the labor long noncontributory spells are substantially market with a low level of education. Infor- more frequent in Jordan than in Chile: 40 mality among graduates with a tertiary educa- percent of contributory spells and 70 percent tion is in the range of 24–41 percent in Egypt, of noncontributory spells exceed three years Iraq, Jordan, Lebanon, Morocco, Syria, and (figure 5.10c and 5.10d, respectively). This 154 JOBS FOR SHARED PROSPERITY FIGURE 5.10 Months of contributory and noncontributory spells of social insurance for men and women in Chile (2008) and Jordan (2010) Percent a. Chile, contributory period b. Chile, noncontributory period 80 80 70 70 60 60 50 50 Percent 40 40 30 30 20 20 10 10 0 0 <6 6–12 12–24 24–36 >36 <6 6–12 12–24 24–36 >36 Months Months Men Women Men Women c. Jordan, contributory period d. Jordan, noncontributory period 80 80 70 70 60 60 50 50 Percent 40 40 30 30 20 20 10 10 0 0 <6 6–12 12–24 24–36 >36 <6 6–12 12–24 24–36 >36 Months Months Men Women Men Women Source: Administrative data from Social Security Corporation (Jordan) and Pension Superintendency (Chile). “stickiness” is consistent with low mobility current trends point toward the provision between formal and informal jobs and, in a of social insurance under a unified scheme. broader sense, with the existence of a bound- The unification of social insurance provi- ary between covered insiders and uncovered sion, at least for new labor market entrants, outsiders (see Gatti et al. 2012). has occurred in Egypt, Jordan, and Syria. The legislative mandate for unification was Segmentation of coverage between public also approved in Djibouti and Iraq, where and private sector employees the preparatory work for implementation is Social insurance benefits and the rules of in progress. In Saudi Arabia, distinct social participation in the system have historically insurance schemes for private and public sec- been specific to the sector of employment tor workers prevail, but the benefit packages in MENA (Pallares-Miralles, Romero, and are almost identical, with full portability Whitehouse 2012), beginning with the civil arrangements in place for those who change service and only later expanding to the pri- the sector of employment. In the absence of vate sector. This development has generated such arrangements, the social insurance sys- a rather fragmented landscape of systems tem has limited labor mobility and effectively (Palacios and Whitehouse 2006), although segmented the labor market, because vested MENA’S INSTITUTIONAL FRAME WORK : REINFORCING THE STATUS QUO 155 social insurance rights were lost when work- FIGURE 5.11 Life-cycle earning profiles according to sector of ers moved between the private and the pub- employment and the generality of sector-specific pension benefits lic sectors. Strong segmentation of the social in Morocco insurance system is still the norm in the 2.5 Maghreb. For example, Morocco has four Proportion of average earnings distinct mandatory social insurance schemes. 2.0 The Caisse Nationale de Securité Sociale covers private sector workers, the Regime 1.5 Collectif d’Allocations et de Retraites cov- ers workers at state-owned enterprises and 1.0 contractual workers in the private sector, the 0.5 Caisse Marocaine de Retraite provides pen- sion benefits to civil servants and the mili- tary, and the Caisse Nationale d’Organismes 0 21 41 61 81 de Protection Sociale provides health insur- Age (years) ance to public sector workers. Public sector Public sector with private-sector pension Beyond the institutional segmentation lies Private sector Informal sector a great heterogeneity in the social insurance Source: Bodor, Robalino, and Rutkowski 2008. benefits and contribution requirements: not only do public sector workers tend to earn more, but also their social insurance benefits are more generous. percent of their country’s GDP. These con- As figure 5.11 illustrates, higher aver- tingent liabilities are also increasing. One age earnings do not fully explain the higher can calculate these contingent liabilities as defined-benefit pensions for public sector the present value of all future pensions to be workers; the pension of a representative, full- paid to current pensioners plus the present career-path civil servant in Morocco is 15–20 value of all future pensions to be paid to cur- percent higher than his or her pension calcu- rent contributors (active employees), based lated under the rules of the private sector pen- on contributions completed to date. sion scheme. As mentioned, such generous MENA’s pension system benefits are not pension schemes discourage labor from mov- perceived as generous, but if the value of con- ing to private employment and strengthen the tributory streams, the source of funding, and segmentation between the public and the pri- the pension benefit streams are compared, vate employment sectors. then the overall packages are quite gener- ous. The implicit pension debt is increasing, Financial sustainability—or not because any unit of contribution creates more The exclusion from access to social insurance earned pension rights for the contributor runs deeper than just lack of coverage, since than the value of this unit of contribution: the pension systems currently in place redis- that is the core of the pension system sustain- tribute income to insiders at the expense of ability problem. At some point, the unfunded outsiders. Pay-as-you-go (PAYG) is the pre- implicit liabilities must be financed, and dominant mode of financing pension systems there are only two options: defer the burden in MENA. Under PAYG, benefits to current to future plan members (that is, intergenera- pensioners are fi nanced through the contri- tional redistribution) or subsidize pension butions of those currently in formal jobs, but payments from the government’s general rev- PAYG schemes are operating on unsustain- enues. As those with jobs in the formal sector able financial paths. Robalino and Bogomo- are generally in a better income position than lova (2006) show that the pension systems of the rest of the population, the unsustain- a set of MENA countries incurred unfunded able and accelerating accumulation of pen- contingent liabilities in the range of 55–105 sion liabilities points toward solutions that 156 JOBS FOR SHARED PROSPERITY channel general revenues toward those who and they need to leave the country upon are already better off. conclusion of the fi xed-term (often three- year) contract unless their employer seeks Segmentation in the migrant-dominated the renewal of their work permit on their private labor markets of the GCC behalf. As a rule, the sponsorship system Moving abroad for better work opportunities allows no mobility within the host country is widespread in MENA, with the potential labor market. for further growth.5 Remittances received in Recent reforms have started to allevi- Egypt, Morocco, and the Republic of Yemen ate these economically costly restrictions. amount to 5.0–9.5 percent of GDP; the cor- Bahrain formally abolished the sponsor- responding figures are 20.3 percent in Jor- ship system, theoretically allowing firm- dan and 22.0 percent in Lebanon (World to-firm mobility for foreign workers but with Bank 2011).6 The GCC as a region is one of little institutional support thus far. Saudi the world’s key destinations for temporary Arabia introduced the Nitaqat Saudization migrants; the share of the migrant popula- firm classification system, which allows tion there varies between 27.8 percent in foreign workers of fi rms with lower-than- Saudi Arabia and 86.5 percent in Qatar, with required threshold Saudization performance remittance outflows in the range of 6.8–11.3 to move to better-performing fi rms. Within percent of GDP in Bahrain, Oman, and Saudi the so-called free zones of the United Arab Arabia (World Bank 2011). While the GCC Emirates, fi rm-to-fi rm mobility of foreign primarily hosts migrant workers from East workers has been in place for some time. and South Asia, it is also a relevant destina- Social protection is differentiated across tion for workers throughout MENA: more nationals and foreign workers and heavily than 30 percent of the migrants in Kuwait, favors the former. The only standard benefit Qatar, and Saudi Arabia are from other Arab offered to foreign workers is an end-of-service countries.7 cash payment conditional on past wages and Institutional segmentation between the the length of employment with the fi rm, but citizens of a country and migrant workers anecdotal evidence suggests that employers is the international norm: national govern- do not necessarily comply with this mandate. ments tend to protect their domestic labor With the exception of Bahrain’s unemploy- markets in favor of their citizens. The GCC ment insurance provision to foreign workers, labor migration phenomenon is a unique mandatory social insurance coverage in the case. GCC governments do not constrain GCC hardly extends beyond covering work- the access of fi rms to foreign labor; instead, related injury. they establish distinct rules for the employ- On the one hand, the GCC economies ment and social protection of nationals and provide jobs to millions of workers from foreign workers. The employment of GCC low- and middle-income countries where nationals is backed by strong employment income-generating opportunities are worse; protection legislation and the provision on the other hand, though, the employment of social insurance benefits. In addition, conditions of foreign workers in the GCC administrative requirements are in place are often considered exploitative. Although for firms to employ a threshold share of social protection provisions for foreign nationals in their workforce (so-called Bah- workers in the GCC do not meet interna- rainization, Emiratization, Kuwaitization, tional standards, the good news is that the Omanization, Qatarization, and Saudiza- GCC governments have initiated efforts tion policies). In contrast, foreign workers to improve policies affecting the mobil- are bound by the rules of the sponsorship ity and protection of foreign workers. The system. Their legal presence in the host rationale for such initiatives includes cap- country is legitimized only through their turing the productivity gains from using contract with their employer (sponsor), the human capital of migrant workers in MENA’S INSTITUTIONAL FRAME WORK : REINFORCING THE STATUS QUO 157 a better way. An interesting emerging idea FIGURE 5.12 Age-specific composition of the labor force across is the use of “mobility savings accounts.” pension system contributors, employed noncontributors, and the These accounts, which would replace the unemployed in Jordan, 2010 employer mandate for the end-of-service benefit, would consist of an employee- and 70,000 employer-funded mandatory saving mech- 60,000 anism that would enable such saving to “sponsor” the foreign worker’s job search in 50,000 the host country for a given period after the Labor force initial fi xed-term work contract ends. 40,000 30,000 Demographic shifts and the need for reforming MENA’s social insurance systems 20,000 Threats to the financial sustainability of MENA’s pension schemes are at least 10,000 partially hidden by the current demographic 0 transition and pension system matura- 15 20 25 30 35 40 45 50 55 60 tion. MENA is at a stage where the ratio Age (years) of contributors to beneficiaries is favorable Contributor Employed, noncontributing Unemployed to current fi nances. Although contribution Source: Administrative data from Social Security Corporation (Jordan) and Jordan’s LMPS (2010; see payments to PAYG schemes in MENA gen- appendix for more information). erate more pension entitlements than the value of those contributions, paradoxically, the current fi nancial balances of PAYG pen- the variation of individual contribution pat- sion schemes tend to be in surplus, owing to terns over the active life cycle. The situation the age structure of contributors and pen- resembles that of borrowers who improve sion beneficiaries. For example, Jordan’s their current financial position by borrowing age cohort of people older than 50 who are at an interest rate of 6–14 percent but who approaching (or already partially entering) can afford to service the debt in the long run retirement has fewer than 10,000 contribu- at an interest rate of only 2–4 percent. tors (fi gure 5.12). In contrast, the scheme A further concern is that the active age has more than 30,000 contributors between population in M ENA’s middle-income 22 and 30 years of age. countries is leveling off or even starting to From a fiscal or even a political perspec- decline. Under these circumstances, the dete- tive, it is clear that the positive PAYG bal- riorating financial position of PAYG schemes ance coincides with the scheme’s increasing will become difficult to conceal. These implicit indebtedness. Until the parameters schemes could gain a short-term boost by of the defined-benefit pension systems are the politically popular move of eliminating adjusted to sustainable levels, the higher the boundary between insiders and outsiders current revenues from large, young entry and enrolling current noncontributors (see cohorts actually worsen the system’s long- “noncontributing” employed in figure 5.12), term financial position. The benchmark who would pay large contributions with- range for the long-term sustainable internal out claiming benefits for decades. Without rate of return (IRR)—that is, the implicit reform, however, these gains are likely to be percentage return the pension system pays illusory, because they would raise the cur- on contributions in the form of pension rent fi nancial balance of the pension system payments—for a PAYG pension scheme is while increasing its unfunded liabilities over 2–4 percent. In contrast, the pension system the long term. A more realistic option would IRRs in Egypt and Syria are 6–10 percent be to give outsiders access to social risk- and 7–14 percent, respectively, depending on management tools and at the same time reset 158 JOBS FOR SHARED PROSPERITY the IRRs of the pension systems to sustain- provisions are generous—often there is no able levels (for example, by adjusting pen- reduction in benefits if an individual retires sion system parameters, realigning income before the statutory age (table 5.4). A large replacement rates with contribution rates, number of workers in the formal sector retire and increasing the effective age of retirement when they are in their early 50s and begin to and the length of contributory paths). claim their pension benefits while they con- tinue to be economically active. In Egypt, Effects of flawed pension system designs Jordan, and Syria, one of the most lucrative Because the IRR for participating in pension strategies for participating in the pension schemes in MENA countries is high, even system is for an individual to enroll upon compared to returns from alternative saving entering the labor market and to claim pen- options, those who are allowed to join social sion benefits as soon as he or she meets the insurance systems have strong incentives to requirements for the vesting period and age do so, at least under the current generous of early retirement (Gatti et al. 2012). conditions. By international standards, mini- Such design flaws undermine the financial mum vesting periods are short in MENA’s sustainability of the pension schemes and give pension schemes, and early retirement rise to unfunded government liabilities that TABLE 5.4 Pension design parameters of mandatory formal sector schemes in MENA, 2009–11 Statutory Minimum vesting Vesting for early Life retirement age period Early retirement age retirement age expectancy at birth Female Male Female Male Female Male Female Male Algeria 73.5 55 60 10 15 Any age/45 Any age/50 32/14 32/20 Djibouti 58.5 55 55 — — Any age Any age 25 25 Egypt, Arab Rep. 73.5 60 60 10 10 Any age Any age 20 20 Iran, Islamic Rep. 73.3 55 60 19 19 Any age/45 Any age/50 35/30 35/30 Iraq 70.2 60 60 25 25 55/50 55/50 25/30 25/30 Jordan 73.6 55 60 15 15 50 50 22 25 Libya 75.1 60 65 20 20 — — — — If employer If employer Morocco 72.5 60 60 14 14 55 55 pays pays Syrian Arab Republic 76.1 55 60 15 15 Any age/50 Any age/55 25/20 25/20 Tunisia 74.4 60 60 10 10 50 50 30 30 Country Average wage measure Algeria Maximum of the average of the wages of 5 last years and the average of 5 best years without revalorization Djibouti Average of the wages of the last 10 years/last month’s wage Egypt, Convoluted combination of unrevalorized average basic wage of the last 2 years and unrevalorized full career average variables Arab Rep. wages Iran, Average of unrevalorized wages in the last 2 years Islamic Rep. Iraq Gradually expanding to revalorized full career average wages in public sector/unrevalorized average of last 3 years of wages in private sector Jordan Average of unrevalorized wages in the last 8 years Libya Average of unrevalorized wages in the last 3 years Morocco Average of unrevalorized wages in the last 8 years Syrian Arab Average of last year’s unrevalorized wages with wage increase not to exceed 15% and 30% in the last 2- and 5-year periods Republic respectively Tunisia Average revalorized wages in the last 10 years in main scheme; different rules in 4 other national schemes Source: Gatti et al. 2012. Note: — = not available. MENA’S INSTITUTIONAL FRAME WORK : REINFORCING THE STATUS QUO 159 represent implicit subsidies of formal sector Unemployment insurance: From protecting employment. They also indirectly limit the gov- jobs to protecting income ernment funds available for encouraging volun- In addition to covering only part of the tary entry into the social insurance system—in population, the scope of social insurance other words, to extend the system’s coverage to in MENA is limited; for example, adequate individuals and households outside the formal unemployment coverage is the exception sector that have limited saving capacity. rather than the rule (table 5.5). Lebanon, The other high-return strategy for gaming Libya, Morocco, Syria, the United Arab the pension systems in Egypt and Syria is for Emirates, and the Republic of Yemen have an individual to concentrate the minimum neither unemployment insurance (UI) (which number of contributory years toward the end is a contribution-based tool intended to sup- of his or her career. In this way, the natu- port consumption during spells of unem- rally high earnings at the end of an individ- ployment between jobs) nor unemployment ual’s career are registered (or even artificially assistance (UA) (which is a noncontributory inflated) by the flawed average-wage measure tool for poverty reduction specifically among of the formulas for defi ned pension benefits, the unemployed). As a social insurance which, in this case, consider only wages in program that covers the risk of temporary the few years before retirement (table 5.4). income loss when a person loses a job, UI is In summary, these flaws in defined-benefit a means of sharing the social costs of reallo- pension systems exacerbate the financial cating labor among firms through periods of unsustainability of PAYG schemes, heighten unemployment. their advantages for formal workers, and Algeria has a contributory unemploy- operate against expanding social insurance ment benefit program, but it deviates from coverage to informal workers. In other words, the standards of UI in its ad hoc mechanism they do little to diminish the formal-informal for determining benefits. Egypt’s current segmentation of MENA’s labor markets. UI mechanism has been in place since 1975 TABLE 5.5 Unemployment benefit programs in MENA, 2011 Non-GCC MENA countries Type of unemployment benefit Remarks Algeria Unconventional unemployment benefit Ad hoc benefit formula is an average of last salary and the minimum wage with a ceiling of 3 times the minimum wage Egypt, Arab Rep. UI Hardly any UI beneficiaries due to restrictive conditions Iran, Islamic Rep. Unconventional unemployment benefit Resembles early retirement schemes with maximum benefit length in excess of 4 years Jordan UI + UA Introduced in 2011 Lebanon None — Lybia None — Morocco None — Syrian Arab Republic None — Tunisia Unconventional unemployment benefit Not UI as wage and benefit are disattached by minimum wage ceiling Yemen, Rep. None — GCC countries Bahrain UI + UA Introduced in 2006 Kuwait None Benefit conditional on unemployment may not be consistent with citizenship status-based cash transfer Oman UA Introduced in 2011 for a year Saudi Arabia UA Introduced in 2011 for the short term United Arab Emirates None — Source: United States Social Security Administration and International Social Security Administration (ISSA) (2010, 2011), and World Bank operational work in Oman and Saudi Arabia. Note: — = not available. MENA = Middle East and North Africa; GCC = Gulf Cooperation Council; UA = unemployment assistance; UI = unemployment insurance. 160 JOBS FOR SHARED PROSPERITY but is hardly operational because of highly According to the exit-voice theor y restrictive conditions on obtaining benefits. (Hirschman 1970), citizens can respond to The Islamic Republic of Iran’s unemploy- declining quality in (public or private) goods ment benefit scheme mixes family support by switching providers and making use of objectives with long-term benefit eligibility competition (exit) or by complaining to pro- (in excess of four years) as a means of helping viders and asking for quality improvements beneficiaries in their inactivity rather than as (voice). Both strategies can lead to higher a way to move them into a new job. Tunisia’s quality and productivity by pressuring firms unemployment benefit scheme does not live to improve. up to the principles of UI; by setting the ben- However, as recent Gallup data show, efit ceiling at the minimum wage, it delinks a culture in which citizens and clients the prior lay-off wage from the unemploy- express their opinions to public officials is ment benefit. Oman and Saudi Arabia intro- only nascent. People in MENA’s non-GCC duced ad hoc temporary unemployment countries voice their opinions significantly assistance schemes in 2011 as a response to less than people in comparator countries social tensions around unemployment. UI in from Eastern Europe and Latin America its more complete sense exists only in Bah- (figure 5.13). As will be discussed in the rain (introduced in 2006) and Jordan (2011). concluding chapter of this report, until Income protection for the unemployed in recently, the political climate in many coun- MENA could be far more effective, espe- tries did not encourage freedom of speech cially given its potentially significant role or voice. 8 With few worries about exit or in addressing the region’s key labor market voice, fi rms in MENA face even less pres- challenges. sure to improve or to hire the best workers, without regard to segments. Why does segmentation persist in MENA? What happens if segmentation Hiring in segments—giving priority to persists? prime-age workers over youth, or to men The divide between labor market insiders and over women, for example—is not efficient. market outsiders may be bridged as countries It bypasses talent in lower-priority groups, in the region continue to integrate their trade which remains unemployed or stagnates in with Europe and the rest of the world. Jordan precarious jobs that lack growth opportu- and Tunisia in particular have made great nities. On average, choosing to hire in seg- strides (Diop and Ghali 2012). Integration ments contributes to the low-productivity will invariably translate into stronger compe- equilibrium we observe in labor markets in tition and induce domestic fi rms to operate MENA. at their most profitable level. Firms that want A primary reason why MENA fi rms can to maximize profits will need to hire the afford to hire along segments is that they face most productive workforce and will therefore little competition. Chapter 4 provided ample discriminate less along the traditional lines evidence that firms in MENA have very few of youth and gender. Evidence on export- local competitors and experience very little ing fi rms in Europe confi rms this tendency creative destruction. One of the consequences (Heinze and Wolf 2010). of this stagnation is that in MENA, firms are Yet if segmentation persists in MENA’s under less pressure to hire the most produc- labor markets, a number of important oppor- tive workers. Another reason that can explain tunities will be missed. First, countries will why segmentation endures is that “voice” fail to obtain the best development results does not have a long tradition in MENA, and from their young and increasingly edu- among other consequences, limited voice can cated population. The institutionally forced reduce the pressure on firms to perform. marginalization of masses of outsiders in MENA’S INSTITUTIONAL FRAME WORK : REINFORCING THE STATUS QUO 161 FIGURE 5.13 Percentage of those who have “voiced” an opinion to a public official in the past month in selected world economies, 2010 a. Total population b. By gender Slovenia Spain Colombia Slovenia Bahrain Portugal Portugal Poland Kuwait Greece France France Spain Colombia Tunisia Bulgaria Brazil Brazil Bulgaria Yemen, Rep. United Arab Emirates United Arab Emirates Iraq Tunisia Yemen, Rep. Syrian Arab Republic Syrian Arab Republic West Bank and Gaza Algeria Morocco West Bank and Gaza Lebanon Morocco Kuwait Lebanon Jordan Jordan Iraq Poland Egypt, Arab Rep. Greece Bahrain Egypt, Arab Rep. Algeria 0 5 10 15 20 25 30 35 40 45 0 5 10 15 20 25 30 35 40 45 50 Percent Percent Women Men Source: Gallup World Poll 2011. See the appendix for more information on the poll. low-quality, informal jobs prevents them extend the current social insurance benefit from realizing their true value-adding poten- package to masses of young system entrants tial and limits economic development out- would almost certainly be broken, because comes. In an era when human capital is the systems are not fi nancially constituted emerging as the primary asset of production, to deliver benefits sustainable in the long this missed opportunity can have particularly term. However, widespread participation serious repercussions. in a reformed social insurance system, Second, MENA countries could fail to realigned to establish a closer relationship capture the potential for building social between contributions and benefits, would cohesion that is associated with widespread distribute the system’s unfunded liabilities access to social insurance. Social risk man- more equitably, both between and within agement usually works best with universal generations. access. Institutional exclusion from such pub- Finally, youth are a sizable and increas- lic mechanisms may very well have been the ingly vocal part of the population in all subtext of a dominant message of the Arab MENA countries. The population under Spring—the lack of dignity felt by outsid- 25 years of age ranges from about 40 percent in ers. Failure to respond could undermine the Tunisia to nearly 70 percent in the Republic achievements of that movement. of Yemen. The Arab Spring was largely A third opportunity—linked to integrat- youth led, enabled by social media tools ing the large number of young noncontribu- that youth wield with dexterity. Given this tors into pension systems—must go hand relative knowledge advantage, and the inde- in hand with close attention to the system’s pendence of social media from state control, fi nancial sustainability. A simple promise to the voice of this group will inevitably be 162 JOBS FOR SHARED PROSPERITY heard—preferably in constructive ways, in (for example, Neumark 1999) have opened open public debate. a debate. Elmeskov, Martin, and Scarpetta (1998) and Nickell and Layard (1997) find no significant impact of the minimum wage The unsustainable status quo on overall unemployment but conclude from existing research that higher minimum wages Labor regulations are not the prime driver of increase unemployment of all groups whose segmentation in MENA, but they cement the productivity falls, or might fall, below the disadvantages of disadvantaged and excluded wage, such as youth. groups. Strict legislation on employment pro- 4. Regarding international evidence on this phe- tection discourages access to jobs by new nomenon, see Mankiw (2006). entrants such as youth and women. For exam- 5. All references in the contractual section come ple, youth have less stable labor contracts than from raw data in World Bank (2012), unless prime-age workers in MENA and face longer otherwise noted. unemployment spells. Labor regulations also 6. For a comprehensive overview of the situa- exacerbate the barriers to private sector par- tion and potential for migration in the MENA region, see Luthria, Pouget, and Trenner ticipation discussed in the previous chapter: (forthcoming) and Zlaoui and Koettl (2009), restrictive legislation limits productivity gains as well as Keller (2010). by limiting the movement of labor, and labor 7. Both Jordan and Lebanon struggle with the turnover is indeed low in the region. phenomenon of skilled outmigration combined The current degree of segmentation in with unskilled immigration. See Jordan Ministry MENA’s labor markets is likely to be unsus- of Planning and International Cooperation tainable economically or socially. MENA’s (2011) and Robalino and Sayed (2012) for a social insurance systems operate on a fi nan- more in-depth discussion of this topic. cially unsustainable path and have started 8. According to the 2009 Polity IV rankings running deficits in some countries. Even for MENA, voice and accountability scores with sustainable parameters, most coun- are significantly lower than income in com- parator countries. According to the Freedom tries’ pension systems cannot afford to House Freedom of Press Index, only Egypt exclude young, unemployed contributors. and Morocco are rated “partially free”; all Moreover, the Arab Spring was a clear other MENA countries had a “not free” press expression of the frustration and sense of in 2009. exclusion generated by this equilibrium in labor markets. References Ahsan, A., and C. Pages. 2007. “Are All Labor Notes Regulations Equal? Assessing the Effects of Job 1. Assar Lindbeck and Dennis Snower devel- Security, Labor Dispute and Contract Labor oped what is known in labor economics as Laws in India.” Social Protection Discussion the insider-outsider theory, which studies Paper 0713, World Bank, Washington, DC. the behavior of economic agents in markets ———. 2009. “Are All Labor Regulations Equal? where some participants have more privileges Evidence from Indian Manufacturing.” Journal than others. of Comparative Economics 37 (1): 62–75. 2. The following arguments are taken from Aidt, T., and Z. Tzannatos. 2002. Unions and Grun et al. (2008). Collective Bargaining: Economic Effects in a 3. Most minimum wages in the world have no Global Environment. Washington, DC: World measurable impact on overall unemploy- Bank. ment, but they do affect youth unemploy- Almeida, R., and P. Carneiro. 2009. “Enforcement ment. Although work by Card and Krueger of Labor Regulation and Firm Size.” Journal (1997) found no significant aggregate of Comparative Economics 37 (1): 28–46. employment effect, critiques of this work (for Angel-Urdinola, D., and K. Tanabe. 2012. example, Deere, Murphy, and Welch 1995) “M icro -Determinants of I nformal and empirical work with a different result Employment in the Middle East and North MENA’S INSTITUTIONAL FRAME WORK : REINFORCING THE STATUS QUO 163 Africa Region.” Social Protection Discussion Boeri, T., and J. F. Jimeno. 2005. “The Effects Paper 1201, World Bank, Washington, DC. of Employment Protection: Learning from Assaad, R. 1992. “The Effects of Public Sector Variable Enforcement.” European Economic Hiring and Compensation Policies on the Review 49(8): 2057–77. Eg yptian Labor Market.” World Bank Boughzala, M. 2011. “Labor Market Regulations Economic Review 11 (1): 85–118. and Institutions: The Case of Tunisia.” Autor, D., W. Kerr, and A. Kugler. 2007. “Do Tunisia background paper, World Bank, Employment Protections Reduce Productivity? Washington, DC. Evidence from U.S. States.” IZA Discussion Caballero, R ., K. Cowan, E . Engel, and Paper 2571, Institute for the Study of Labor A. Micco. 2004. “Effective Labor Regulation (IZA), Bonn, Germany. and Microeconomic Flexibility.” N BER Baccaro, L., and D. Rei. 2005. “Institutional Working Paper 10744, National Bureau of Determinants of Unemployment in OECD Economic Research, Cambridge, MA. Countries: A Time-Series Cross-Section Caballero, R., and M. L. Hammour. 2000. Analysis (1960–98).” International Institute “Creative Destruction and Development: for Labor Studies Discussion Paper 160, Institutions, Crises, and Restructuring.” International Institute for Labour Studies, NBER Working Paper 7849, National Bureau Geneva. of Economic Research, Cambridge, MA. Bassanini, A., and R. Duvall. 2006. “Employment Card, D., and A. B. Krueger. 1997. Myth and Patterns in OECD Countries: Reassessing the Measurement: The New Economics of the Roles of Policies and Institutions.” OECD Minimum Wage. Princeton, NJ: Princeton Economics Department Working Paper 486, University Press. Organisation for Economic Co-operation and Chartouni, C. 2011. “Impact of the Public Development (OECD), Paris. Sector on Reservation Wages in the Private Bassanini, A., and D. Venn. 2007. “Assessing Sector (UAE).” Working paper, World Bank, the Impact of Labour Market Policies on Washington, DC. Productivity: A Difference-in-Differences Deere, D., K. M. Murphy, and F. Welch. 1995. Approach.” OECD Social, Employment and “Employment and the 1990–1991 Minimum- Migration Working Papers 54, OECD, Paris. Wage Hike.” American Economic Review Belot, M., J. Boone, and J. van Ours. 2007. 85 (2): 232–37. “Welfare Improving Employment Protection.” Diop, N., and S. Ghali. 2012. “Are Jordan Economica 74: 381–96. and Tunisia’s Expor ts B ecoming More Benassy, J.-P. 1995. “Nominal Rigidities in Wage Technologically Sophisticated? Analysis Using Setting by Rational Trade Unions.” Economic Highly Disaggregated Export Databases.” Journal 105 (430): 635–43. MENA Working Paper Series 54, World Bank, Bentolila, S., J. Dolado, and J. Jimeno. 2011. Washington, DC. “Reforming an I nsider- Outsider Labor Elmeskov, J., J. Martin, and S. Scarpetta. 1998. Market: The Spanish Experience.” CEPR “Key Lessons for Labour Market Reform: Discussion Paper 8691, Centre for Economic Evidence from OECD Country Experiences.” Policy Research, London. Swedish Economic Policy Review 5 (2): 205–52. Besley, T., and R. Burgess. 2004. “Can Labor Eslava, M., J. Haltiwanger, A. Kugler, and Regulation Hinder Economic Performance? M. Kugler. 2010. “Factor Adjustments after Evidence from India.” Quarterly Journal of Deregulation: Panel Evidence from Colombian Economics 119 (1): 91–134. Plants.” Review of Economics and Statistics Blanchard, O. 1998. Revisiting European 92 (2): 378–91. Unemployment: Unemployment, Capital Forteza, A., L. Lucchetti, and M. Pallares- Accumulation and Factor Prices. Research Miralles. 2009. “Measuring the Coverage Series. Dublin: Economic and Social Research Gap.” In Closing the Coverage Gap: The Role Institute (ESRI). of Social Pensions and Other Retirement Bodor, A., D. Robalino, and M. Rutkowski. In c o m e Tra n sfe rs , e d . R . Hol z m a n n , 2008. “How Mandatory Pensions Affect D. Robalino, and N. Takayama, 23 – 40. Labor Supply: Decisions and Human Capital Washington, DC: World Bank. Accumulations? Options to Bridge the Gaertner, M. 1981. “A Politico-Economic Model Gap between Economic Theory and Policy of Wage Inflation.” De Economist 129 (2): Analysis.” Bank I Kredit 2 (July): 3–18. 183–205. 164 JOBS FOR SHARED PROSPERITY Gatti, R., D. Angel-Urdinola, J. Silva, and A. Washington, DC: International Bank for Bodor. 2012. Striving for Better Jobs: The Reconstruction and Development and World Challenge of Informality in the Middle East Bank. and North Africa Region. Washington, DC: Kugler, A., and G. Saint-Paul. 2000. “Hiring World Bank. and Firing Costs, Adverse Selection and Grun, R., L. Etter, and I. Jillson. 2010. “Arab the Persistence of Unemployment.” CEPR Republic of Egypt: Management and Service Discussion Paper 2410, Centre for Economic Quality in Primary Health Care Facilities in Policy Research, London. the Alexandria and Menoufia Governorates.” Luth ria, M ., Y. Pouget , and S . Tren ner. Working paper, World Bank, Washington, DC. Forthcoming. “Integrating Labour Markets.” Grun, R., L. Etter, J. Silva, and J. F. Giret. Proceeding from a program of the Center for 2008. “Labor Demand, Skills Supply and Mediterranean Integration, Marseille, France. Employment: Towards an Integrated Strategy Machin, S., A. Manning, and L. Rahman. 2003. for Job Creation —Phase II (MILES).” Tunisia “Where Minimum Wage Bites Hard: The Policy Note, World Bank, Washington, DC. Introduction of the UK National Minimum Haltiwanger, J., S. Scarpetta, and H. Schweiger. Wage to a Low Wage Sector.” Paper Presented 2010. “Cross Country Differences in Job at the Royal E conomic Society A nnual Reallocation: The Role of Industry, Firm Conference, London, April 7–9. Size and Regulations.” Working Paper 116, Mahjoubi, A. 2003. Les Aspects Pratiques des European Bank for Reconstruction and Salaires, Primes et Avantages. Arab Edition. Development, London. Self publication, BP 21 Tunis 1075. Heckman, J. J., and C. Pages. 2000. “The Cost of Mankiw, N. G. 2006. Macroeconomics. 6th rev. Job Security Regulation: Evidence from Latin ed. New York: Worth Publishers Inc. American Labor Markets.” NBER Working Messina, J., and G. Vallanti. 2007. “Job Flow Paper 7773, National Bureau of Economic Dynamics and Firing Restrictions: Evidence Research, Cambridge, MA. from Europe.” Economic Journal 117 (521): Heinze, A., and E. Wolf. 2010. “The Intra-Firm 279–301. Gender Wage Gap: A New View on Wage Micco, A., and C. Pages. 2006. “The Economic Differentials Based on Linked Employer– Effects of Employment Protection Legislation: Employee Data.” Journal of Population Evidence from International Industry-Level Economics 23 (3): 851–79. Data.” IZA Discussion Paper 2433, Institute Hendy, R., and I. Selwaness. 2011. “Employability for the Study of Labor (IZA), Bonn, Germany. in the Middle East and North Africa Region: Mryyan, N. 2011. “Beyond the Insider-Outsider The Case of Egypt.” Background paper, World Structure of the Labor Market: The Case of Bank, Washington, DC. Jordan.” Jordan background paper, World Hirschman, A. O. 1970. Exit, Voice and Loyalty: Bank, Washington, DC. Responses to Decline in Firms, Organizations, Musette, M., and M. Mohamed-Meziani. 2011. an d St ates . Ca mbridge, M A : H a r va rd “Beyond the Insider-Outsider Structure of the University Press. Labor Market.” Algeria background paper, ILO-KILM (International Labour Organization– World Bank, Washington, DC. Key I ndicators of the Labour Market) Neumark, D. 1999. “The Employment Effects of (database). International Labour Organization, Recent Minimum Wage Increases: Evidence Geneva, Switzerland. http://kilm.ilo.org / from a Pre-Specified Research Design.” NBER kilmnet/. Working Paper 7171, National Bureau of Jimeno, J. F., and L. Toharia. 1996. “Effort, Economic Research, Cambridge, MA. Absenteeism, and Fixed-Term Employment Nickell, S., and R. Layard. 1997. “The Labour Contracts.” Revista Española de Economía Market Consequences of Technical and 13 (1): 105–19. Structural Change.” Discussion Paper Series Jordan Ministry of Planning and International 23, Centre for E conomic Performance, Cooperation. 2011. National Employment London. Strateg y 2011–2020 , by Omar Razzaz. Nicoletti, G., A. Bassanini, E. Ernst, S. Jean, Consultant report, Jordan. P. Santiago, and P. Swaim. 2001. “Product Keller, J. 2010. Labor Migration from North and Labour Market Interactions in OECD Africa: Development Impact, Challenges, Countries.” OECD Economics Department and Policy Options. Vol. I: Main Report. Working Paper 312, OECD, Paris. MENA’S INSTITUTIONAL FRAME WORK : REINFORCING THE STATUS QUO 165 OECD (Organisation for Economic Co-Operation ———. 2011. Social Security Programs throughout and Development). 1999. Employ ment the World: Africa, 2011. Washington, DC: Outlook. Paris: OECD. SSA Publication. ———. 2006. Employment Outlook. Paris: OECD. Social Security Corporation (Jordan). http:// Palacios, R., and E. Whitehouse. 2006. “Civil- www.ssc.gov.jo/. Service Pension Schemes around the World.” World Bank. 2009. “Egypt Investment Climate Social Protection Discussion Paper 0602, Assessment 20 09: Accelerating Private World Bank, Washington, DC. Enterprise-Led Growth.” Working paper, Pa l l a re s - M i ra l le s , M . , C . Rom ero, a nd World Bank, Washington, DC. E. Whitehouse. 2012. “International Patterns ———. 2011. Migration and Re mit tances of Pension Provision II: A Worldwide Overview Factbook 2011. 2nd ed. Washington, DC: of Facts and Figures.” Social Protection World Bank. Discussion Paper 70319, World Ban k, ———. 2012. Doing Business 2012: Doing Washington, DC. Business in a More Transparent World . Pension Superintendency (Chile). http://www Washington, DC: World Bank. .spensiones.cl /por tal /instit ucional /578/ ———. 2013. World Development Report 2013: w3-propertyvalue-5925.html. Jobs. Washington, DC: World Bank. Robalino, D. A., and T. Bogomolova. 2006. World Bank and Institut National de la Statistique “Implicit Pension Debt in the Middle East (INS). Forthcoming. “Recent Labor Market and North Africa: Magnitude and Fiscal Trends and Main Characteristics of the Labor lmplications.” MENA Discussion Paper 46, Market in Tunisia.” Working paper, World World Bank, Washington, DC. Bank, Washington, DC. Robalino, D. A., and H. Sayed. 2012. “Good Youssef, T. 2004. “Development, Growth and Jobs Needed: The Role of Macro, Investment, Policy Reforms in the Middle East and North Education, Labor and Social Protection Africa since 1950.” Journal of Economic Policies (“MILES”).” Working paper, World Perspectives 18 (3): 19–116. Bank, Washington, DC. Zlaoui, L., and J. Koettl. 2009. Shaping the Social Securit y Administration and ISSA Future: A Long-Term Perspective of People (International Social Security Association). and Job Mobility for the Middle East and 2010. Social Security Programs throughout North Africa Region . Washington, DC: the World: Asia, 2010. Washington, DC: SSA World Bank. Publications. Unemployable and Unemployed: Skills Gaps and a Meritocracy 6 Deficit in MENA Main findings • Despite the achievements in enrollment rates, education systems in MENA continue to lag behind in providing the quality education and skills that are relevant for private sector employment. The system continues to operate under a logic of “selecting” candidates for public employment rather than a logic of fostering learning. • Many employers make hiring decisions based on criteria other than candidates’ skills and qualifications because of the inability of formal education to signal candidates’ ability. • The “meritocracy deficit” in the private sector reduces the incentives among youth to seek the most relevant education and relieves pressure on education to focus on quality. • Vocational and training systems continue to operate in isolation from the private sector; existing partnerships are poorly institutionalized and operationalized. • In addition to the mismatch between supply and demand for skills, there is a mismatch between graduates’ aspirations and the availability of acceptable and decent jobs. Becoming employable in MENA: A matter of educational quality and relevance The “road not traveled for education reform” in the MENA region also refers to the fact that education systems must now travel over new and relatively unexplored ter- rain. Education systems have a tendency to spawn a new set of challenges for every problem resolved. Once everyone is in school, we must ensure that they do not drop out. Once they remain seated, we must make sure that they learn something. Once they appear to be ready to learn, we must make sure that the material is useful for their future. World Bank, 2008a 167 168 JOBS FOR SHARED PROSPERITY O ne of the key goals of all educational have brought enrollment in tertiary educa- and training systems is to generate tion in line with comparator countries (see employable graduates. Although figure 6.3). Notable outliers are some of employability entails the capital accumula- the oil-rich countries in the Persian Gulf— tion of skills, competencies, academic cer- Oman, Qatar, Saudi Arabia, and the United tificates, and professional qualifications and A rab Emirates —which have achieved refers to the capacity of graduates to func- enrollment rates near the world average of tion in a job, it does not imply the acquisi- 30 percent but are still below the level pre- tion of a job (Yorke 2006). An employable dicted by their high income. The enrollment graduate is not just one who demands a job in tertiary education in Djibouti, Morocco, but rather one who is in a position to offer and the Republic of Yemen is also below relevant skills to the labor market. Today’s what their income would predict. graduates in countries in the Middle East As has been the case in many other coun- and North Africa (MENA) have invested tries around the world, the rapid expansion heavily in education but fall short of options of secondary and higher education has led in the labor market. As one young woman in to increasing expectations from families and a rural part of the Arab Republic of Egypt employers alike. Just as employers expect to puts it, “ Education will not help us find see readily employable graduates equipped employment. Many university graduates with relevant skills and competencies, an can’t fi nd work. They are exactly like those increasing number of parents expect to see with primary education or like farmers. At their investment pay off when their children least farmers have land. But university grad- complete secondary and tertiary education. uates have nothing.”1 As far as students are concerned, the increas- ing expectations from society, coupled with large increases in educational attainment, Great progress, greater expectations, have led to even greater aspirations for their and even greater aspirations employment. However, it appears that the Following independence, most MENA coun- dramatic expansion of post–basic education tries sought to rapidly expand access to edu- in MENA is not living up to the expectations cation, particularly to higher education. As a of employers, families, and young gradu- result, impressive progress has been made in ates: many graduates are not getting jobs (see enrollment and completion rates in both sec- chapters 1 and 2); many employers claim that ondary and tertiary education (see figure 6.1). young graduates are not well equipped with Enrollment in secondary schooling increased relevant skills; and the quality of education almost threefold between 1970 and the mid- in MENA is low by international standards, 2000s, and the increase was nearly fivefold in with too many students not learning in higher education (World Bank 2008a). This school. In short, youth in MENA are not just process brought about a broadening of edu- at risk of being unemployed but, even worse, cational opportunities and a narrowing of are also at risk of being unemployable. gender and socioeconomic gaps in access to education. Figure 6.2 shows average gross Low quality and low relevance: Skill enrollment in secondary and tertiary educa- gaps and mismatches tion by region and gender. In MENA, where enrollment in primary education is nearly uni- The low quality and relevance of education versal, average enrollment is over 70 percent are widely seen as the most important rea- in secondary school and close to 30 percent in son for the failure of MENA’s educational tertiary education for both men and women. and training systems to produce employable Educational systems have been very graduates, endowed with the knowledge and responsive to the increased demand for ter- skills required for the labor market and for tiary education. Most countries in MENA life (ETF and World Bank 2005; World Bank U N E M P LOYA B L E A N D U N E M P LOYE D: S K I L L S G A P S A N D A M E R I TO C R AC Y D E F I C I T I N M E N A 169 FIGURE 6.1 Educational attainment in secondary and tertiary education in MENA, 1991–2010 a. Gross enrollment ratio in secondary education b. Gross enrollment ratio in tertiary education Bahrain 97 51 Bahrain Qatar 94 Lebanon 50 Oman 91 West Bank 91 Jordan 38 and Gaza United Arab 91 33 Emirates Tunisia Kuwait 91 Iran, Islamic 32 Algeria 90 Rep. 89 Saudi Arabia 32 Tunisia Jordan 86 United Arab 30 Emirates Lebanon 82 Egypt, Arab 29 Iran, Islamic 78 Rep. Rep. Syrian Arab 71 Oman 26 Republic Egypt, Arab 67 25 Rep. Algeria Morocco 52 Morocco 12 Iraq 51 26 Yemen 10 Djibouti 0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 Average, 1991–95 Average, 2005–09 Source: EdStats (database). Note: MENA = Middle East and North Africa. 2008a). In the words of a university student by comparing eighth graders in MENA at in Egypt, “[Employers] will tell you that what different levels of achievement to an inter- you learned at university is not what you will national benchmark. 3 The figure shows use for work.” To be sure, despite remarkable that, with the exception of Jordan and Leba- progress in access and completion rates, evi- non, more than three-fourths of students in dence on learning outcomes—as measured by MENA possess only a basic or below basic the Trends in International Mathematics and knowledge of mathematics in 2007, well Science Studies (TIMSS) among eighth grad- below the world median. While MENA has a ers and by the Programme for International large number of low and very low achievers, Student Assessment (PISA) among 15-year- it has few high performers at the other end olds—points to the relatively low quality of of the scale. There are virtually no students education in the region.2 in the “advanced” category, with Jordan and In absolute terms, MENA countries fail Lebanon in 2007 leading the MENA coun- to raise even half their student population tries with 10 and 9 percent in the “high” cat- to “low” levels of learning. Figure 6.4 illus- egory. Recent data for 2011 suggest a signifi- trates the region’s poor quality of education cant improvement in competencies in GCC 170 JOBS FOR SHARED PROSPERITY FIGURE 6.2 Gross enrollment in secondary and tertiary education FIGURE 6.3 Gross enrollment in tertiary by gender in selected world regions, 2009 education in selected economies in MENA by GDP per capita, 2005–09 a. Secondary 100 Tertiary gross enrollment Sub-Saharan Africa ratio (latest year) South Asia 50 Lebanon West Bank and Gaza Bahrain Tunisia Iran, Islamic Rep. Saudi Arabia Egypt, Arab Rep. Algeria OmanUAE World Yeme, rep. Morocco 0 Djibouti Qatar East Asia and Pacific 6 7 8 9 10 11 Log GDP per capita Middle East and North World MENA Fitted values Africa Latin America and the Caribbean Source: Based on EdStats and World Development Indicators. Note: GDP = gross domestic product; MENA = Middle East and North Africa; North America UAE = United Arab Emirates. the labor market, as critical reasoning skills Europe and Central Asia are increasingly seen as central to success in 0 20 40 60 80 100 high-value-added jobs.4 Percent On average, MENA countries under- b. Tertiary perform in TIMSS and PISA compared to Sub-Saharan Africa countries at similar levels of income per capita, with the notable exception of Jordan, South Asia Lebanon, Tunisia, and the Syrian Arab Republic. According to these measures, World learning outcomes are particularly poor in East Asia and Pacific the Gulf countries, given their high rates of economic development. For some (such as Middle East and North Africa Egypt, Morocco, Saudi Arabia, and the West Latin America and Bank and Gaza), the performance in TIMSS the Caribbean even worsened between 2003 and 2006. Europe and Central Asia Results for mathematical literacy among 15-year-olds in the four countries participat- North America ing in PISA are even less encouraging. While 0 20 40 60 80 100 students in Jordan scored as expected given Percent its level of income, Qatar, Tunisia, and the Female Male United Arab Emirates performed signifi- cantly worse compared with countries with Source: EdStats (database). similar levels of income (figure 6.5). TIMSS and PISA are administered to eighth graders and 15-year-old students and countries, driven by Qatar and Saudi Arabia, provide a solid assessment of the quality of and the addition of the United Arab Emir- education at these ages. There is, however, ates. In contrast, non-GCC MENA countries no comparable assessment at higher levels of improved only marginally (such as Tunisia) or education and training (tertiary and techni- even deteriorated (such as Jordan). Overall, cal and vocational education and training, or these fi ndings indicate that the educational TVET). Moreover, these tests measure qual- system is not capable of producing a critical ity of learning only among those who are in mass of students who have the fundamentals the school system, while there are many who to perform well in labor markets. This fail- have dropped out of school who might score ure is bound to have detrimental impacts on even lower on these measures. U N E M P LOYA B L E A N D U N E M P LOYE D: S K I L L S G A P S A N D A M E R I TO C R AC Y D E F I C I T I N M E N A 171 Effect of poor educational quality on FIGURE 6.4 Selected economies in MENA, the region as a whole, attendance and school completion and the OECD benchmarked against the international median in The dropout rate provides an alternative indi- mathematics, 2007 and 2011 cator of educational quality, provided quality is defined as the capacity of the system to give a. TIMSS 2007 a personalized educational response to every Jordan 39 26 24 student by responding to individual needs, Lebanon 26 38 26 interests, and characteristics.5 Recent survey Egypt, Arab Rep. 53 26 16 Tunisia 39 40 18 data indicate that school dropout is highest Bahrain 51 30 16 in lower-secondary, followed by primary, Syrian Arab Republic 53 30 14 education (figure 6.6). West Bank and Gaza 61 24 12 Many factors in the schools themselves Oman 59 27 12 condition the quality of instruction, such Morocco 59 28 12 Algeria 59 34 7 as inadequate facilities, overcrowded class- Kuwait 71 23 6 rooms, inappropriate language of instruc- Qatar 84 12 4 tion, and teacher absenteeism; these may Saudi Arabia 82 15 2 contribute to student dropout (Colclough, Rose, and Tembon 2000). In particular, there MENA average 54 27 14 OECD average 11 23 32 is evidence that students and parents take the International median 25 29 31 perceived quality of education into account 0 20 40 60 80 100 when making decisions about whether to stay Percent in school or drop out. For example, evidence from Egypt shows that student decisions Very low score (<400) Low score (400–474) on whether to stay in school are influenced Intermediate score (475–554) High score (550–624) Advanced score (>625) by the quality of their school, measured as the gain in achievement that a student can b. TIMSS 2011 expect from attending a given school for an additional year. As Hanushek, Lavy, and Morocco 2 10 24 64 Oman 4 12 23 61 Hitomi observe, “Holding constant the stu- Syrian Arab Republic 3 14 26 57 dent’s own ability and achievement, a stu- Saudi Arabia 4 15 27 53 dent is much less likely to remain in school West Bank and Gaza 6 18 27 48 if attending a low quality school rather than Bahrain 7 18 27 47 a high quality school” (2006, 2). Data from Qatar 8 19 25 46 Jordan 6 20 29 45 2005–06 show that about one in five parents Iran, Islamic Republic 6 18 29 45 in Egypt cites poor school quality as a rea- Tunisia 5 20 36 39 son that their child dropped out of primary Lebanon 8 29 35 27 school and about one in 10 for dropping out United Arab Emirates 12 28 31 27 of lower-secondary school (El-Zanaty and MENA GCC average 7 18 27 47 Gorin 2007). MENA non-GCC average 5 18 29 46 OECD average 21 32 24 12 Pervasive skill mismatches International median 14 29 29 25 In addition to quality of education, the 0 20 40 60 80 100 relevance of graduates’ skills is crucial Percent for employability. The extent of skill mis- Very low score (<400) Low score (400–474) matches, however, is hard to quantify, given Intermediate score (475–549) High score (550–624) the lack of data and limitations in measur- Advanced score (>625) ing the skill content of demand. Skill mis- matches, therefore, are mostly inferred Source: Based on Trends in International Mathematics and Science Study (TIMSS). Note: GCC = Gulf Cooperation Council; OECD = Organisation for Economic Co-operation and rather than measured directly. Stakehold- Development; MENA = Middle East and North Africa. ers’ perceptions of constraints to graduates’ 172 JOBS FOR SHARED PROSPERITY FIGURE 6.5 Quality of math education in selected countries in FIGURE 6.6 Dropout rates by highest level of MENA as measured by TIMSS and PISA, 2011 and 2009 education attained in the Arab Republic of Egypt, 2009 a. 2011 600 Post-secondary/university/ TIMSS mathematics scores, 2011 post graduate 3.51 Completed vocational 9.51 500 United Arab Completed secondary 5.94 Lebanon Emirates Tunisia Iran, Islamic Rep. Bahrain 400 Jordan Qatar Saudi Arabia Completed preparatory 12.87 Morocco Syrian Arab Republic Oman Completed primary 39.90 300 7 8 9 10 11 Can read and write 28.25 Natural log GDP per capita 2010 b. 2009 Illiterate 0 550 0 5 10 15 20 25 30 35 40 45 PISA mathematics scores, 2009 Percent 500 Source: Based on data from the Egypt’s Survey of Young People in Egypt 450 United (SYPE) 2009. See the appendix for more information on the survey. Arab Emirates 400 Jordan Tunisia Qatar 350 and firm growth.7 This share is the high- est in all developing regions of the world, 8 9 10 11 comparing with 14 percent in South Asia Natural log GDP per capita and about 7 percent in Germany and the World Middle East and North Africa Fitted values Republic of Korea, two countries with strong educational and training systems (figure 6.7). Source: TIMSS and IMF World Economic Outlook (database), January 2013 update. Other sources corroborate these perceptions, Note: For panel a, the GDP per capita of year 2010 is an IMF estimate, not final figures, for Botswana, Chile, Georgia, Ghana, Honduras, Lebanon, Oman, South Africa, the United Arab Emirates, and the including the Arab World Competitiveness United Kingdom. GDP = gross domestic product; MENA = Middle East and North Africa; Report 2011–12 (WEF and OECD 2011), PISA = Programme for International Student Assessment; TIMSS = Trends in International Mathematics and Science Study. which ranks inadequate education as the fourth-most-constraining issue to economic growth, after access to fi nancing, restrictive education-to-work transition and the impor- labor regulations, and inefficient government tance of skill mismatches often stem from bureaucracy (figure 6.8). qualitative work, such as focus groups with Two different types of skills are relevant employers (see box 6.1 for an example in to employers: hard skills (job-related or tech- the West Bank and Gaza) or large-scale nical) and soft skills (such as the creativ- employer surveys. ity, the ability to communicate clearly, and According to surveys of private sec- problem-solving and interpersonal skills).8 tor employers, the skills of job applicants Employer perceptions of skill mismatches have low relevance to their firms’ business seem to be stronger for TVET than for uni- needs, and thus these employers question versity graduates, and somewhat higher for the system’s ability to produce employable technical and cognitive (hard) skills than for graduates.6 Overall, more than one-third of soft skills (figure 6.9). About one-third of employers in MENA identify skill shortages employers interviewed in Egypt, Jordan, and as a major constraint to business operation the Republic of Yemen (IFC and ISDB 2011) U N E M P LOYA B L E A N D U N E M P LOYE D: S K I L L S G A P S A N D A M E R I TO C R AC Y D E F I C I T I N M E N A 173 BOX 6.1 Perceptions of the main constraints to the education-to-work transition in the West Bank and Gaza Six focus groups in the West Bank and five groups The focus group participants clearly identifi ed the in Gaza met between March 17 and 23, 2011, to relevance of skill to the needs of the labor market, help stakeholders better understand the constraints the level of competency, and amount of experience facing graduates as they transition from education of graduates among the most often named con- to work. The groups represented a mix of business straints to employment for youth. When asked to owners and principals, students, and graduates identify the constraints young women face to get- of technical colleges and universities. The Min- ting into a paid job upon graduation, participants istry of Education and Higher Education and the noted that educational and training programs are World Bank conducted the interviews jointly. The mainly male oriented, that support for reconciling participants were split into small groups of three work and home duties is lacking, and that women to fi ve and were asked to identify the “four con- are confi ned to employment in professions with straints faced by young people to get into a paid limited absorptive capacity. job upon graduation.” Figure B6.1.1 groups the Participants noted specific skills and labor mar- responses and their frequencies into 11 categories. ket mismatches: FIGURE B6.1.1 Perceptions of the main constraints in transitioning from education to work in the West Bank and Gaza Political situation 38 Education (relevance) 36 Market assessment/coordination 29 Education (competency and 26 experience) Capacity of tertiary 24 education institutions Market size 23 Culture 15 Nepotism/queing 12 Salary 9 Entrepreneurship support 8 Student orientation 5 0 5 10 15 20 25 30 35 40 Number of respondents answering “yes”a Source: Brodmann et al. 2012a. Note: Queuing = voluntary unemployment to wait for a public sector job. a. That is, reporting that this is their main constraint. (continued next page) 174 JOBS FOR SHARED PROSPERITY BOX 6.1 Perceptions of the main constraints to the education-to-work transition in the West Bank and Gaza (continued) • Programs and skills not oriented to market needs They also identified gaps in competence and • Limited enrollment in technical specialties experience: • Limited attention to vocational education • Lack of job-relevant experience • Large number of graduates in same specialties • Lack of technical experience • Lack of life skills • Industry’s lack of capacity and experience in • Lack of continuous education training students • Education becoming commercialized; too many colleges Source: Brodmann et al. 2012a. FIGURE 6.7 Share of firms identifying inadequately educated and relevance appears to be predictive of workforce as a major constraint to growth, by region and selected a wide range of outcomes, 9 such as edu- countries, 2005–11 cational attainment and employment out- comes, often more than cognitive skills.10 40 However, achievement tests do not mea- sure those soft skills, and the school cur- 35 ricula in MENA countries do not include 30 them explicitly, with the exception of some “life skills” modules usually developed and 25 implemented in training by nongovernmen- tal organizations (NGOs). These interven- Percent 20 tions do not reach the majority of students, 15 and their impact on student performance and skill acquisition is not systematically 10 evaluated. In fact, whether soft skills are malleable and can be taught and learned in 5 school and university is an open question.11 0 In MENA, some evidence from a rigorous impact evaluation of an entrepreneurship re y a p. a G e in a sia co ica nt ean sia e O aran ific an Car ca ut s -in b- nd rld So trie Ko man Re pe e Afri hA a r lA c a o Af Sp d ibb co Sah Pa training program in Tunisia suggests that un W a, ra r Eu and orth Ce the intervention affected a range of behav- D N a EC ro th i ica d As er an ioral skills (personality traits), in particular, st gh Su Am ast m Ea tin e E agreeableness and extraversion (Premand La iddl et al. 2012; see also chapter 7, box 7.12).12 M hi Region or country Another indication of prevalent mis- Source: Enterprise Surveys (global dataset). matches is the disconnect between labor Note: OECD = Organisation for Economic Cooperation and Development. market demand and patterns of students’ demand for higher education, as shown, for assert that recently hired university graduates example, in shares of enrollment in different lacked soft skills. fields of study. Figure 6.10 exemplifies the Soft skills are usually not part of second- mismatch in Tunisia, where graduates in cer- ary school curricula and are not measured tain fields are in high demand in the labor by achievement tests. However, this addi- market, whereas other fields have less labor tional dimension of educational quality market relevance. Graduates from technical U N E M P LOYA B L E A N D U N E M P LOYE D: S K I L L S G A P S A N D A M E R I TO C R AC Y D E F I C I T I N M E N A 175 study fields (including architecture, medicine, FIGURE 6.8 Most problematic factors for doing business in the telecommunications, electricity, engineering, Arab world and sciences) and health are recruited for jobs that correspond to their level of qualifi- Access to financing cation, whereas graduates from law, humani- Restrictive labor regulations ties, and social sciences initially often work Inefficient government in jobs that require lower levels of qualifica- bureaucracy Inadequately educated workforce tion (figure 6.10). Corruption This indicator of the education-to-work Poor work ethic in national transition one-and-a-half to three years after labor force Inadequate supply of infrastructure graduation, however, does not account for the longer-term impact of field of study on Policy instability employment outcomes. International evi- Government instability/coups dence suggests that life-cycle implications Inflation for education matter. Hanushek, Machin, Tax rates and Woessmann (2011), for example, use Tax regulations the International Adult Literacy Survey to Foreign currency regulations show that individuals with general education Poor public health initially face worse employment outcomes Crime and theft but that they experience more and better 0 2 4 6 8 10 12 14 employment opportunities as they become Percentage of responses older than individuals with more vocational- oriented degrees and diplomas. Source: World Economic Forum/OECD 2011. Note: From a list of 15 factors, respondents were asked to select the 5 most problematic for doing Overall, enrollment in different fields of business in their country and to rank them between 1 (most problematic) and 5 (least problematic). tertiary education is unbalanced in MENA. The bars in the figure show the responses weighted according to their rankings. A very high percentage of students studies humanities, education, and social sciences. In the West Bank and Gaza, for example, FIGURE 6.9 Mismatch in hard and soft skills of newly hired graduates in selected countries in MENA, 2010 73 percent of students are enrolled in these fields of study, with a meager 17 percent enrolled in scientific, technical, or engineering 50 fields (figure 6.11). 40 Barriers to employability: Why 30 Percent are there skill gaps? 20 At least three sets of arguments can help 10 explain the prevalence of these widespread mismatches and low relevance of skills: 0 p. . a n co ep bi a Re • Because the private sector and the edu- oc rd ,R ra b Jo or en iA ra M cational and training sector operate in m ud t, A Ye Sa yp isolation, information and signaling fail- Eg ures occur on both sides. University graduates’ hard skills University graduates’ soft skills • The public sector is the main “client” of Vocational graduates’ hard skills Vocational graduates’ soft skills the educational and training system and is thus the main shaper of students’ choices and expectations. Source: IFC and ISDB 2011. Note: Managers agreeing that graduates hired in the past year have the appropriate skills. Numbers • A “logic of selection”—through rigid track- show the percentage of the company’s graduates hired in the past years that have appropriate skills. ing in secondary education, high-stakes MENA = Middle East and North Africa. 176 JOBS FOR SHARED PROSPERITY FIGURE 6.10 Proportion of graduates employed below their level examinations, and the low status of TVET— of qualification, by field of specialization, in Tunisia, 2005 and 2007 prevails over a “logic of learning” in educa- tional and training systems. Architecture 0 Medicine 2 Isolation of the private sector from the Telecommunication and education and training sector 7 electricity In MENA, the private sector and the edu- Health and social services 16 cational and training sector tend to operate Agricultural engineering 26 in isolation, resulting in skill gaps and mis- Engineering 32 matches (see, for example, ETF and World Bank 2005; IFC and ISDB 2011; World Sciences 24 Bank 2008a). The lack of communication Languages 35 and coordination between the sectors is both Information technology 38 cause and consequence of major informa- Business, economics, tion and knowledge gaps on both sides. As 52 and finance a result, the educational and training system Commerce and administration 55 lacks the information necessary to respond to Humanities 69 the needs of the private sector, whereas the private sector lacks the capacity and interest Law 72 to play its role in a demand-driven skill devel- 0 10 20 30 40 50 60 70 80 opment system. This issue is particularly rele- Percent vant in the TVET subsector, where the role of 3.5 years after graduation 1.5 years after graduation employers is by definition crucial in ensuring that the skills acquired are relevant for access Source: MFPE and World Bank 2009. to the labor market. A similar mechanism is Note: Calculated based on tracer survey from graduates of 2005 and 2007. in play with respect to firms’ incentives to FIGURE 6.11 Distribution of university graduates by field of study train their workforce. Firms in MENA oper- in selected economies in MENA, 2004–10 ate in a limited-competition environment and do not fully internalize the benefits of a skilled workforce. West Bank and Gaza There is little tradition of an institution- Qatar alized dialogue between the educational and Lebanon Algeria training system, on the one hand, and the Saudi Arabia private sector, on the other. Efforts to adapt Jordan TVET to the requirements of the private sec- Oman tor include broadening the participation of United Arab Emirates other social partners in the governance of Iraq Morocco TVET systems. Several countries in MENA Tunisia have created national and regional TVET Iran, Islamic Rep. councils with a view toward providing a sta- Djibouti ble governance body bringing together public 0 20 40 60 80 100 and private stakeholders.13 Such councils are Percent meant to coordinate, facilitate, and incentiv- Scientific, technical, and engineering Health ize private sector involvement in curriculum Education and humanities Social sciences and program development, to partner in prac- Other tical training and apprenticeship schemes, to contribute to funding for training, and, most Sources: Edstats (database); Tunisian data from Bureau des Etudes de la Planification et de la important, to bridge the information gaps Programmation 2010, 2011. and failures between the private sector and Note: MENA = Middle East and North Africa. Data from Tunisia refer to students enrolled during the academic year 2010–11. the educational and training sector. U N E M P LOYA B L E A N D U N E M P LOYE D: S K I L L S G A P S A N D A M E R I TO C R AC Y D E F I C I T I N M E N A 177 Many initiatives to include employer and social partners. Box 6.2 illustrates the exam- union representatives in governance have ple of Egypt, Jordan, and Tunisia. Even in not succeeded because of the ad hoc nature cases where such public-private partnerships of committees and the lack of operational are purposefully institutionalized, they often responsibilities among the participating fail to work, because the top leadership lacks BOX 6.2 Private sector involvement in skill development systems in the Arab Republic of Egypt, Jordan, and Tunisia In Egypt, the Supreme Council for Human Resources by all other stakeholders. Thus the council has not Development (SCHR D), a ministerial council addressed policy issues in the sector beyond the chaired by the minister of manpower and migration, scope of the Ministry of Labor, and the private sector is the highest formal authority for formulating poli- employers have been less motivated and less involved cies for human resource development. The council is in the activities of the council. In short, the lack of a a tripartite body and includes among its members all consolidated vision for the sector and a clear man- the relevant ministries (Education and Higher Edu- date for the council and the absence of an empowered cation, as well as the sectoral ministries in charge of secretariat that can provide the necessary leadership vocational training), representatives from the private have resulted in a de facto situation in which policy sector through the federations of employers, and issues, directives, and key decisions in the TVET sec- trade unions. The SCHRD was established in 2003 tor are increasingly being discussed, processed, and but has never met in session, among other reasons implemented outside the purview of the E-TVET for lack of an executive arm and difficulties in coor- Council. As a result, the mere presence of the three dinating the large number of high-profi le members key TVET suppliers (Ministry of Education, Min- of the board, as well as the creation of parallel struc- istry of Higher Education, and Ministry of Labor) tures with similar functions under other ministries. in one governance structure may not be enough to In 2007–09, the prime minister initiated a process make it functional. Recent efforts on the part of the to reactivate the SCHRD under the leadership of government to coordinate the roles of the three min- the minister of manpower and migration. The newly istries in the council vis-à-vis the private sector may established executive arm of the SCHRD met regu- result in a more successful E-TVET Council. larly during this period to prepare a new agenda for In Tunisia, a strategy to improve business this governing body, which still needs to be decided competitiveness included a reform of professional upon by the prime minister. In the interim, the Min- training through the Program for Upgrading the isterial Committee for Employment and Human Vocational Training and Employment System. To Resources Development (chaired by the minister of promote the involvement of the private sector, the state for military production, who represents the Ministry of Vocational Training and Employment government), the Social Fund for Development, and introduced steering committees to help manage public the National Authority for Quality Assurance and training centers after the reform. The committees Accreditation in Education have begun policy dis- were headed by a private sector representative and cussions on skills development. composed of other private sector stakeholders and In Jordan, an E-TVET (Employment-Technical delegates from the Tunisian Vocational Training and Vocational Education and Training) Council was Agency. These steering committees, however, have set up by law in 2008 as the central element of sec- been heavily criticized, as their role has been merely tor governance to recommend policy directives and consultative. Due to the lack of a detailed regulatory funding that could support priorities leading to high- framework that clarifies the role of private and public employment outcomes. Despite strong expectations, stakeholders, the private sector does not have a true the potential of the institution has remained mostly voice in training centers’ activities, thereby limiting untapped. The institutional setup of the E-TVET its ability to demand and contribute to developing Council and the secretariat accounts for this, as they the skills and competencies it needs. are both perceived as part of the Ministry of Labor Sources: El-Ashmawi 2011; Mornet-Cariou and Rajadel 2011; and World Bank. 178 JOBS FOR SHARED PROSPERITY commitment or employers lack involvement. functioning councils and committees, how- Evidence from focus group discussions with ever, private sector involvement is mostly employers in the West Bank and Gaza and absent or limited. To be sure, in some cases Tunisia14 reveals that employers do not feel the private sector contribution to curricu- particularly represented by employer asso- lum development has been strong, but these ciations (whose funding often depends heav- examples are small scale and often initiated ily on state subsidies), and thus they tend through the personal initiative of a few peo- to develop stronger ties with the state than ple, such as, for example, the cooperation with the private sector they are supposedly between the Jordanian Vocational Training representing. Corporation and the hotel industry. As a result, employers are often unaware The resource allocation and management of private sector involvement in the TVET of publicly run TVET centers often reflect a system at the national level. In Tunisia, strong supply-driven approach, mostly with for example, the interviewed employers the private sector providing mandated financ- knew nothing about the work of the Cen- ing without believing that it receives sufficient tre National de Formation de Formateurs returns. In Tunisia, for example, TVET cen- et d’Ingénierie de Formation (CENAFFIF) ters are fi nanced 95 percent through public with professional federations in the design spending and only 5 percent through student of training courses. In fact, some had never fees and training services purchased by com- heard of training centers’ steering commit- panies. Incentives for centers to complement tees (comités d’établissement). It appears their budget are weak or even negative, as that close collaboration between employers these additional funds are deducted from the and training centers depends mostly on indi- next year’s budget allocation. Incentives to viduals, that is, on personal and professional innovate and respond to demand are limited, relationships more than on institutional as spending allocations are negotiated yearly frameworks. Local employers thus have little based on projected enrollment figures rather say in the type of training courses provided, than on performance. especially for initial vocational education and As for the private sector contribution to training (VET). public funds, in Tunisia, the vocational edu- cation and training tax is the country’s oldest Effect of the disconnect on the effectiveness instrument for fi nancing education, consist- of the TVET system ing of a 2 percent compulsory payroll tax Because of the absence of regular dialogue and amounting to about 90 million Tunisian between the private sector and the educa- dinars (US$60 million) in 2010. The fund tional and training sector, information and was reformed in 2009 because of disap- knowledge failures occur on both sides. The pointing results, with less than 40 percent private sector, for example, has few incentives of eligible businesses having taken part in or opportunities to contribute to curriculum the system. The new vocational training development and fi nancing, little interest in and apprenticeship promotion fund seeks to identifying its precise skill and competency motivate small and medium fi rms to engage needs, and little capacity to provide practical in training. With social partners having training to complement theoretical education. limited say in the management of the fund, As for curriculum development, the responsi- and given the small number of staff and the bility lies with a central department—mostly lack of communication on existing financial the Ministries of Education, Technical and instruments, many fi rms still shy away from Vocational Education, and Higher Educa- providing training (Mornet-Cariou and tion. The national education, TVET, and Rajadel 2011). higher education councils offer a path for The private sector also has little involve- the private sector to contribute and become ment in providing space for practical train- involved. Given the difficulties in establishing ing, either for universities, where there are U N E M P LOYA B L E A N D U N E M P LOYE D: S K I L L S G A P S A N D A M E R I TO C R AC Y D E F I C I T I N M E N A 179 practically no links to the private sector, or and programs. In MENA, diplomas and cer- for technical and vocational education and tificates do not provide information on what training. The development of dual apprentice- students know and are able to do. End-of- ship has been identified as a priority in some program and end-of-cycle examinations and countries’ strategies for TVET,15 for exam- testing are common in the region. However, ple, in Jordan, Tunisia, and the Republic of how well this testing captures the quality Yemen. Setting up dual apprenticeship sys- of the training imparted is questionable. In tems, however, requires the deep involvement some cases, testing is biased toward factual of employers, whose priority is not training, and academic knowledge and does not seek and smaller businesses in particular, which an adequate assessment of job-related skills are short on staff and rarely have a mid- or and employability. Secondary school leaving long-term strategy that would translate into examinations, for example, are designed as an active human resource policy. selective mechanisms for allocating scarce As for the educational and skills develop- further educational opportunities. TVET ment system, problems in the quality of the graduation exams in Egypt are a good illus- evaluation and certification system, weak tration (see box 6.3). Moreover, secondary or nonexistent counseling and orientation school leaving examinations, for exam- services for students, and insufficient or ple, are designed as selective mechanisms no tracking of outcomes and labor market for allocating scarce further educational research lead to information failures. As a opportunities. result, neither the administrators in the sys- tem itself nor those in the institutions can Victims of the information failure make informed decisions, and, at the same Timely and correct information is crucial for time, they cannot be held accountable for enabling prospective and current labor mar- the results achieved and for not using those ket participants to make informed choices. results to improve their policies, institutions, In fact, one could say that “unemployment BOX 6.3 TVET end-of-year examination examples from the Arab Republic of Egypt The following example is drawn from a review of While the practical test amounts to perform- a sample of end-of-year examinations and tests in ing a task, the structure of the test does not allow TVET schools in Egypt. The assessment tool for assessment of whether the student would be able to Enterprise Training usually contains a theoretical perform adequately in a production process. The part and practical part. The theoretical exam con- task is a simulation of a work process without a real tained 15 open questions that called exclusively for job context or a relevant job result. The disconnect reproduction of factual knowledge. For example, with the real world runs even deeper, as there is no one of the questions for shoe manufacturing in the evidence that the equipment used in TVET centers leather sector was to “list three non-leather mate- is similar to that found in the workplace, and it rials used in manufacturing the upper part of the is well known that not all students have access to shoe.” Meanwhile, the practical examination on the equipment for practical training during the year. same level is task oriented: “Make a prototype or As a result, the knowledge and skills tested in these pattern on a sheet of packing paper of about 140 × exams are unlikely to be linked to particular com- 140 cm using the single shoe vamp piece to replicate petencies, and there is no evidence of links to labor the cutting of textile vamp lining. Use the best lay- market requirements or needs. out with correct direction of pieces. Textile lining Source: El-Ashmawi 2011. material could also be used for cutting the pattern.” 180 JOBS FOR SHARED PROSPERITY cannot be reduced in the short and medium in Morocco among adults, parents seem to term without a realignment of expecta- have limited ability to provide advice. Only tions” (World Bank 2008b, 48). However, 9 percent of youth reported making their as a result of these information and signal- choices based on the advice received from a ing failures, both in the private sector and in professional figure outside their family, such the educational and training system, young as a teacher, an NGO operator, or a school people have limited information on which counselor (La Cava et al. 2012). to base their educational choices. In focus The private sector has an important role to group discussions in Morocco, for exam- play, as it needs to send signals to the educa- ple, youth were asked how they made their tional and training system and to students on educational choices following compulsory the skills and competencies demanded. lower-secondary education. Most young The isolation of the private sector from people reported choosing the degree that the educational and training system is also they would enjoy most; less than a quar- manifested in the lack of continued work- ter chose their educational path accord- force training. In contradiction of employers’ ing to what they perceived would be the complaints about skills, the investment in job most useful degree on the market (La Cava training by the private sector in MENA is et al. 2012). This fi nding is striking, given rather low compared to that in other coun- that in all focus groups youth indicated full tries with similar economic development awareness of the labor market situation; they and workforce education. Figure 6.12 illus- seemed to be lacking the tools to be strategic trates the point with data for Algeria, Egypt, about their choices. Data from youth surveys Jordan, Morocco, Oman, Saudi Arabia, show that about 80 percent of young people Syria, the West Bank and Gaza, and the in Morocco reported making their educa- Republic of Yemen.16 On average, only one tional choices completely on their own; the out of four fi rms in the MENA region invests main source of information and advice was in formal training programs. Moreover, firms siblings. Given the very low level of education that invest tend to offer formal training to a relatively small share of the workforce. On average, only 33 percent of workers benefit FIGURE 6.12 Incidence of job training by selected economies’ GDP from these programs. In stark contrast, more than 50 percent of firms in East Asia and the 0.8 Pacific and more than 40 percent of firms in Latin America and the Caribbean invest in Lebanon job training. 0.6 In MENA, based on findings from the % of firms training Enterprise Surveys, larger fi rms are three to 0.4 four times more likely to invest in training Saudi Arabia Algeria than smaller firms; these differences hold Yemen, Rep. West Bank and Gaza Jordan even after controlling for differences in sec- 0.2 Syrian Arab Republic Oman Morocco tor, technological intensity, education of the Egypt, Arab Rep. workforce, and fi rm openness. Evidence also 0 shows that in the MENA region, the degree 10 of education of the workforce and on-the- 4 6 8 job training are strongly complementary: GDP per capita Job training countries (mean) Job training, MENA countries (mean) fi rms with larger shares of skilled workers Fitted values are much more likely to invest in training. In addition, innovation and training are Source: Almeida, Gatti, and Saenz. 2012; calculation using the Enterprise Surveys’ standardized data also complementary, as fi rms adopting and (one wave per country). Note: Figure reports the average share of firms that report investing in job training around the adjusting their technology more frequently world and the country’s level of GDP per capita. GDP = gross domestic product. are more likely to invest in training. Even U N E M P LOYA B L E A N D U N E M P LOYE D: S K I L L S G A P S A N D A M E R I TO C R AC Y D E F I C I T I N M E N A 181 though it is not possible to disentangle such formal guarantees, the ongoing struc- correlation from causality, these findings tural adjustments, and a rising number of strongly suggest that the differences in the graduates, competition for entry into the investment in training across fi rm size are public sector has become fi ercer. A young not solely explained by the greater “net woman in Morocco comments on the cur- benefits” of this investment among larger rent difficulty of moving into public sector fi rms. Most likely, there are also market fail- employment, noting that some graduates ures related to lack of access to credit, lack “want to work only in public jobs. They of information, or a greater worker turn- refuse to look for other options. But gov- over disproportionately affecting smaller ernment capacity recruitment has become employers. limited. The young people cannot count anymore on the government to give them a job. They need to change this old mentality.” Public sector as the main client Students participating in focus group of the educational and training discussions in Egypt and Jordan expressed system their strong desires to continue on to higher The government plays an important role education, mainly because “society appreci- in providing jobs in MENA. On average, ates people with university degrees.” Parents one-fourth of jobs are in the public sector, and students alike perceive “good jobs” as ranging from 4 in 10 positions in Jordan those requiring a university degree; both of and Saudi Arabia to less than 2 in 10 in the these beliefs suggest a “social aspiration” in Islamic Republic of Iran, Morocco, and the MENA. In the words of a secondary student United Arab Emirates (see chapter 5). Most from Jordan, “It is simply not prestigious MENA countries have considerably higher if you hold less than a bachelor’s degree.” employment shares in the public sector than Despite the increasingly insecure and slow countries such as Turkey (15 percent) and transition from education to public employ- Spain (22 percent). In some countries in the ment, aspiration to insider jobs in the public region, structural adjustment of the public sector in many countries of the region has sector has already happened (for example, established a strong incentive for more par- Morocco), while others are in the process of ents and students to invest in and pursue uni- reduction: in Egypt and Qatar, the employ- versity education, particularly women.19 As ment shares in the public sector dropped a result, the educational system has created by 10 percentage points between the early signals for public sector hiring rather than and the late 2000s. In the West Bank and equipping graduates with the employability Gaza, however, the public sector expanded capital needed to succeed in the wider labor over the same period, from 27.4 percent market. in 2000 to 34.9 percent in 2008. Employ- In conclusion, the public sector not ment in the public sector is the preferred only regulates the educational and train- career choice for the majority of youth in ing system—with very little involvement of MENA, with the exception of Lebanon, employers—but also is its main client. This Morocco, and the West Bank and Gaza.17 long-standing relationship creates a histori- Public sector employment is even considered cal inertia that provides little incentive for a “right” in some countries, not least due reform and keeps the educational system to the tradition of job guarantee schemes isolated from the private sector. Such iner- for high school and university graduates tia misguides many young people who are in Egypt and Morocco in the 1960s that still making choices under the assumption encouraged youth to stay in formal educa- that they will get a public sector job. With tion and thus served important social objec- those jobs in increasingly short supply and tives when fi rst established (Salehi-Isfahani with no employability capital in the pri- and Dhillon 2008).18 With the abolition of vate sector, these youth are at risk of not 182 JOBS FOR SHARED PROSPERITY finding a suitable opportunity to realize backgrounds, who tend to perform less well their potential. in early selection (Ireson, Hallam, and Hur- ley 2005; Jakubowski et al. 2010; OECD 2007). One disadvantage of highly stratified A “logic of selection” over a systems is that transitioning from a lower to “logic of learning” a higher track is difficult and thus not com- Secondary and tertiary educational systems in mon. In many MENA countries, tracking MENA are characterized by high selectivity happens relatively early in pupils’ lives, and throughout, starting with early selection and few transition pathways—and viable second rigid tracking by the end of lower-secondary chances—exist (see box 6.4). High selectivity education (grades 8 or 9), and leading to fierce is then compounded by high rigidity, which competition for the most in-demand univer- renders it impossible to reverse tracking deci- sity faculties and programs in the last grade sions made early in students’ lives. For exam- of upper-secondary education. As a result, ple, in Egypt, only 5 percent of graduates children and young people are forced into from technical secondary schools transitioned learning strategies that increase their chances into postsecondary vocational education in of succeeding in the high-stakes examinations 2008–09, down from 8 percent in 2003–04. that are crucial to gaining access to preferred The remaining graduates transitioned straight options in higher education. With the “logic into the labor market (El-Ashmawi 2011). of selection” prevailing over the “logic of learning” in the educational system, students, National examinations as a high-stakes parents, and educators in MENA focus more screening mechanism on examination scores than on the acquisi- tion of relevant skills (aptly named the “scores As a result of the expansion of secondary versus skills debate”).20 education and the increase in the demand for Early tracking and streaming 21 have tertiary education, the secondary education negative consequences on subsequent graduation exam, such as the baccalaure- educational and labor market outcomes, par- ate in Algeria, Djibouti, Lebanon, Morocco, ticularly for pupils from lower socioeconomic Syria, and Tunisia; the Tawjihi in Jordan and BOX 6.4 Transition paths in the Tunisian educational and training system In Tunisia, vocational education and training start opted for vocational education and training in 2009. in the eighth grade and henceforth offer virtually no The latter is provided by public or private training crossover to the general education track. Starting in centers. Students follow a two-year course to com- the eighth grade, after seven years of basic education, plete a professional certifi cate. The overwhelming students can opt for the vocational education system. majority of students (about 80 percent, according During the fi rst two years, they are theoretically free to expert opinion) holding a professional certificate to switch from the general track to the vocational enter the labor market. A small proportion opts for one and vice versa. In practice however, transitions an additional two-year course to obtain a techni- occur only from general to vocational education. cian certificate. Finally, students holding a technician Students unable to complete the ninth grade of basic certificate can take a competitive entrance exam to vocational education can pass a competency certifi- apply for another two-year program to complete a cate, a short course of 6 to 12 months validating one specialized technician certificate. Their average suc- specifi c skill. An estimated 15 percent of students cess rate on these competitive exams is low (about who attained the ninth grade, general or vocational, 7 percent, according to expert opinion). Graduates (continued next page) U N E M P LOYA B L E A N D U N E M P LOYE D: S K I L L S G A P S A N D A M E R I TO C R AC Y D E F I C I T I N M E N A 183 BOX 6.4 Transition paths in the Tunisian educational and training system (continued) FIGURE B6.4.1 Transition paths in the Tunisian educational and training system education Higher education Universities, Technical bachelor’s BTS Higher VET engineering degree (ISET) 2nd Year schools Professional 1st Year bachelor’s degree Competitive- entrance exam Labor market General Technical baccalaureate baccalaureate BTP VET 2nd Year 4th Year 1st Year 3rd Year Secondary education School guidance CAP VET 2nd Year 2nd Year 1st Year 1st Year General basic education Vocational basic certificate education certificate basic education Compulsory 9th grade - 9th grade - General basic education Vocational basic education CC 8th grade - 8th grade - General basic education Vocational basic education 7th grade - Basic education Source: Adapted from Toumi 2009. Note: CC = competency certificate; BTP = technician certificate; CAP = professional certificate; BTS = specialized technician certificate; ISET = institute for higher technological studies; VET = vocational education training. with a technician certificate are in direct competition technical or professional degrees. Figure B6.4.1 sum- with students holding a technical baccalaureate. The marized this discussion. latter also have the option of going into the higher Source: Mornet-Cariou and Rajadel 2011. education system but are then more likely to choose the West Bank and Gaza; and the Thanawiya Despite increases in enrollment, graduation amma in Egypt, Oman, Qatar, and Saudi from upper-secondary education is chal- Arabia, has become a key defi ning moment lenging, and students who fail the exami- in a student’s life. These exams also serve nation leave with few relevant skills and no as entry examination for higher education. marketable diploma. A secondary student in 184 JOBS FOR SHARED PROSPERITY Jordan put it this way, “Tawjihi is a destiny Following a long tradition, these national determinant.” public examinations not only condition the These national examinations share three wider social perception of what counts as fundamental traits: (1) they serve the double quality of education but, in doing so, also function of graduation and university access, essentially narrow the number of “avenues of so that failing the exam means that the stu- success” for students in the educational sys- dent does not graduate from upper-secondary tem: the score on the examination is there- education; (2) they are used as a selection fore perceived as the only measure of success. mechanism for determining access to univer- Thus, the combined effect of early track- sity in general and to the most sought-after ing with a high-stakes, examination-based institutions and fields of study;22 and (3) they upper-secondary educational system is the are themselves a highly influential institution exclusion of a large proportion of students in shaping teaching and learning practices from educational opportunities and acquisi- in secondary schools. As such, they are key tion of relevant skills and competencies. In factors in shaping school curricula (Eckstein other words, a “logic of selection” prevails and Noah 1996; Moreno 2006). over a “logic of learning” in educational and The primary objective of these high- training systems. At the same time, this sys- stakes examination systems is to identify tem continues to sustain the wedge between top academic performers—by measuring the what employers and other stakeholders acquisition of facts and knowledge through expect from the educational system (skills) memorization rather than through criti- and the wider social perception of what really cal and independent thinking or problem- counts (scores). solving skills—and to give them access to the most sought-after institutions and pro- Equity challenge in higher education grams in higher education. To get there, In spite of rapid enrollment growth through- not only do students and teachers spend out the MENA region, tertiary education time preparing for these examinations at remains largely elitist, with strong dispari- the expense of developing a broader range ties in access and success persisting in high-, of skills and competencies (Bray 2009), but middle-, and low-income Arab countries. In also families in some countries spend con- Egypt, for example, the latest household sur- siderable sums on private tutors. In Egypt, vey indicates that the probability of access for example, 40 percent of young people to tertiary education is eight times higher report having received private tutoring for children from the richest quintile than after school (based on the Survey of Young for those from the lowest quintile. Inequal- People in Egypt 2009; see appendix). Par- ity in tertiary education is, to a large extent, ents invest heavily in private tutoring not an extension of inequality at lower levels because they believe that schools are low of education, reflecting structural barriers quality but also because the game at stake (income, ethnicity, gender, language, culture, is “cut-throat selective competition” for religion, disability, and the like) and affect- scarce places in universities and programs ing the economic and social opportunities in high demand. Because only a small frac- of many talented and capable young people tion of students excel in these examina- (Salmi and Malee Bassett, forthcoming). tions, stakeholders perceive the system as The main additional barriers at the ter- academically sound and rigorous. Overall, tiary level are of two kinds, financial and however, due to the structure and allure of nonfi nancial. The latter include inadequate university-tracked education, students and information, motivation, academic prepara- their parents end up spending time, energy, tion, and social capital. In particular, to cope and money acquiring knowledge that is not with the demographic pressure on the systems valued by private sector employers (Dhillon and maintain fiscal affordability, access to et al. 2008). some of the faculties was limited and made U N E M P LOYA B L E A N D U N E M P LOYE D: S K I L L S G A P S A N D A M E R I TO C R AC Y D E F I C I T I N M E N A 185 conditional on grades from the secondary TVET as a low-quality option school leaving exam. Quotas were adopted Tracking can be particularly detrimental for the majority of science faculties—the when the option for those tracked into tech- ones that are more expensive to run, as they nical training produces low-quality skills. require functioning laboratories and other In fact, nowhere are problems of relevance equipment—but also the ones that are more more severe than in TVET, which is meant relevant to private sector needs. With high- to be the part of the system most quickly and stakes examinations and prevalent tutoring, directly geared to employability. According the access to faculties that would be more to the evidence, over the past decade most likely to lead to high-paying jobs in the pri- countries in the region have seen a decrease vate sector is thus highly skewed toward in student demand for TVET, despite the young people from more affluent families. fact that demand for upper-secondary and The evidence suggests that hiring for tertiary education increased dramatically higher-paying private sector jobs in larger during the period. Overall, student enroll- firms is relatively more meritocratic than hir- ment in technical or vocational secondary ing in smaller and informal fi rms and thus education as a proportion of total secondary in principle is more open to people from education varies considerably across MENA all social strata. However, since the access countries, from less than 5 percent in Kuwait, to the faculties that are more conducive to Qatar, and the West Bank and Gaza to about these jobs is still highly correlated with fam- 20 percent in Egypt (figure 6.13a). Enroll- ily background, opportunities to access more ment figures have been decreasing drastically desirable jobs outside the public sector con- in countries such as Bahrain, Egypt, Jordan, tinue to be inequitably distributed. and Syria. The decreasing demand for TVET FIGURE 6.13 Enrollments in technical and vocational training in MENA and worldwide, 1999, 2000, and 2009 a. Middle East and North Africa b. World Egypt, Arab Rep. Arab states Tunisia Central and Eastern Europe Bahrain East Asia and Pacific Iran, Islamic Rep. Latin America and the Djibouti Caribbean Jordan North America and Western Europe Syrian Arab Republic South and West Asia Kuwait Sub-Saharan Africa West Bank and Gaza Qatar World 0 5 10 15 20 25 30 0 10 20 30 40 50 60 % of secondary enrollment % of secondary enrollment 2000 2009 1999 2009 Source: World Bank Edstats; UNESCO 2011. 186 JOBS FOR SHARED PROSPERITY is a worldwide phenomenon but is particu- The second transition: From larly dramatic in the Arab states, which have employability to employment seen a decrease of 14 percentage points from 1999 to 2009: that is, from 34 percent to Even if we acquire the right skills, even 20 percent of total secondary education if new jobs are created, what does it enrollment (figure 6.13b). While the share of matter if I do not carry the “right” TVET enrollments has also fallen across the family name? My less-talented class- board in all other world regions—with the mate with the connections will end up exception of Sub-Saharan Africa—the reduc- landing the job anyway. tion in MENA is by far the most dramatic Young woman, Morocco of all. Young people in MENA are avoiding In M ENA , being employable is not what is perceived as an unreformed low- enoug h for yout h — a nd out sider s i n quality option that is visibly associated general—to transition from education to with academic failure in the context of the work. Instead, youth need to succeed in a highly selective logic described above. 23 “double transition”: in addition to obtain- Some employers in focus group discussions ing the competencies and credentials to be in Egypt even voiced preferences for hiring employable, they need to position them- young nondiploma holders who have not selves in the labor market. This section gone through the technical secondary sys- analyzes the second part of the “double tem. 24 Hence, student demand increasingly transition” from education to work— concentrates on higher education, with the namely, from employability to work—to result that the overwhelming majority of characterize job search and hiring prac- young people in MENA have only one valid tices and how they relate to the structure avenue for success within the educational of the labor markets in MENA and likely and training system: the academic and highly reinforce segmentations in those markets. selective upper-secondary education leading The issues that matter for graduates’ sec- to the high-stakes graduation and university ond transition are whether employabil- entrance examination. ity (skills and competencies) counts when Students going into TVET are normally employers make hiring decisions, whether those excluded from secondary general edu- education (credentials and qualifi cations) cation by the system of examinations and matters to employers, and whether hiring marks. In addition, students in TVET pro- is transparent and based on merit. grams are often from lower-income back- The question of who gets hired into which grounds, such as in Jordan, where 95 percent type of jobs is crucial to understanding pat- of those pursuing the academic secondary terns of exclusion in the labor market. Many track come from middle- and upper-income factors can affect the process of search and backgrounds (Dhillon et al. 2008). Once in matching. the VET system, graduates have a low social In labor markets, there is usually an asym- status that is very strongly felt, for example, metry of information between the job appli- in the Egyptian social system. 25 There are cant and the hiring employer. The applicant exceptions on a small scale, such as graduates may know his or her level of ability, but the of the Don Bosco Technical Institute or the employer cannot observe it and uses certain Mubarak-Kohl-Initiative in Egypt. Despite “signals” to make inferences about an appli- the relatively higher quality of these students cant’s ability to perform a certain job (Spence and graduates and the early workplace expo- 1973). Prospective employers might lack the sure they receive, these graduates still repre- information necessary to judge the quality sent a very small fraction (1–2 percent) of the of graduates effectively. First-time job seek- new entrants to the labor market every year ers typically rely on the signaling value of (El-Ashmawi 2011). diplomas and credentials, as they often lack U N E M P LOYA B L E A N D U N E M P LOYE D: S K I L L S G A P S A N D A M E R I TO C R AC Y D E F I C I T I N M E N A 187 previous job experience or professional net- because the employment statuses of indi- works. With educational systems that have viduals within a network are correlated served predominantly as screening devices (Calvo-Armengol and Jackson 2004). for access to insider positions in the public A relatively large share of employees finds sector, the signaling value of diplomas for jobs through informal networks. According the private sector in MENA is often limited. to recent studies in the United States and In a competitive labor market, fi rms decide Germany, around 44 percent of workers in the number and quality of employees to hire the United States (Franzen and Hangartner to maximize profits. However, limited com- 2006) and 34 percent in Germany (Caliendo, petition in product markets—as observed in Schmidl, and Uhlendorff 2010) found their MENA—is likely to decrease fi rms’ demand jobs through informal social networks. for mechanisms that resolve these informa- Survey data show that the importance of tion asymmetries, both the signaling value of informal channels for matching and recruit- credentials and the public and private labor ment is particularly high in MENA, where market intermediation services. 80 percent of private sector employees report The prevalence of job search through having found their job through friends or informal channels and its impact on the labor relatives in Lebanon, 26 66 percent in Syria, market will be explored in the following 53 percent in Iraq, and 40 percent in the sections. Republic of Yemen—compared with, for example, 33 percent in Turkey (figure 6.14). For the vast majority of the unemployed Importance of informal channels in MENA, informal networks constitute The importance of networks and connections important channels of job search. Figure 6.15 for job matching is well established in the liter- shows that in places with very small private ature. Personal networks—including friends, sectors, such as Iraq and the West Bank and relatives, colleagues, or acquaintances—work Gaza, virtually all the unemployed rely on well for job seekers and hiring firms alike friends, family, or their own contacts with because of information flows in an envi- ronment of shared social capital and trust (Calvo-Armengol and Jackson 2004; Mont- FIGURE 6.14 Percentage of the employed who gomery 1991; Mortensen and Vishwanath found their job through friends or relatives in the 1994). Granovetter (2005) finds that rely- private sector in selected countries in MENA, 2005–10 ing on existing social contacts and networks is less costly than a more formal search 90 80.2 through job intermediaries. This observa- 80 tion is particularly important for individuals 70 65.6 with less education—often from less affluent 60 52.9 Percent 50 40.3 neighborhoods—who rely more extensively on 40 33.3 informal contacts for employment, whereas the 30 better educated use a mixture of contacts and 20 10 formal search methods (Ioannides and Loury 0 2004; Whaba and Zenou 2005). n ic q . ey ep no Ira bl rk ,R pu These preexisting networks and their ba Tu en Re Le m unequal distribution do, however, introduce b Ye ra nA inequalities into the labor market and poten- ria Sy tially reinforce divides. Evidence suggests a link between the quality of the network Source: Based on Iraq’s Household Socioeconomic Survey (HSES) 2006–07, of professional contacts and an individual’s Lebanon’s Employer-Employee Survey (EES) 2011, the Syrian Arab Repub- lic’s Employer-Employee Survey (EES) 2009, Turkey’s Labor Force Survey transition into the labor market (Margolis (LFS) 2010, and the Republic of Yemen’s Household Budget Survey (HBS) and Simonnet 2002; Montgomery 1991), 2005–06. See the appendix for more information. 188 JOBS FOR SHARED PROSPERITY FIGURE 6.15 Percentage of the unemployed FIGURE 6.16 Percentage of the unemployed using informal networks in their job search in using friends or relatives in their job search selected economies in MENA, 2005–10 in selected economies in MENA, by level of education, 2005–10 United Arab Emirates 47 West Bank and Egypt, Arab Rep. 57 Gazaa United Arab Lebanona 57 Emirates Jordan 63 Tunisia Yemen, Rep. 68 Egypt, Arab Rep. Morocco 69 Lebanona West Bank and Gaza 86 Syrian Arab Republic 86 Jordan Iraq 91 Yemen, Rep. 0 10 20 30 40 50 60 70 80 90 100 Percent Morocco Friends/relatives Contacts with employers Iraq 0 10 20 30 40 50 60 70 80 90 100 Source: Based on Egypt’s Labor Market Panel Survey (LMPS) 2006, Iraq’s HSES 2006–07, Jordan’s Labor Market Panel Survey (LMPS) 2010, Lebanon’s Percent Employer-Employee Survey (EES) 2011, Morocco’s Household and Youth Primary Secondary Tertiary Survey (HYS) 2010, Syria’s EES 2009, Turkey’s LFS 2010, the United Arab Emirates’ Labor Force Survey (LFS) 2009, the West Bank and Gaza’s Labor Force Survey (LFS) 2008, and the Republic of Yemen’s HBS 2005–06. See Source: Based on Iraq’s HSES 2006–07, Jordan’s LMPS 2010, Lebanon’s EES the appendix for more information on these surveys. 2011, Morocco’s HYS 2009, the United Arab Emirates’ LFS 2009, the West a. Mutually exclusive answers. Bank and Gaza’s LFS 2008, and the Republic of Yemen’s HBS 2005–06. See the appendix for more information on these surveys. a. Mutually exclusive answers. employers during the job search. To a lesser FIGURE 6.17 Ways university graduates found extent, the same is true in countries with public and private sector jobs in Tunisia, 2005 more employment in agriculture, such as Morocco and the Republic of Yemen. Othera 2.0 45.6 Consistent with the literature, the intensity Educational 3.1 of using networks differs between lower- and institution 7.3 higher-skilled job seekers (figure 6.16). In Newspaper 1.2 7.2 Morocco, for example, 74 percent of those Local authorities 0.6 12.3 with primary education report using networks 17.9 Employment offices 8.6 in their job search, in contrast to 47 percent Contacted firm 37.0 among those with tertiary education. This 17.3 fi nding is consistent with a relatively higher Friends or relatives 32.2 7.8 prevalence of public sector employment (and 0 10 20 30 40 50 the use of concours —that is, the competi- Percent tive entrance examination—for employment Private Public admission) among skilled workers. Data from Tunisian university graduates (figure 6.17) Source: World Bank and Ministère de l’Emploi et de l’Insertion Profession- nelle des Jeunes 2005. support this hypothesis, indicating a signifi- a. Mostly concours (competitive entry examinations) for entering public cantly lower use of informal networks among sector. U N E M P LOYA B L E A N D U N E M P LOYE D: S K I L L S G A P S A N D A M E R I TO C R AC Y D E F I C I T I N M E N A 189 public sector employees. Some evidence also larger fi rms are more likely to use formal suggests that parental employment and con- channels in addition to informal ones. In nections have had increasing importance in Jordan, for example, about 30 to 40 percent securing jobs in the formal sector, particu- of employees in larger fi rms (with 50 and larly for skilled workers and especially in the more employees) report having found their face of declining job opportunities in the pub- job through informal channels, compared lic sector (Binzel 2011; Pierre and Robalino with 45 percent among small and medium 2012; Gatti et al. 2012). In particular, Binzel fi rms (10–24 and 25–49 employees, respec- argues that the gradual suspension of the pub- tively) and 60 percent among micro or small lic employment guarantee scheme in Egypt fi rms with 5–9 employees (figure 6.18). adversely affected intergenerational mobility Firms’ use of specific recruitment channels among the increasingly well-educated youth, (formal as opposed to informal) seems to be as access to desirable jobs depends more and correlated with the importance of creden- more on personal connection and parental tials in the job-matching process. In MENA, background. smaller firms attach greater value to personal references and contacts and therefore “trust” when recruiting, while larger firms attach Employers and informal hiring greater value to credentials (Pierre and networks Robalino 2012; see also box 6.5). In inter- Evidence on employers’ search and recruit- views, when asked whether they believe that ment practices is limited. Firms’ character- hiring is done in a meritocratic way, private istics as well as the type of vacancy to be recruitment agencies and headhunters28 in fi lled affect the choice of recruitment chan- Jordan and Lebanon unanimously stated nel (Sabatier 2010). Evidence from interna- that this was not the case (box 6.6). Recruit- tional literature suggests that fi rms seeking ers agree that “usually and most of the time” qualified staff are more likely to use adver- hiring is based on personal contacts, as most tisements to fi ll vacancies—especially when of their clients are small or medium enter- competition between job seekers is low— prises. One recruiter believes that there have whereas employers seeking to hire in the been considerable improvements in the hir- second segment of the labor market (that is, ing process, especially for medium and large fi xed-term, part-time, or low-skill jobs) are enterprises, but because small fi rms cannot more likely to use local employment agencies afford to search for candidates, they hire (Gorter, Nijkamp, and Rietveld 1993, 1996; Russo, Hassink, and Gorter 2005). Findings from France suggest that public and private FIGURE 6.18 Importance of personal networks in finding a job in institutional intermediaries are overall more Jordan, by firm size, 2010 effective than personal and professional net- works, with private agencies being more effi- 100 cient at fi lling skilled vacancies and public 90 agencies more efficient at fi lling nonskilled 80 70 vacancies (Sabatier 2010).27 60 32 39 Percent 46 44 The choice of recruitment channel also 50 61 depends on the employer’s capacity to use 40 50 30 a certain channel. Using data from the 20 Netherlands from 1995 to 1999, Russo, 10 Hassink, and Gorter (2005) fi nd that for- 0 1–4 5–9 10–24 25–49 50–99 100+ mal searches cost seven times as much as Visiting institutions and workplaces Relatives, friends, family informal ones. Not surprisingly, fi rm size Setting up own business Other matters, as smaller firms tend to favor personal contacts in recruitment, whereas Source: Based on Jordan’s LMPS 2010. See the appendix for more information on the survey. 190 JOBS FOR SHARED PROSPERITY BOX 6.5 Employers’ hiring practices in Jordan and the Arab Republic of Egypt Jordan: The importance of wasta ripple effect is an often-neglected issue in labor I applied to the phone office, they invited me for an economics. interview but they told me that people pay to get hired. Maybe they hire employees’ children. Egypt: “Sons of Employees” hiring scheme Egyptian labor law stipulates that all citizens have a Young woman, Egypt right to access jobs based on merit in an equal and In a study of the new social security law in equitable manner. However, in some public profes- Jordan, employers, employees, and key stakeholders sions, such as university professorships, police work, describe the varying, and often contradictory, and lower levels of the judiciary, sons of employ- hiring practices of Jordanian employers. There ees are given preference. In other cases, the son of is increasing focus on the merits of the potential a person who is about to retire or of someone who employee—on experience and on the knowledge has reached 55 years of age is hired. This practice is and skills related to the job for which he or she practically nonexistent in the private sector and is is applying—as well as on the attitude of the formally discouraged by the Egyptian government, applicant and the confidence that the employer which has issued regulations to curtail the hiring of has in the applicant’s potential as a dedicated and sons of existing employees in government factories responsible employee. However, the customs of and other public offi ces. Nevertheless, the practice favoritism and wasta a continue to be practiced, persists. Indeed, there are documented complaints and cultural challenges have not substantially to government ministries from aggrieved family abated. Gender discrimination and inequality are members who have not been hired, even though not uncommon, although such discrimination may their mother or father served in the ministry for have shifted over time: veiled women may not be many years. Such petitioners believe that because hired, for example, and employers may hire women their parents have worked in the office, they have the over men because women are willing to work for right to a job themselves. a lower salary. The latter is a troublesome trend: Sources: Brodmann et al. 2012b; World Bank 2013. while more women may be employed under these a. Wasta translates as the use of personal connections to get things done or, in circumstances, their work is undervalued and the other words, to accomplish things through favoritism. BOX 6.6 Hiring criteria of headhunters and private recruitment firms Recruitment for management and executive positions job. Figure B6.6.1 shows the frequency of responses is often performed by specialized headhunters or pri- of three agencies in Lebanon and one in Jordan vate recruiters. In phone interviews with such agencies (the response ranked fi rst got three points; the one in Jordan and Lebanon, the human resource special- ranked second, two points; and the one ranked third, ists and recruitment managers were asked a number one point). of questions related to the recruitment process.a Do candidates’ diplomas and academic Which signal is most important in making credentials reflect their skills and hiring decisions? competencies? Recruiters were asked to rank the three most All of the interviewed recruiters stressed that important indicators in signaling whether a diplomas give an indication of candidates’ hard candidate who is a fresh graduate is right for the skills but that soft skills are lacking and not (continued next page) U N E M P LOYA B L E A N D U N E M P LOYE D: S K I L L S G A P S A N D A M E R I TO C R AC Y D E F I C I T I N M E N A 191 BOX 6.6 Hiring criteria of headhunters and private recruitment firms (continued) FIGURE B6.6.1 Signals important for hiring Professional experience (previous jobs) and proof of performance in those jobs Professional letters of recommendation Personal and family references Tests and/or interviews conducted by your own human resources staff Visible outputs: products of the candidate's work (publications, contracts, reports, presentations, etc) Schools and universities where the candidate has studied Diplomas and other academic credentials 0 2 4 6 8 10 Source: Phone interview with recruiters, February 2012. reflected in credentials. Evidence from a recent required personal characteristics; (3) applicants employer-employee survey in Lebanon suggests lacked required job-related skills; (4) applicants that skill gaps are in fact larger for managers than lacked requ i red p erson a l it y t ra it s; a nd (5) for those in nonmanagerial positions, with mis- applicants expected higher wages than firm can matches being largest in the following three areas offer. Overall, the number-one reason named was for managerial staff: noncomputer office skills, applicants’ wage expectations, closely followed by computer skills, and ability to work independently lacking personality traits and lacking job-related (Pierre and Robalino 2012). One recruiter says, experience. According to one recruiter, “Fresh “There is a preconceived idea that students from graduates are not prepared for the job market, they top universities—such as the American University do not know how to prepare a CV, have the right in Beirut—tend to have better soft skills than those attitude in a job interview, etc.” from other universities.” Do you think hiring is done in a meritocratic What are the three most important problems way? encountered when hiring fresh graduates? All recruiters unanimously said that hiring is not Recruiters were asked to name the three most done in a meritocratic way. important problems from the following list that a. The phone interviews were based on a standardized questionnaire with recruiters face when hiring fresh g raduates: open-ended questions and asked of four agencies. (1) number of applicants; (2) applicants lacked 192 JOBS FOR SHARED PROSPERITY individuals with whom they have a close are more exposed to competition: that is, relationship. where fi nding good matches matters most. Overall, the evidence for MENA corrobo- This evidence, though still limited and quali- rates the hypothesis that firms operating in a tative, is consistent with empirical fi ndings more competitive environment rely on more by Heinze and Wolf (2010) in Germany indi- formal job search mechanisms and offer cating that wage discrimination margins are better employment conditions. In Tunisia, lower in sectors like exports, which are more for example, data from the Tunisia Higher exposed to competition. Education Graduate Tracer Survey show that university graduates who found their jobs through formal and thus more competi- A meritocracy deficit? tive mechanisms earn on average 10 percent The increasing demand for transparency more than those who found jobs through and equality of opportunity in accessing friends or relatives. Moreover, interviews jobs makes the importance of meritocracy conducted in Morocco with headhunters and clear rules in job search and hiring even serving high-technology clusters confi rmed more salient. Young people and their fami- that the search for talent tends to be more lies have high expectations for the future, meritocratic in those sectors and fi rms that they invest heavily in education and skills, and they expect these investments to pay off eventually. Yet among students, graduates, and employers in MENA countries, there is a widespread perception that education cre- FIGURE 6.19 Perceptions of youth about constraints to getting a dentials serve a minor role in employers’ hir- job in selected economies in MENA, 2009 ing decisions. Educated youth in MENA have received a clear message from the labor mar- ket: to access one of the few insider jobs, you must wait your turn or already belong to an Libya insider family. Morocco Qatar Tunisia Factors beyond individual control Egypt, Arab Rep. The perception that connections are critical Kuwait undermines the sense of dignity. In recent Saudi Arabia opinion polls, young people in MENA voice Djibouti deep concerns about not succeeding in the United Arab Emirates second transition because of factors they feel West Bank and Gaza are beyond their control, with the lack of job Jordan opportunities and the meritocracy deficit per- Lebanon ceived as greater constraints than the lack of Syrian Arab Republic Algeria training received. Yemen, Rep. Figure 6.19 shows how young people per- Iraq ceive the main constraints to getting a job and Bahrain starting a family. In Egypt and Tunisia and, 0 20 40 60 80 100 to a lesser extent, Syria, the lack of good jobs Percent is the predominant constraint. In all coun- Jobs given based No good jobs tries, the fact that available jobs are thought on connections available to be accessible only through connections is Lack of training Other seen as a source of concern. Interestingly, lack of training is named a prominent constraint only in Morocco (28 percent), Djibouti Source: Based on Gallup World Poll 2009. See the appendix for more information on the poll. (23 percent), and countries in the Gulf Coop- Note: MENA = Middle East and North Africa. eration Council (GCC) (7–16 percent). U N E M P LOYA B L E A N D U N E M P LOYE D: S K I L L S G A P S A N D A M E R I TO C R AC Y D E F I C I T I N M E N A 193 The prevalent use of nonformal job- FIGURE 6.20 World scores on proxies of matching methods is consistent with a num- meritocracy in hiring, 2011 ber of explanations. Firms might have little or no incentive to invest in searching over wide pools of talent for several reasons: the struc- MENA, non-GCC 3.5 ture of product markets29 —little competi- countries tion, distribution of fi rms by size—but also Europe and 3.8 by the pervasiveness of arbitrary enforcement Central Asia of rules, which makes other qualities, such as trust, more relevant. Also, hiring can be risky, South Asia 4.0 especially if it is difficult to fire employees or if they require costly training. Improving com- Africa 4.1 petition and transparency is likely to increase incentives to fi nd better matches. However, Latin America and 4.2 the Caribbean little infrastructure and regulation are in place to increase information flows. To some extent, East Asia and 4.5 this void reflects the degree of development Pacific and features of the labor markets in MENA, MENA, 4.9 for example, high informal employment. GCC countries Informal networks and information and OECD countries 5.3 signaling failures 0 1 2 3 4 5 6 The widespread use of informal networks is a sign of information and signaling failures. These information asymmetries can be over- Source: Based on Executive Opinion Survey 2011. Note: GCC = Gulf Cooperation Council; OECD = Organisation for Economic come to some extent through intermediation Co-operation and Development; MENA = Middle East and North Africa. See services that increase the pool of available WEF and OECD (2011, 75) for a detailed description of the Global Executive Survey. Most questions in the survey ask respondents to evaluate, on a scale workers and jobs. Another advantage of such of one to seven, one particular aspect of their operating environment. One formal job-matching mechanisms is that they represents the worst possible situation; seven represents the best. reduce the perception of the influence of net- works and connections. Comparative data on FIGURE 6.21 Employer-reported skill shortages versus reliance the extent to which hiring is done in a trans- on professional recruitment in selected countries in MENA and in parent and meritocratic way is not available countries outside the region, 2006–11 per se. Firms’ self-reports on how much they rely on professional management in making 80 hiring decisions rather than on families and friends can be used as a proxy. Non-GCC Syrian Arab Republic Index, skill shortages 60 Lebanon MENA countries show the lowest scores of Egypt, Arab Rep. all world regions on this indicator of meritoc- Algeria 40 racy in hiring (figure 6.20). Jordan Morocco The relatively high use of informal Yemen, Rep. recruitment mechanisms of fi rms in MENA 20 stands in sharp contrast to the perceived skill mismatches reported by employers. 0 Figure 6.21 plots employers’ reports of skill 2 3 4 5 6 shortages (from the Enterprise Survey) as Index, reliance on professional recruiters opposed to the extent to which hiring deci- MENA countries Non-MENA countries sions in MENA are based on professional Fitted values, all countries management (from the Global Executive Source: Based on the Global Executive Survey and Enterprise Survey. Survey). Overall, the higher the reliance Note: MENA = Middle East and North Africa. The reliance on professional management data ( x-axis) on professional recruitment, the lower the is for year 2011. 194 JOBS FOR SHARED PROSPERITY reported level of skills shortages—a find- with job seekers, and they even provide assis- ing that supports the hypothesis that rely- tance in the screening and selection process. ing on a limited pool of suitable candidates For job seekers, they provide an overview of increases the probability of not fi nding the open vacancies, and they can also assist in best fit for the job in skills and competencies. preparing for interviews or in updating quali- The intense use of informal recruitment fications to increase the chances of a good job mechanisms devalues the education creden- match. In many countries, employment agen- tials of job seekers, de facto rendering the cies work in close cooperation with social employability of graduates insufficient for benefit organizations, social partners, pri- succeeding in the transition from education vate service providers, temporary work agen- to work. Although the quantitative data are cies, and educational institutions, as well as suggestive and not defi nitive, the data from with other public organizations and NGOs interviews and focus groups are stagger- (Andersen et al. 2009). ing in their uniform belief that employers Public employment services have been make their hiring decisions on the basis of established only recently in countries such as criteria that have little to do with the capi- Jordan (2006), Morocco (2001), and Syria tal of employability brought by candidates. (2001) (table 6.1). Private intermediation, Although this area requires further research however, is forbidden in some countries, such and data, the tentative conclusion drawn is as Tunisia, and has only recently been legal- that hiring in MENA suffers from a “meri- ized in Syria. The efficiency of PES is low tocracy deficit.” in many MENA countries (Angel-Urdinola, Kuddo, and Semlali 2013). Cost-effective counseling and labor intermediation pro- Role of public employment grams, such as job search skills, training services in job matching programs, career and job counseling, job Public and private employment agencies can clubs, job vacancy fairs, employer contact play an important role in the job-matching (intermediation) services, and the like, hardly process. For employers, they facilitate contacts exist in the region.30 In terms of effectiveness, TABLE 6.1 Public employment services in selected countries in MENA, 2009 Number of Legal name of national PES, Number of private and/or the ministry responsible Year of Number of NGOs providing employment Country for employment services establishment regional offices services agencies Egypt, Arab Ministry of Manpower and Migration 1961 307 3 54 Rep. Jordan Department of Employment and Training under 2006 14 — 45 the Ministry of Labor Lebanon National Employment Office, Ministry of Labor 1977 3 4 — Morocco National Agency for the Promotion of 2001 74 — — Employment and Competence (ANAPEC), Ministry of Employment and Vocational Training Syrian Arab Central Nomination Unit at the Directorate of 2001 28 4 Legalized in 2010 Republic Labor, Ministry of Social Affairs and Labor Tunisia Independent National Agency for Employment 1993 91 2 Illegal and Labor (ANETI), Ministry of Vocational Training and Employment Yemen, Rep. Ministry of Social Affairs and Labor 1996 20 150 — Source: Angel-Urdinola, Kuddo, and Semlali 2013. Note: — = not available; MENA = Middle East and North Africa; NGO = nongovernmental organization; PES = public employment service. U N E M P LOYA B L E A N D U N E M P LOYE D: S K I L L S G A P S A N D A M E R I TO C R AC Y D E F I C I T I N M E N A 195 only 0.3 and 2.8 percent of those employed in Role of active labor market Lebanon and the Republic of Yemen, respec- programs in facilitating the tively, report having found a job through transition to work public or private employment services, while the corresponding number is between 30 and Active labor market programs (ALMPs) 50 percent in Egypt and Iraq. can be an effective strategy for helping An important factor contributing to the youth master the transition from education success of intermediation services is the insti- to work. The type of intervention depends tutional capacity of national employment on the specific constraints. For example, services, including the network of offices, training to overcome skills deficits can yield the legal framework in which they operate, promising results; wage subsidies or labor- and especially the number and professional intensive public works programs can be use- level of the staff at local employment offices. ful in an environment with limited demand; Available data show wide variations in levels skill certifi cation programs can be useful of staffing among surveyed MENA countries when workers have the rights skills but (table 6.2). In 2009, a very high staff caseload face difficulties in communicating them to was reported in Syria, exceeding 14,000 reg- potential employers; and entrepreneurship istered job seekers per a single PES staff mem- programs or microfi nance have been used ber. A high staff caseload does not allow to overcome constraints to business start- employment services to deliver personalized up (see, for example, World Bank 2010b). job intermediation. Overall, PES in MENA Publicly provided ALMPs in MENA face a predominantly follow a passive approach, as number of shortcomings in program design, they expect employers to post open vacan- targeting, skill provision, certification, and cies and few resources are put into market- assessment. Angel-Urdinola, Kuddo, and ing their services. The ratio of registered Semlali (2013) summarize the findings of an job seekers per one vacancy in Egypt and ALMP inventory collected through face-to- Lebanon (Beirut only) is comparable to inter- face interviews in Egypt, Jordan, Lebanon, national standards but is relatively high in Morocco, Syria, Tunisia, and the Republic Jordan and Morocco. Finally, job placement of Yemen between 2010 and 2011. Simi- rates are not sufficient to absorb the unem- larly, Angel-Urdinola, Semlali, and Brod- ployed. In Egypt, less than 5 percent of the mann (2010) review nonpublicly provided registered job seekers are employed every ALMPs from Algeria, Egypt, Jordan, Leba- month, in Jordan less than 3 percent, and in non, Morocco, Syria, Tunisia, the West Bank Morocco, about 1 percent. and Gaza, and the Republic of Yemen and TABLE 6.2 Performance of public employment services in selected countries in MENA, 2009 Number of Number of Ratio of Average Job placements registered registered job seekers placements per 1,000 job seekers, job vacancies, per one per month, job seekers Country (thousands) (thousands) registered vacancy (thousands) per month Egypt, Arab Rep. 895.1 222.9 4.0 40.1 45 Jordan 28.0 2.6 10.8 0.7 25 Lebanona 12.2 3.6 3.4 — — Morocco 517.0 27.7 18.7 4.4 9 Tunisia 105.4 — — 1.6 — Source: Angel-Urdinola, Kuddo, and Semlali 2013. Note: — = not available; MENA = Middle East and North Africa. a. Beirut only. 196 JOBS FOR SHARED PROSPERITY conclude that the majority of these programs international best practice (Angel-Urdinola, lack the necessary mix of design features that Semlali, and Brodmann 2010). make programs effective. Focusing resources on those most in The majority of ALMPs provided by PES need—that is, targeting—is essential in light in the Middle East and North Africa focus of fiscal constraints. Identifying the proper on enhancing employability rather than target group and adjusting the program on supporting job matching or providing design accordingly is crucial for obtaining employment services. As noted, job-matching promising results (World Bank 2010b). Tight services play a relatively minor role, and targeting is difficult to achieve in practice; in their importance even decreased between MENA, given the data restrictions, target- 2008 and 2010 (figure 6.22). In contrast, ing for ALMPs is generally ad hoc and cat- training plays an increasingly prominent egorical, so that programs are targeted to role, accounting for about one-third of all specific groups, such as fi rst-time job seek- ALMPs provided in 2010. The majority of ers, women, and the long-term unemployed training programs (62 percent) are in class, (Angel-Urdinola, Semlali, and Brodmann 23 percent are in class and on the job, and 2010). Many countries practice “creaming,” 15 percent are provided only on the job whereby programs are targeted to the most (Angel-Urdinola, Kuddo, and Semlali 2013). qualified applicants. For example, about Training outcomes are often not monitored half of publicly provided ALMPs in Egypt, and hardly ever evaluated, which is particu- Jordan, Lebanon, Morocco, Syria, Tunisia, larly problematic, given that the majority of and the Republic of Yemen target highly training is in the classroom and thus is mostly skilled persons (figure 6.23). supply driven, that is, not coordinated with Most training programs in the ALMP the private sector. In general, a mix of in- inventory focus on providing hard rather class and on-the-job training, in combination than soft skills (Angel-Urdinola, Kuddo, with additional employment services such as and Semlali 2013). About half the train- counseling or job matching, is regarded as ing programs delivered by PES in MENA provide some type of certification, but cer- tification practices vary from country to FIGURE 6.22 Distribution of active labor market programs by country. Countries like Syria and the Repub- program type, 2008 and 2010 lic of Yemen largely lack standard certifica- tion and accreditation systems and national 40 qualifications networks, whereas countries 35 35 like Jordan and Lebanon have more devel- 30 30 oped accreditation systems (Angel-Urdinola, 25 25 Kuddo, and Semlali 2013). Although far Percent 20 20 18 18 from ideal, the public sector is doing better 15 15 15 than the private sector when it comes to pro- 10 9 9 gram certification. A previous review of pri- 5 vately provided ALMPs revealed that only 5 3 0 10 percent of training programs included in an inventory granted beneficiaries some es s g ge n es ed ive in io tiv vic ka bl at in at type of “recognized” credential at program isa en c er Tra cre iti Pa ts rd nc in ob en completion (Angel-Urdinola, Semlali, and ti fo p tj en t-u m m c oy m ra re ar Brodmann 2010). oy pl og Di St Em pl Pr Em An efficient allocation of resources 2008 2010 requires regular monitoring and periodic evaluation of ALMPs. Most programs Source: Angel-Urdinola, Kuddo, and Semlali 2012. Note: Countries included are the Arab Republic of Egypt, Jordan, Lebanon, Morocco, the Syrian Arab included in the ALMP inventory include out- Republic, Tunisia, and the Republic of Yemen. put-based monitoring but lack results-based U N E M P LOYA B L E A N D U N E M P LOYE D: S K I L L S G A P S A N D A M E R I TO C R AC Y D E F I C I T I N M E N A 197 FIGURE 6.23 Active labor market program general lack of job opportunities, the use targeting as a percentage of all programs, 2009 of low-wage, labor-intensive techniques, and finally the predominant use of infor- Refugees 0 mal job search and recruitment channels in the labor market are the major demand-side Abroad 6 constraints. 6 In a region that is generating far too few Disabled jobs, and far too few high-quality jobs, and Specific sector 12 where jobs are allocated not on the basis of 12 merit but on “nonmarket” criteria, the sig- Rural nals coming from the labor market do not Urban 12 emphasize skill acquisition as the route to employment. The signal instead is to acquire Illiterate 12 the “right” degree from the “right” university Low-income individuals 12 and then queue for a public sector job. This message reinforces many of the educational Youth 18 and pedagogical elements that emphasize suc- First-time job 18 cess in the secondary school leaving and uni- seekers versity access examination. The end results Women 18 are relatively poor outcomes at the end of the Poor areas 18 compulsory schooling cycle, little pressure to change traditional patterns of pedagogy (that LTU 24 is, high selectivity and rigid tracking), and a Vocational 35 rather moribund TVET system. education In a small, protected, and uncompeti- Low skills 35 tive private sector, the labor market does Young adults 35 not seem to consider and acknowledge the employability capital of graduates, and the Highly skilled 47 educational and training system, in turn, 0 5 10 15 20 25 30 35 40 50 45 does not seem to have the incentives or the % of all programs capacity to second-guess a labor market that fails to signal. Hence, there is a low-level Source: Angel-Urdinola, Kuddo, and Semlali 2013. equilibrium trap: if the labor market does Note: Data from the Arab Republic of Egypt, Jordan, Lebanon, Morocco, the Syrian Arab Republic, Tunisia, and the Republic of Yemen. LTU = not really demand much, the educational and long-term unemployed. training system is not compelled to deliver much beyond a pipeline of prospective public sector employees. or impact evaluations. The cost-effectiveness Dignity and transparency are values in of ALMPs in MENA is largely unknown. their own right, and demand for them is ris- ing strongly in the region. Efforts to measure these concepts precisely, based on hard data, The employability challenge remain unsatisfactory. But the evidence from Both supply- and demand- side factors con- opinion polls, focus groups, and interviews strain graduates’ chances of cashing in their highlights the frustration that individual “employability capital”—or, as described investment in education and training has not in this report—making the “second transi- been sufficient to secure access to jobs. The tion.” On the supply side, educational and strong and rising social expectations for edu- skill development systems fail to produce and cational systems in MENA are not being ful- signal the relevant skills required by employ- filled. This failure could severely undermine ers. As mentioned in previous chapters, the public confidence in education and training 198 JOBS FOR SHARED PROSPERITY and even lead to questioning the massive pub- to this definition, high means that “students can lic investments made in the sector. apply their understanding and knowledge in a variety of relatively complex situations and explain their reasoning,” whereas low indicates that “students have some basic mathematical Notes knowledge.” 1. The quotations in this section draw on mul- 4. Evidence from the United States suggests tiple sources. As part of the background work that the importance of such skills is increas- for this report, focus groups were carried out ing as the task content of jobs has shifted in March 2011 in the West Bank and Gaza over the past decades. In the United States, (Brodmann et al. 2012a) and in Egypt and the proportion of the labor force employed Tunisia in October 2011 (El-Ashwami2011; in occupations that make intensive use of Mornet-Cariou and Rajadel 2011). Other nonroutine cognitive tasks has increased focus groups and interviews were conducted substantially since 1960, whereas the per- as background work for La Cava (2012), centage of the labor force employed in La Cava et al. (2012), Dawani and Shanti occupations involving routine cognitive and (2011), and Brodmann et al. (2012b). manual tasks, as well as nonroutine manual 2. Research suggests that student achievement activities, has declined (Autor, Levy, and in the first years of schooling is linked to Murnane 2003). children’s emotional and social skills (Ladd, 5. Comparable data on student dropout rates Kochenderfer, and Coleman 1997; O’Neil et are not available. al. 1997; Wentzel and Asher 1995). Many 6. These firm surveys are representative only of studies show that high-quality child care the formal sector. supports the positive social, emotional, and 7. Skill shortages seem more severe for large cognitive development of young children and firms compared with small firms in Morocco, fosters school readiness (Peisner-Feinberg et Syria, and the Republic of Yemen, whereas al. 2001; Vandell and Wolfe 2002). Early the reverse is true in Algeria, Jordan, and the investments in education through families West Bank and Gaza. and public interventions can reduce inequali- 8. With the exception of evidence that draws on ties in educational achievement and reduce secondary research and data, the classification exclusion (Esping-Andersen 2008). For of skills used in this chapter mostly follows example, skills measured before second- the simple distinction introduced by World ary school are important for a host of later Bank (2010a): “cognitive,” “technical,” and outcomes, including educational attainment, “soft” skills. Note that there are overlaps. employment status, wages, and involvement For instance, communication skills are often in crime and health (Carneiro, Crawford, classified as soft skills, as they may be related and Goodman 2007). Therefore, getting to assertiveness (in the socio-emotional field), off to a good start is crucial, as the skills but they may also be related to literacy and developed in early childhood form the basis verbal abilities (cognitive field). of future learning and labor market suc- 9. Skills gaps and poor learning outcomes are cess (World Bank 2010a). Coverage of early phenomena that are not related just to what childhood education is still relatively low in happens in secondary schools or tertiary MENA, with an average of 20 percent enroll- institutions. These are processes that start ment versus almost universal enrollment in very early in life and affect early childhood high-income countries (EdStats). MENA development, child nutrition and health care, countries are very diverse, however, with the access to preschool education, and school Gulf countries (with the exception of Saudi readiness. For references, see note 2. Arabia), Lebanon, and Morocco having 10. International evidence, for example from the rather high rates of preprimary enrollment, United States, suggests that soft skills such as whereas Djibouti, Iraq, Libya, and Syria have self-discipline, perseverance, and passion for rates of less than 10 percent. long-term goals surpass intelligence quotient 3. TIMSS uses five points on the scale as interna- (IQ) as a predictor of academic performance tional benchmarks: “advanced” (>625), “high” (Duckworth and Seligman 2005; Hanushek, (550–624), “intermediate” (477–554), “low” Machin, and Woessmann 2011; Heckman, (400–474), and “below low” (<400). According Stixrud, and Urzua 2006). U N E M P LOYA B L E A N D U N E M P LOYE D: S K I L L S G A P S A N D A M E R I TO C R AC Y D E F I C I T I N M E N A 199 11. There is growing evidence that in fact certain 20. Expression used by Mona Mourshed soft skills such as “executive-function skills” (McKinsey and Company) at a presentation or self-control are relatively malleable—quite delivered to Bahrain’s annual education con- possibly more malleable than IQ, which is ference in 2009. notoriously hard to increase over a sustained 21. Tracking and streaming are alternative period (National Scientific Council on the denominations of similar practices of the Developing Child 2009). ability grouping of students. While tracking 12. These findings are consistent with Cobb-Clark would usually refer to totally different study and Tan (2010), who find that agreeableness programs for different groups of students of (the degree to which a person needs pleas- the same age (for instance, VET programs ant and harmonious relations with others) is in upper-secondary education versus general negatively associated with the probability of ones), streaming refers to ability grouping in being a manager or business professional. selected curriculum areas or subjects. 13. Such councils have been recently created in 22. Ease of access to higher education differs by Egypt, Jordan, and the West Bank and Gaza, field of study. While passing the baccalaure- and they also exist in Morocco and Tunisia. ate is a condition for university entrance in 14. These focus groups were carried out as part of Tunisia, for example, access has traditionally the background work for this report in October been subject to an annual competition (con- 2011 in Egypt and Tunisia and in March 2011 cours) for architecture, engineering, medi- in the West Bank and Gaza. cine, and pharmacy but open to everybody 15. Students are to acquire a certain amount of who passed the upper-secondary examina- competencies during the first few months tion (baccalaureate) for law, humanities, and they spend at the TVET center and then social sciences. Probably as a consequence, acquire another set of competencies on the repetition rates in the latter faculties are higher, job, before returning to the center for the and passage from the first to the second year is next stage, and so forth. The learning process achieved only by about 50 percent of students. is conceived as a linear path. Overall exam passage rates are also modest, at 16. Almeida and Aterido (2011) show that the barely 70 percent (World Bank 2008c). stringency of labor market regulations does 23. Tunisia is an exception, as enrollment in not significantly affect the extent to which TVET has been increasing (see box 6.4). firms invest in job training in the developing 24. One employer referred to graduates from tech- world. nical secondary schools as “technically defeated 17. According to a representative opinion poll and morally destroyed” (El-Ashmawi 2011). among youth in MENA, close to half or more 25. Focus group discussions with students in than half of youth state they would prefer to Egypt, October 2011 (El-Ashmawi 2011). work in the public sector, assuming that the 26. An explanation specific to Lebanon could be pay and working conditions were similar to that both hiring and job search are conducted working in other sectors or to being self- within social and religious networks, reflect- employed (Gallup 2009). ing the high ethnic fractionalization within 18. Public sector employment guarantees for the country. high school and university graduates in 27. To some extent, this result may be particular Egypt and Morocco in the 1960s encouraged to the case of France, where private and public youth to stay in formal education and thus employment agencies have traditionally special- served important social objectives when first ized in recruiting for different kinds of jobs. established. Public agencies attract more low-skilled job 19. See discussion in chapter 1 that preferences seekers, whereas private agencies focus on job for public sector work are stronger among offers for skilled people (Sabatier 2010). This women and that women tend to prefer study- practice, known as “skimming,” however, is ing subjects more suitable for public sector not uncommon and is also observed in MENA jobs. This preference may reflect social norms (Angel-Urdinola, Semlali, and Brodmann 2010). about “suitable subjects for women” but 28. Headhunters usually deal with the elite seg- could equally reflect a preference for public ment of the labor market—often referred sector work. Jobs in the public sector offer to in the literature as the “war on talent”— terms more conducive to family life, such as looking to fill managerial and specialized flexible work hours, leave, and job security. positions in high demand. 200 JOBS FOR SHARED PROSPERITY 29. The use of informal networks for hiring is Social Protection Discussion Paper 55673, less prevalent among firms with foreign or World Bank, Washington, DC. mixed ownership. In Tunisia, for example, 34 Autor, D. H., F. Levy, and R. J. Murnane. 2003. percent of employees with a university degree “The Skill Content of Recent Technological working in private firms with Tunisian own- C h a n ge : A n E mpi r i c a l E x plorat ion .” ership report having found their job through Quarterly Journal of Economics 118 (4): friends or relatives compared with 24 percent 1279–334. in firms with foreign or mixed ownership Binzel, C. 2011. “Decline in Social Mobility: (based on a Graduate Tracer Survey of the Unfulfilled Aspirations among Eg ypt’s graduation cohort of 2004). Educated Youth.” IZA Discussion Paper 6139, 30. A typical example is Lebanon, where the Institute for the Study of Labour (IZA), Bonn. role of the National Employment Office as Bray, M. 20 09. C onfronting the Shadow a core provider of employment services and Education System: What Government Policies training programs has remained peripheral, for What Private Tutoring? Paris: UNESCO– partly as a result of its limited financial and IIEP (United Nations Educational, Scientific, institutional capacities. Its mandate is in and Cultural Organization)–IIEP (International practice limited to (1) running an electronic Institute for Educational Planning). labor intermediation platform; (2) conduct- Brodmann, S., E. Cuadra, M. I. Allouche, and ing occasional studies and labor market S. A. Hillis. 2012a. “From Education-to- needs assessments; and (3) subsidizing some Work: Opportunities and Challenges in the vocational training programs implemented West Bank and Gaza.” Quick Notes Series 55, by NGOs (Angel-Urdinola, Kuddo, and World Bank, Washington, DC. Semlali 2013). Brodmann, S., I. Jillson, N. Hassan, J. Jude Sajdi, and A. Zidan. 2012b. “Social Insurance Reform in Jordan: Awareness and Perceptions References on Employment Opportunities for Women.” Abu-Qara, F. 2012. Palestinian Children: Issues Working paper, World Bank, Washington, DC. and Statistics. Annual Report. Ramallah: Bureau des Etudes de la Planification et de la Palestinian Central Bureau of Statistics. Programmation. 2010, 2011. Répartition des Almeida, R., and R. Aterido. 2011. “On-the- étudiants par domaine d’étude (CITE). [Tunisia Job Training and Rigidity of Employment Ministry of Higher Education and Scientific Protection in the Developing World: Evidence Research, Higher Education in Numbers from Differential Enforcement.” Labour 2010 –2011], http://www.mes.tn/francais/ Economics 18 (Supplement 1): S1–160. donnees_de_base/depliantfr10_11.pdf. Almeida, R., R. Gatti, and J. Saenz. 2012. Caliendo, M., R. Schmidl, and A. Uhlendorff. “On-the-Job Training in the MENA Region: 2 010. “ S o c i a l N e t work s , Job S e a rc h Looking at the I nvestment Pat terns to Methods and Reservation Wages: Evidence Inform Policy.” Working paper, World Bank, for Germany.” IZA Discussion Paper 5165, Washington, DC. Institute for the Study of Labour (IZA), Bonn. Andersen, T., M. E. Hansen, J. Moltesen, L. Calvo-Armengol, A., and M. Jackson. 2004. “The Feiler, R. Götz, T. Wilthagen, I. Borghouts, Effects of Social Networks on Employment and A. Nunn. 2009. The Role of the Public and Inequality.” American Economic Review Employment Services Related to ‘Flexicurity’ 94 (3): 426–54. in the European Labour Markets . Public Carneiro, P., C. Crawford, and A. Goodman. Employment Services Final Report. Brussels: 2007. “The Impact of Early Cognitive and European Commission. Non-Cognitive Skills on Later Outcomes.” Angel-Urdinola, D., A. Kuddo, and A. Semlali. Discussion Paper 0 092 , Centre for the 2013. Building Effective Employment Services Economics of Education, London. for Unemployed Youth in the Middle East and C obb - Cla rk , D. A ., a nd M . Ta n. 2010. North Africa. Washington, DC: World Bank. “Noncog n itive Sk ills, Occupational Angel-Urdinola, D., A. Semlali, and S. Brodmann. Attainment, and Relative Wages.” Labour 2010. “Non-Public Provision of Active Labor Economics 18 (1): 1–13. Market Programs in Arab-Mediterranean Colclough, C ., P. Rose, and M. Tembon. Countries: An Inventory of Youth Programs.” 2000. “Gender Inequalities in Primary U N E M P LOYA B L E A N D U N E M P LOYE D: S K I L L S G A P S A N D A M E R I TO C R AC Y D E F I C I T I N M E N A 201 Schooling: The Roles of Poverty and Adverse Gatti, R., D. Angel-Urdinola, J. Silva, and A. Cultural Practice.” International Journal of Bodor. 2012. Striving for Better Jobs: The Educational Development 20: 5–27. Challenge of Informality in the Middle East Dawani, Z., and E. Shanti. 2011. “Student Aid and North Africa Region. Washington, DC: Focus Groups Study in Jordan.” Report World Bank. transmitted to the government of Jordan, Gorter C., P. Nijkamp, and P. Rietveld. 1993. World Bank, Washington, DC. “The Impact of Employers’ Recruitment Dhillon, N., D. Salehi-Isfahani, P. Dyer, T. Yousef, Behaviour on the Allocation of Vacant Jobs A. Fahmy, and M. Kraetsch. 2008. “Missed by to Unemployed Job Seekers.” Empirical the Boom, Hurt by the Bust: Making Markets Economics 18: 251–69. Work for Young People in the Middle East.” ———. 1996. “Employers’ Recruitment Behaviour Wolfensohn Center for Development–Dubai and Vacancy Duration: An Empirical Analysis School of Government Working Paper, Middle for the Dutch Labour Market.” Applied East Youth Initiative, Brookings Institution, Economics 28: 1463–74. Washington, DC. G ra nove t ter, M . 20 05. “T he I mpac t of Duckworth, A. L., and M. E. P. Seligman. 2005. Social Structure on Economic Outcomes.” “Self-Discipline Outdoes IQ in Predicting Journal of Economic Perspectives 19 (1): Academic Performance of Adolescents.” 33–50. Psychological Science 16: 939–44. Hanushek, E . A., V. Lavy, and K. Hitomi. Eckstein, M., and H. Noah. 1996. Secondary 2006. “Do Students Care about School School Examinations . New Haven: Yale Quality? Determinants of Dropout Behavior University Press. in Developing Countries.” NBER Working EdStats (database). World Bank, Washington, Paper 12737, National Bureau of Economic DC. http://go.worldbank.org/ITABCOGIV1. Research, Cambridge, MA. El-Ashmawi, A. 2011. “T V ET in Egypt.” Hanushek, E., S. Machin, and L. Woessmann, Background paper prepared for the MENA eds. 2011. Handbook of the Economics of Regional Flagship, World Bank, Washington, DC. Education. Vol. 3. Amsterdam: Elsevier. El-Zanaty, F., and S. Gorin. 2007. “Egypt Heckman, J. J., J. Stixrud, and S. Urzua. 2006. Household Education Survey 2005–2006.” “The Effects of Cognitive and Noncognitive Cairo: El Zanaty and Associates and Macro Abilities on Labor Market Outcomes and International, Inc. Social Behavior.” Journal of Labor Economics Enterprise Surveys. Various years. International 24 (3): 411–82. Finance Corporation (IFC) and World Bank, Heinze, A., and E. Wolf. 2010. “The Intra-Firm Washington, DC. http://data.worldbank Gender Wage Gap: A New View on Wage .org/data-catalog/enterprise-surveys. Differentials Based on Linked Employer– Esping-Andersen. 2008. “Childhood investments Employee Data.” Journal of Population and Skill Formation.” International Tax Economics 23: 851–79. Public Finance 15: 19–44. ICA (Investment Climate Assessment) (database). ETF (European Training Foundation) and World World Bank, Washington, DC. http://web Bank. 2005. Reforming Technical Vocational .worldba n k.org / W B SI T E / E X T E R NA L / Education and Training in the Middle East and COUNTRIES/AFRICAEXT/EXTAFRSUMA North Africa: Experiences and Challenges. FTPS/0,,contentMDK:22472299~menuPK:68 Luxembourg: Office for Official Publications 19957~pagePK:51246584~piPK:51241019~th of the European Communities. eSitePK:2049987,00.html. Executive Opinion Survey (database). World IFC (International Finance Corporation) and Economic Forum. http://www.weforum.org/. ISDB (Islamic Development Bank). 2011. Franzen, A., and D. Hangartner. 2006. “Social Education for Employment: Realizing Arab Networks and Labour Market Outcomes: Youth Potential. Washington, DC: IFC and T he Non-Monet a r y B enef its of S o cia l ISDB. Capital.” European Sociological Review 22 IMF (International Monetary Fund) World (4): 353–68. Economic Outlook (database). 2013. January Gallup. 2009. “The Silatech Index: Voices of 2013 World Economic Update. http://www Young Arabs.” http://www.somalilandtimes .imf.org/external/pubs/ft/weo/2013/update/01/ .net/sl/2010/422/Silatech_Report_Final.pdf. index.htm. 202 JOBS FOR SHARED PROSPERITY Ioannides, Y., and L . Lour y. 20 04. “Job for the MENA Regional Flagship, World I n formation Net works , Neig hborhood Bank, Washington, DC. Effects and Inequality.” Journal of Economic Mortensen, D. T., and T. Vishwanath. 1994. Literature 42 (4): 1056–93. “Personal Contacts and Earnings: It Is Who You Ireson, J., S. Hallam, and C. Hurley. 2005. “What Know!” Labour Economics 1 (2): 187–201. Are the Effects of Ability Grouping on Gcse National Scientific Council on the Developing Attainment?” British Journal Educational Child, Harvard University. 2009. http:// Research Journal 31 (4): 443–58. developingchild.harvard.edu /index.php? Jakubowski, M., H. A. Patrinos, E. E. Porta, cID=89. and J. Wisniewski. 2010. “The Impact of the OECD (Organisation for Economic Co-operation 1999 Education Reform in Poland.” Policy and Development). 2007. PISA 2006: Science Research Working Paper 5263, World Bank, Competencies for Tomorrow´s World. Paris: Washington, DC. OECD. La Cava, G. 2012. “Reclaiming Their Voice: O’Neil, R., M. Welsh, R. D. Parke, S. Wang, and Young Women and Men’s Perspectives from C. Strand. 1997. “A Longitudinal Assessment Upper Egypt.” Social Protection Study 71674, of the Academic Correlates of Early Peer World Bank, Washington, DC. Acceptance and Rejection.” Journal of Clinical La Cava, G., T. Vishwanath, M. Morgandi, Child Psychology 26 (3): 290–303. U. Serajuddin, and L. Jilson, 2012. Kingdom Peisner-Feinberg, E. S., M. Burchinal, R. M. of Morocco: Promoting Youth Opportunities Clifford, M. Culkin, C. Howes, S. L. Kagan, and Participation. Washington, DC: World and N. Yazejian. 2001. “The Relation of Bank. Preschool Child Care Quality to Children’s Ladd, G. W., B. J. Kochenderfer, and C. C. C o g n it ive a nd S o c i a l D e velopm e nt a l Coleman. 1997. “Classroom Peer Acceptance, Trajectories through Second Grade.” Child Friendship, and Victimization: Distinct Development 72 (5): 1534–53. Relational Systems That Contribute Uniquely Pierre, G., and D. Robalino. 2012. “Skills Analysis to Children’s School Adjustment.” Child of the Lebanese Labor Market.” Working Development 68 (6): 1181–97. paper, World Bank, Washington, DC. Margolis, D. N., and V. Simonnet. 2002. Premand, P., S. Brodmann, R. Grun, and “Technical / Professional versus General R. Almeida. 2012. “Entrepreneurship Training Education, Labor Market Networks and Labor and Self-Employment among University Market Outcomes.” International Journal of Graduates: Evidence from a Randomized Trial Manpower 23 (5): 471–92. in Tunisia.” Paper presented at the Institute for MFPE (Ministry of Vocational Training and the Study of Labour (IZA)/World Bank/OECD E mploy ment) a nd World B a n k. 20 09. Conference, “Activation and Employment Employment Dynamics and Relevance of the Support Policies,” Istanbul, April 30 and May 1. Training Received by University Graduates Russo, G., W. H. J. Hassink, and C. Gorter [Dynamique de l’emploi et adéquation de la 2005. “Filling Vacancies: A n Empirical formation parmi les diplômes universitaires]. Analysis of the Cost and Benefit of Search in Tunis: MFPE. the Labour Market.” Applied Economics 37: Montgomery, J. D. 1991. “Social Networks 1597–606. and Labor-Market Outcomes: Toward an Sabatier, M. 2010. “Filling Vacancies: Identifying Economic Analysis.” American Economic the Most Efficient Recruitment Channel.” Review 81 (5): 1407–18. Economics Bulletin 30 (4): 3355–68. Moreno, O. J. M. 2006. “The Dynamics of Salehi-Isfahani, D., and N. Dhillon. 2008. Cu rricu lu m Desig n a nd Development: “Stalled Youth Transitions in the Middle East: Scenarios for Curriculum Evolution.” In A Framework for Policy Reform.” Middle East School Curricula for Global Citizenship: Youth Initiative Working Paper 8, Brookings Comparative and Historical Perspectives Institution, Washington, DC. on Educational Contents , ed. A. Benavot Salmi, J., and R. M. Bassett. Forthcoming. and C. Braslavsky, 195–211. Hong Kong: Opportunities for All: A Global Study on C ompa rat ive E ducat ion C enter of t he Equity in Tertiary Education. Washington, University of Hong Kong. DC: World Bank. Mornet-Cariou, O., and T. Rajadel. 2011. “TVET Spence, M. 1973. “Job Market Signaling.” Profi le Tunisia.” Background paper prepared Quarterly Journal of Economics 87 (3): 355–74. U N E M P LOYA B L E A N D U N E M P LOYE D: S K I L L S G A P S A N D A M E R I TO C R AC Y D E F I C I T I N M E N A 203 TIMSS (Trends in International Mathematics ———. 20 08a. T he Roa d Not Travell ed : and Science Study). Institute of Education Education Reform in the Middle East and Sciences, National Center for Education Nor th Afric a . Washing ton, DC: World Statistics. http://nces.ed.gov/timss/. Bank. Toumi, B. 2009. “Le Système de Financement de ———. 2008b. “Hashemite Kingdom of Jordan: la Formation Professionnelle en Tunisie” [The Resolving Jordan’s Labor Market Paradox fi nancing system for professional education of Concurrent Economic Growth and High in Tunisia]. Powerpoint presentation at the Unemployment.” Policy Note, World Bank, Education for Development and Competitivity Washington, DC. conference. Rabat, January 21. http://info. ———. 2008c. “Tunisia Labor Demand, Skills worldbank.org/etools/docs/library/250694/ Supply a nd E mploy ment: Towa rds a n day%205Tunisia%20Toumi.pdf. Integrated Strategy for Job Creation—Phase U N E S C O ( Un ite d Nat ion s E duc at ion a l , II.” Policy Note, World Bank, Washington, DC. Scientific, and Cultural Organization). 2011. ———. 2010a. Stepping Up Skills for More Jobs “Global Education Digest 2011 Focus on and Higher Productivity. Washington, DC: Secondary Education.” Paper presented at World Bank. a regional workshop, Almaty, Kazakhstan, ———. 2010b. “Active Labor Market Programs November 15. for Youth: A Framework to Guide Youth Vandell, D. L., and B. Wolfe. 2002. “Child Care Employment Interventions.” Employment Quality: Does It Matter and Does It Need to Policy Primer 16, World Bank, Washington, Be Improved?” Special Report, University of DC. Wisconsin, Institute for Research and Poverty, ———. 2013. Opening Doors: Gender Equality Madison. in the Middle E ast and Nor th Afric a . WEF (World Economic Forum) and OECD. Washington, DC: World Bank. 2011. Arab World Competitiveness Report World Bank and Ministère de l’Emploi et 2011–2012. Geneva: World Economic Forum. de l’Insertion Professionnelle des Jeunes. Wentzel, K. R., and S. R. Asher. 1995. “The 2005. Tunisia Higher Education Graduate Academic Lives of Neglected, Rejected, Tracer Survey, 2005 [Enquête sur l’insertion Popular, and Controversial Children.” Child professionnelle des diplômes de l’enseignement Development 66 (3): 756–63. supérieur de 2005]. Whaba, J., and Y. Zenou. 2005. “Density, World Development Indicators (database). World Social Networks and Job Search Methods: Bank, Washington, DC. http: //data.worldbank. Theory and Application to Egypt.” Journal of org/data-catalog/world-development-indicators. Development Economics 78 (2): 443–73. Yorke, M. 2006. “Employability in Higher World Bank. 2006. West Bank and Gaza Education: What It Is—What It Is Not.” Education Sector Analysis. Washington, DC: Learning and Employability Series 1. York, World Bank. United Kingdom: Higher Education Academy. Policy Options PART 3 A little knowledge that acts is worth infinitely more than much knowledge that is idle. Advance, and never halt, for advancing is perfection. Advance and do not fear the thorns in the path, for they draw only corrupt blood. — Khalil Gibran Introduction T he third part of the report discusses the policy options that can address the barriers to achieving higher productivity, private sector–led job creation and improve job opportunities, especially among those who suffer most from the current low-productivity equilibrium. While the effort has been made to organize these policy options along a labor-demand, -supply, and -intermediation frame- work, it must be emphasized from the outset that these policies are strongly com- plementary and mutually reinforcing. In particular, chapter 7 covers a range of policy options intended to realign the incentives toward private sector–led job creation and increased labor productivity and innovation, as well as to foster countries’ entrepreneurship potential. The follow- ing chapter discusses how to improve labor market regulations and social insurance systems, with a view toward achieving greater protection against risk and reduce the wedge between labor market outsiders and insiders. Finally, chapter 9 discusses mea- sures to refocus education and training systems on generating the skills that enhance young people’s employability and on creating interventions to improve matching and intermediation in the labor market. 205 Aligning Incentives to Invest, Innovate, and Generate 7 Employment Main findings • The creation of more and better-quality jobs requires removing the distortions that repress labor demand and encouraging the process of creative destruction that drives labor productivity. • Priorities include reforming the financial sector to incentivize lending to small enterprises and removing the bias against labor-intensive activities by reforming energy and agricultural subsidies. • Enhancing the transparency and accountability of public administrations and establishing effective competition authorities are key to reducing the scope of the discretionary application of regulations and making long-term investments possible. • A combination of training, finance, and infrastructural investments can improve the productivity of the existing stock of subsistence entrepreneurs. • The entry of educated and experienced workers into entrepreneurship—those who are best positioned to create high-value-added firms—can be incentivized by programs that enhance skills, improve market access, and reduce risks associated with forgoing wage employment. • The proposed set of reforms would also realign the incentives for firms in MENA to invest in innovation, which can be further enhanced by public-private partnerships, including those between universities and private firms. Introduction regulations. As a result, the private sector in MENA, particularly the high-productivity A s identified in chapter 4, major con- segment, is smaller than optimal and can- straints to generating jobs in the Mid- not expand and diversify rapidly enough to dle East and North Africa (MENA) absorb the growing labor supply. Chapter include distortions that repress the demand 10 will discuss the political economy that for labor, burdensome business regulations, has sustained this system of favoritism and and discretionary enforcement of those rents distribution in the private sector. This 207 208 JOBS FOR SHARED PROSPERITY chapter instead focuses on reform options the adoption of more advanced, energy- and discusses four priority areas for reform- efficient modes of production. Chapter 4 ing job creation strategies: demonstrated that such a reform would level the playing field for smaller businesses • Removing distortions that repress the and private sector firms, given that their demand for labor competitors—large and public firms—benefit • Lifting barriers that prevent fi rms from disproportionately from energy subsidies. competing on a level playing field, includ- It is important to recognize the potential ing the barriers to accessing productive repercussions on the political economy of inputs removing these subsidies. Given the mag- • Promoting incentives for fi rms to innovate nitude of these subsidies, their main ben- • Supporting entrepreneurship eficiaries constitute a powerful constituency Some of these policy recommendations that could oppose their removal, especially are examined in greater detail in two previ- when risks of short-term losses in employ- ous World Bank flagships: From Privilege ment associated with restructuring are high. to Competition (World Bank 2009) and Energy-intensive industries are especially Financial Access and Stability: A Road- likely to lobby, as they will be the most map for the Middle East and North Africa affected. In the Arab Republic of Egypt, for (World Bank 2011). Those publications offer example, energy-intensive industries directly in-depth discussions of measures to reduce consumed more than 80 percent of the coun- the discretionary enforcement of regulation try’s fuel oil and 28 percent of its diesel oil, and to increase access to credit and kick-start which together amount to 50 percent of a more dynamic private sector. In addition, energy subsidies in the country (Abouleinein, the recent World Development Report 2013: El-Laithy, and Kheir-El-Din 2009). Ensur- Jobs (World Bank 2013) presents a thorough ing sustainability and creating the neces- review of the evidence on the importance of sary political support for reform will require competition and a fair and rule-based invest- adequately compensating losers and support- ment climate. ing their transition to more energy-efficient production. This approach is feasible, since removing energy subsidies would free up Strategies to remove distortions substantial fiscal space for subsidizing the that repress labor demand technical assistance and credit that firms Tax and subsidy schemes in MENA concur- will need to make the technological trans- rently tend to place labor at a disadvantage. formation (as long as technical capacity and Thus, reducing labor taxes and subsidies to institutional facilities are in place). In the other factors of production is likely to gener- short run, the freed-up resources are prob- ate more employment opportunities and, in ably best spent on compensating losers as the medium run, accelerate growth through subsidies are being phased out, while in the enhanced allocative efficiency and innovation. medium run, the freed-up resources could be used to reduce labor taxes. The effect of these changes would render labor relatively Removing energy subsidies cheaper, which would result in additional A salient priority is reforming energy sub- demand for labor and could also increase sidies, which make up the largest share of the tax base of formal employment (see subsidies in most MENA countries (Silva, chapter 8). Removing energy subsidies would Levin, and Morgandi 2013) and favor also have important (indirect) effects on energy-intensive production processes that the direct consumption of the poor and on rely on outdated technology. Ensuring that overall consumer prices as well. It is essen- fi rms face appropriate energy prices would tial that well-targeted social safety nets are enhance competitive pressure and stimulate in place to help low-income consumers cope ALIGNING INCENTIVES TO INVEST, INNOVATE, AND GENERATE EMPLOYMENT 209 with the loss in purchasing power. Many removing subsidies on wheat (which account countries around the world have successfully for a large share of agricultural subsidies in managed to remove energy subsidies in this MENA) could realign commodity prices and way. The discussion of the specific design of enhance allocative effi ciency. The impacts social safety nets that favor this transition is on agricultural employment could be impor- beyond the scope of this report but remains tant, because wheat production is relatively absolutely essential and is discussed in detail capital intensive. By contrast, other agricul- in Silva, Levin, and Morgandi (2013) and in tural products, such as olives, require more Fattouh and El-Katiri (2012). labor input, and reforming wheat subsidies might thus boost the demand for labor. Similar to the removal of energy subsidies, Labor-reducing agricultural subsidies the removal of wheat subsidies could free Agricultural subsidies are another area resources to finance investments in agri- where reform can simultaneously increase cultural productivity. Box 7.1 presents an employment and fi scal space. For example, example from the Syrian Arab Republic. BOX 7.1 Agricultural employment and subsidies in the Syrian Arab Republic An estimated 19 percent of the labor force in Syria while for sugar beets the farm-gate price was about works in agriculture. Most farms are operated by 140 percent over parity. Similarly, domestic pro- the owners and their families. The intensity of farm ducer prices of durum and soft wheat were about labor depends on the crop and the relative mecha- 50 percent and 65 percent, respectively, over the nization of farm operations. Cereal production is parity prices in 2009. almost fully mechanized and therefore of low rel- Subsidies make some crops look more profit- evance for rural employment. In contrast, sugar able than they are in reality and encourage farmers beets and oilseeds provide employment in larger to produce them. This can be problematic if “arti- farms, as most small farmers will not engage hired fi cially profitable” crops overuse inputs that are in labor for these crops. Industrial crops are the larg- scarce supply, such as water, or underuse inputs in est providers of casual employment in agriculture. high supply, like labor. Simulations show that for The large cotton-producing areas attract signifi- most governorates in Syria, wheat farming would cant flows of casual labor from across the country. continue to be profitable even if subsidies were Given the labor intensity of cotton picking, which completely removed. In the governorates with no is done mostly by women, even small and medium comparative advantage for wheat production (Deir farms often employ additional labor to complement Ezzor, Reef Dimashk, and Daraa), the removal of family labor, especially at harvest. The second main subsidies would result in net losses, and most farm- provider of seasonal employment is fruit tree crops, ers would likely stop growing wheat. Irrigation- particularly olives, which employ unskilled laborers intensive (but also labor-intensive) crops such as for harvesting and skilled labor for pruning. The sugar beets and cotton would probably not be pro- third major element in Syria’s casual agricultural duced if subsidies were removed. In light of Syria’s labor market is vegetable cultivation, especially water scarcity and labor abundance, and given the seasonally available employment in harvesting field need for better alignment of domestic and interna- vegetables. tional market prices, these crops could be replaced Subsidized crops include wheat, sugar beets, by others that are more labor intensive and have a and cotton. Producer prices of these “strategic high value in the market, such as fruits and vegeta- crops” are administered by the government, often bles, especially oilseeds, nuts, and fruits with limited above international prices. In 2009, for instance, irrigation needs, such as olives and figs. the domestic farm-gate price for cotton was about 115 percent over its international parity equivalent, Sources: Cafiero 2009; Saade, Abdou, and El-Amin 2011. 210 JOBS FOR SHARED PROSPERITY Strategies for competing and For instance, countries could, at very low investing cost, do much to improve the contestability of restricted professions, where barriers to Reducing the costs of entry, exit, and entry are often high and linked to excessive adjustment discretion. In Morocco and Tunisia, some Productivity growth and innovation are professional services, such as accounting, repressed in economic environments that are could be made more accessible without com- not contestable, and the slow pace of tech- promising quality by, among other things, nology adoption, fi rm entry, and churning ensuring that clear criteria are specified for in MENA all attest to a lack of competition. admission and anticompetitive practices are Creating more jobs requires the expansion reformed (see box 7.2). of existing firms, the generation of new Experience from developed and develop- firms, and, paradoxically, the destruction ing economies alike shows that competitive of unproductive fi rms that use inputs ineffi- markets also require institutions that pre- ciently. This change is unlikely to be feasible vent abuse from dominant players and ensure in the absence of greater competitive pres- consumer protection. Empowering— or sure and stronger contestability. establishing—independent antitrust agencies One effective means of enhancing com- is essential for bolstering competitiveness and petitiveness and encouraging competition for ensuring that liberalization and privatiza- is reducing the cost and barriers to entry, tion do not lead to abuse of market power.1 which are formidably high in MENA. Con- Most countries in the region have established versely, protracted and expensive bank- competition agencies, while, besides those in ruptcy procedures hamper the weeding out the Gulf Cooperation Council (GCC), only of inefficient fi rms and limit entry and risk Egypt and Tunisia have established consumer taking, as forward-looking entrepreneurs protection agencies. The recent assessment can become discouraged. of these institutions in MENA noted the BOX 7.2 Liberalizing entry into professions: Accounting in Morocco and Tunisia The regulations for the practice of law and account- the training, candidates need to be accepted by ing place Tunisia among the most restrictive regimes the Tunisian bar association (Ordre des Avocats on the restrictiveness index of the Organisation for Tunisiens). While both steps correspond to interna- Economic Co-operation and Development (OECD), tional practice, there is little public material on the even behind countries with a similar French legal criteria that determine admission in either instance tradition (see figure B7.2.1). (some are still being clarified in decrees). The entry Most restrictions are on the fee scale and nation- to the accounting profession requires a national ality of employees. Not captured by this index but competition for the Accounting Experts, but not also important is the fact that foreign fi rms can- for accountants, who can access the profession with not practice law in Tunisia, only “legal consult- a degree and an internship.b Still, the associations ing,” and that all accounting firms must be 100 that in principle self-regulate admission to the audit percent owned by Tunisians. Still, in principle, a and accounting professions are under the supervi- foreign individual can be authorized to practice sion of the Ministry of Finance, which also strictly as an accountant, if a convention of mutuality regulates the audit practitioners’ scale of fees. exists between Tunisia and the individual’s home Fee ceilings reduce the profi tability and therefore country. a For lawyers, a competition restricts attractiveness of a market, and fee floors preclude entry to the practical poststudy training, and after unknown new entrants from cutting prices. (continued next page) ALIGNING INCENTIVES TO INVEST, INNOVATE, AND GENERATE EMPLOYMENT 211 BOX 7.2 Liberalizing entry into professions: Accounting in Morocco and Tunisia (continued) FIGURE B7.2.1 OECD restrictiveness index for accounting services in selected countries, 2008 Turkey Italy Tunisia Czech Republic Canada France Netherlands Luxembourg Federal Republic of Germany Belgium Sweden Japan Finland Slovak Republic Iceland Korea, Rep. 0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 Restrictiveness index Source: OECD Services Trade Restrictiveness Index. Morocco has a professional Ordre de Compta- imposing irrelevant standards. An additional barrier bles like in Tunisia, but the commission in charge of to profitable practice could be the rule that accoun- admitting new accountants contains both the ordre’s tants are not allowed to advertise. members and members of the administration and Source: World Bank 2008. is presided over by the Ministry of Finance. Given a. Article 15 of law 88–108 of August 18, 1988. the lightness of the legal texts, much discretionary b. The “experts comptables” (law 88–108 previously cited) can access the profession after a national exam, a (“validated”) internship of three years with power remains with the admitting commission. The a professional or in the public administration, and a thesis. If these conditions law gives the committee the power to admit but are fulfilled, admission to the Ordre des Experts Comptables is automatic. For the “comptables” (law n°2002–16 previously cited), access to the profession does not hold it accountable for adhering to stan- depends on the degree diploma and an internship (there is no competitive dards beyond those listed or, more important, for examination). 212 JOBS FOR SHARED PROSPERITY growing role that competition agencies play based on merit or commitment to a devel- but also highlighted how they all continue opment strategy instead of on regional and to lack full autonomy in the de facto exercise sectarian considerations. Finally, strategic of their function (CUTS 2011). For example, incentives for public agencies should reward several agencies have no clear legislation spell- efforts to increase private sector growth and ing out the process of staff appointments, discourage discretion. Creating a perfor- long recognized as a crucial step toward mance tool that incorporates users’ feedback impartial hiring (World Bank 2002). In addi- can highlight the public sector’s unwarranted tion, in all countries, ministers who are in involvement in the private sector and identify charge of state-owned industries—which con- areas of particular arbitrariness in the deci- tinue to enjoy significant protection against sion making of public officials. private and foreign firms—also tend to be Increasing access to information can also the ultimate arbiters of competition cases, an enhance accountability. Business information arrangement that leads to conflict of interests. should be made publicly available through Moreover, the private sector, the media, and online resources, and independent research the public at large are relatively disengaged and surveys should be permitted. Implement- from the operation of these agencies and are ing uniform enterprise identification numbers not proactive in voicing complaints, thus would become a valuable resource for record- minimizing the deterring function that these ing transactions between businesses and pub- agencies could play (CUTS 2011). lic institutions and would allow governments to better detect fraud and law infringements (World Bank 2009). Ensuring that new laws Enhancing transparency and and regulations and the procedure for their institutional accountability implementation are unambiguous and pub- As explained in chapter 4, the discrimina- licly available will make the regulatory envi- tory and inconsistent implementation of ronment more transparent. business regulation—whereby public offi- Investment-promotion agencies could cials tend to favor a select few firms but also play an important role in enhancing the hamper job creation by their (potential) transparency and predictability of the regula- competitors—has arguably been more per- tory environment for potential foreign direct nicious than burdensome regulation itself. investors, who are a very important driver of Discretionary and inconsistent implementa- domestic competitiveness and potentially of tion of regulations creates uncertainty that employment. A recent review of the effective- discourages investment, innovation, and ness of investment-promotion agencies in 124 ultimately job creation. countries suggests that they are particularly Improving the accountability of public effective in countries where bureaucratic red administration can foster the consistent tape and information asymmetries prevail implementation of rules and regulations. (see Javorcik and Harding 2011). These reforms may include several elements. Transferring responsibilities and decision Improving trade openness making to lower tiers of public administra- tion, while making lower-tier civil servants Another means of enhancing competition accountable to a broader base instead of to a is to further open economies to trade. This single minister or high-ranking official, could strategy could pay a double dividend: it enhance accountability and reduce the scope would boost demand and spur productiv- for opportunistic behavior. A complementary ity growth both by serving as a disciplining measure would be to increase the capacity of device and by allowing fi rms to capitalize on workers at lower tiers of public administra- knowledge spillovers. tion. Recruitment and promotion schemes in In addition, international trade agree- public administrations should be restructured ments can signal a government’s commitment ALIGNING INCENTIVES TO INVEST, INNOVATE, AND GENERATE EMPLOYMENT 213 to providing a predictable policy environ- sustainable economic growth (World Bank ment, which is essential to guaranteeing 2011). Improving access to finance for small a return on investment in industries with and medium enterprises should be a prior- long gestation periods, such as those that ity; these businesses currently have very use research and development (R&D) more limited access to credit, in spite of a high intensively. For instance, an expansion led willingness to pay, attesting to severe credit by foreign direct investment (FDI) in Jor- market failures. dan’s pharmaceutical industry was triggered by an international trade agreement, which Strengthen financial infrastructure increased investors’ confidence in the busi- Effective credit reporting systems are a ness environment (see box 7.3). precondition for fi nancial development. In fi nancial systems dominated by the private sector (see below), a critical challenge is to Expanding access to finance so that ensure that private credit bureaus extend firms can grow and invest coverage to include small and medium enter- Whenever entrepreneurs face a very uneven prises. Without sufficient credit information, price of capital because of the discretion- smaller entities cannot obtain the credit ary allocation of credit, fi nancial resources needed to expand their businesses. In econ- are not allocated to their most produc- omies where bureaus are state led, public tive uses in the economy. Policies focused credit registries should be upgraded to oper- on strengthening fi nancial infrastructure, ate as best-practice private bureaus. It is also increasing bank competition, and devel- important that information-collecting poli- oping nonbank financial institutions can cies mandate that all credit-reporting institu- not only improve access to finance in the tions collect in-depth credit information, as MENA region and revitalize the private well as information from microfinance insti- sector but also lay the foundation for more tutions, utilities, and retailers (see table 7.1). BOX 7.3 Jordan’s pharmaceutical sector and trade reforms Trade reforms were crucial in creating an environ- first-mover advantage by starting production of soon- ment for developing the manufacturing sectors that to-expire protected drugs in Jordan. Many renowned support the pharmaceuticals industry in Jordan. global pharmaceutical players—including Astra- Taking advantage of its accession to the World Zeneca, Sanofi -Aventis, Bristol-Myers Squibb, Eli Trade Organization (WTO) in 2000, Jordan signed Lilly, GlaxoSmithKline, Janssen-Cilag, Merck Sharp a free-trade agreement with the United States in & Dohme, Novartis, Organon, Roche, Pfi zer, and which it pledged to harmonize its national legisla- Schering-Plough—have established production sites tion with the WTO Agreement on Trade-Related or expanded their commercial activities in Jordan.a Aspects of Intellectual Property Rights. According Jordan’s pharmaceutical sector has attracted new to the Pharmaceutical Manufacturers Association investments, gained new export markets, and of America, the U.S.-Jordan Free-Trade Agreement engaged in innovative research. According to the has made Jordan’s market more appealing for phar- International Intellectual Property Institute, Pfi zer maceutical research and development, as well as for doubled the number of its local employees, Sanofi - sales and licensing agreements. The benefits include Aventis and Novartis tripled their local labor forces, expanded data protection, elimination of exclusions and Merck increased its employment in Jordan by from patentability for biotechnology inventions, and 500 percent between 2000 and 2004. limitations on compulsory licensing. These reforms allowed many European firms to benefit from a Source: World Bank 2012. 214 JOBS FOR SHARED PROSPERITY TABLE 7.1 Credit registry infrastructure in MENA prices to fall and make services more acces- sible to consumers, and lenders will be more Public registry only Private credit registries likely to penetrate new markets and extend Algeria Bahrain services to underserved communities. Djibouti Egypt, Arab Rep. Jordan Iran, Islamic Rep. Develop nonbank financial institutions Lebanon Kuwait Finally, developing nonbank fi nancial insti- Oman Morocco tutions will help increase access to financial Qatar Saudi Arabia services and expand the range of services pro- Syrian Arab Republic United Arab Emirates Tunisia vided. Improvements in insurance services, West Bank and Gaza which are essential for risk management, Yemen, Rep. would directly contribute to development by reducing risks of large business losses and Source: World Bank. Note: MENA = Middle East and North Africa. encouraging investment. Expanding con- tributions of private sector pension funds to capital market development could also An overhaul of the current collateral contribute to long-term growth. Reverse regimes would increase protection of creditor factoring could be considered as a means of rights. This measure is crucial to increasing providing smaller enterprises with alternative lenders’ willingness to extend credit to the less sources of financing, particularly in areas established and smaller borrowers. Ideally, leg- with limited credit information. These enter- islation would deal with all aspects of secured prises can then rely on the financial standing lending. Such legislation would facilitate of their customers and receive financing at a broader types of credit transactions by allow- lower cost (World Bank 2011). ing generic descriptions of assets listed as col- By developing a market for small loans, lateral, provide flexibility for secured creditors microfinance could make an important con- to enforce their agreements extrajudicially, and tribution to increasing financial access in establish a clear priority scheme for secured MENA. According to a recent study testing lenders. Registries of movable collateral (such the impact of microfi nance in rural areas of as inventory, crops, and equipment) should Morocco, the introduction of microcredit in also be modernized and include a single elec- highly credit-constrained environments sig- tronic database that is searchable and accessi- nificantly increased access to fi nancial mar- ble to the public. Predictability of enforcement kets, quadrupling the borrowing power of is a pivotal precondition for lender confidence. households (Crépon et al. 2011). However, a fi nancial infrastructure that ensures shar- Increase bank competition ing of credit information on small loans will Reforms in reporting credit information be necessary for the continued success of and creditors’ legal rights will improve this market. Without a mutual exchange of financial infrastructure and likely lead to information on borrowers, lenders will not be increased competition because of a more aware of a borrower’s existing lines of credit secure lending environment for creditors. In with other institutions and cannot accurately addition, there is a need for greater supervi- assess the borrower’s creditworthiness or like- sion of competition in the banking system: lihood of default (World Bank 2011). Efforts an authority tasked with monitoring bank to expand women’s access to credit are also competition could both discourage anticom- warranted (see box 7.4). petitive behavior and reward good practices. When implementing licensing criteria, bank regulators should also consider sound com- Facilitating innovation for petition to reduce the obstacles that prevent high-productivity employment reputable banks from entering the market. Technological progress is the source of long- Greater competition among banks will cause term growth, and innovation in its various ALIGNING INCENTIVES TO INVEST, INNOVATE, AND GENERATE EMPLOYMENT 215 BOX 7.4 Improving financial access for women Access to financial services is a problem for men include increasing educational resources for women and women in MENA; yet women tend to face more entrepreneurs and developing more microfinance obstacles in obtaining credit. In Egypt, 92 percent of institutions that aim to improve women’s access to loan applications are rejected because of insufficient credit. collateral. This problem is especially relevant to A few banks in Egypt recognize the potential women, since cultural norms often prevent women benefits of women borrowers and now offer train- from owning or managing their own assets. In addi- ing in business skills to enhance the performance of tion, the fragmented microfi nance sector remains women owners of smaller enterprises. Such train- small, deterring potential credit opportunities for ing programs not only help women take advan- women who are unable to obtain credit through tage of financial services but also simultaneously commercial banks. reduce credit risk to banks. Similar initiatives would New gender-specific targets for reform are needed expand fi nancial access in the region by extending to improve fi nancial access for women. Egypt has resources to a marginalized segment of the popula- taken a number of steps to improve credit infrastruc- tion while also investing in future development of a ture in recent years, such as instituting private credit credit market. bureaus, but further progress is needed to create a more equal environment for women. Policy options Source: Nasr 2010. forms is at the root of technological prog- invention and adoption of new technologies, ress. Innovation—an important engine innovation is typically led by the private for moving economies up the production sector: it is the profi t motive that propels ladder—is at the core of a long-term strategy fi rms to innovate. Crucial preconditions for for more and better jobs. Innovation-based innovating profitably are a favorable busi- growth can also create the types of high- ness climate and a stable regulatory envi- quality jobs to which the increasingly edu- ronment. When firms are able to adopt a cated youth in the MENA region can aspire. medium-term planning horizon and reap Increasing the density of highly innova- the rewards of high-risk, high-return activi- tive enterprises in a country or region can ties, such as research and development or also have broad benefits by creating momen- experimentation, innovation is incentivized tum for reform. Given the relatively early and occurs faster. However, chapter 4 has development of this sector in MENA, few shown that MENA fi rms overall innovate entrenched interests will benefit from limit- less than firms in comparator countries, ing entry in such sectors compared to sectors and they do so even less in countries and that have already matured. Thus, creating an in industries where the regulatory environ- investment climate conducive to highly inno- ment is more discriminatory—that is, least vative sectors may be less fraught with politi- predictable. For this reason, the main role cal problems and, moreover, may benefit that governments can play in facilitating the economy as a whole through knowledge higher adoption of innovative products or spillovers. processes in MENA is to realign the incen- tives (and rewards) for the private sector to invest in innovation. Measures such as the Importance of a conducive investment reduction of barriers to competition; the climate for innovation provision of a stable policy and legal envi- Global experience has shown that while ronment; the reduction of barriers to global some public policies can facilitate the demand, knowledge, and inputs; and the 216 JOBS FOR SHARED PROSPERITY improvement of access to fi nance (in partic- In a relatively short period, the mobile tele- ular, risk-tolerant fi nancing such as venture com sector in MENA countries has reached capital) would all have a positive impact on nearly the same level of dynamism as more job creation. mature markets in Western Europe. It has The introduction of competition among also fostered the rise of regional operators mobile telecommunication providers in originating in MENA economies. MENA is a good example of how opening The prohibition against competition up protected sectors can accelerate invest- with state monopolies in broadband Inter- ment, innovation, and consumer welfare, net service across MENA, however, has at no cost to the government (see box 7.5). led to a stagnation in the sector, with high BOX 7.5 Liberalization, investment, and job creation: The telecom sector in MENA Since the early 2000s, the telecom sector has been FIGURE B7.5.1 Mobile and broadband growth in MENA, the only sector with high-technological input to 2003–11 attain strong investment volumes in MENA. In the early 2000s, all countries in the region started gradually opening up to competition, breaking the 140 Market penetration rates, % monopoly and issuing a second wireless license. As 120 115.23 land-line telecom companies still had limited know- 100 how and a small customer base, the stakes of lib- 80 eralization were limited. Probably for this reason, 60 governments were able to follow international best 40 practices, such as allowing operators to develop their own infrastructure and provide new services 20 4.91 in many markets. Liberalization in some countries 0 2003 2005 2007 2009 2011 went hand in hand with the political economy of privilege. For instance, new entrants had to partner Mobile penetration with local elites to secure a license (Tunisia), and Fixed broadband penetration issuing a third license took a relatively long time in MENA compared to other regions. Yet even mov- Source: Rossotto 2011, based on the International Telecommunication Union World ing from a monopoly to duopolistic competition Telecommunications Indicators database, http://www.itu.int/ITU-D/ict/statistics/. produced staggering results, both in investments Note: MENA = Middle East and North Africa. and new customers (for example, the customer base rose from 1 million to 15 million in two years in and Central Asia. Opening the sector to competition Morocco) (see figure B7.5.1). could create new investments and great potential The telecom sector also reveals the cost of not for employment, especially in the rising outsourcing liberalizing. International calling and broadband industry, which suffers from the cost of connecting services remain under the monopoly of national tele- to the world. Studies suggest that a broadband pen- com companies in virtually all countries in MENA, etration of 10 percent could increase annual growth and, as fi gure B7.5.1 shows, the broadband sector of gross domestic product by 0.24–1.50 percentage has experienced very limited growth. Broadband points and indirectly create 1.5–4.5 jobs (in account- services remain expensive and poor in quality. ing, legal, and other services) for each job it creates In 2009, the costs per megabyte (MB) in Jordan, directly. Morocco, and Tunisia were 5–50 times those in Bulgaria, Lithuania, and Romania, and MB used per Source: Rossotto 2011; Katz et al. 2010; Analysis Mason 2010; McKinsey 2010; capita were between 1/6 and 1/10 of those in Europe Qiang and Rossotto 2009; and Czernich et al. 2009. ALIGNING INCENTIVES TO INVEST, INNOVATE, AND GENERATE EMPLOYMENT 217 prices, small customer bases, and low qual- FDI, which can be a vehicle of knowledge ity of service. The consequences for the rest transfer. of the economy have been severe, especially for the development of an advanced infor- Highly innovative firms in MENA mation technology (IT) sector and out- sourcing industries (box 7.5). In both cases, In many countries in the region, a subset of local political-economy considerations firms has shown a high degree of innovation played a role in governments’ decisions on and the ability to export their products (see whether to foster or limit competition in box 7.6). Although such firms are excep- the sector. tional, they constitute important role models. An attractive investment climate can Successful examples of innovation by also facilitate the entry of export-oriented firms in the region do not seem to be the BOX 7.6 From emerging role models to innovation clusters: The experience of firms in Lebanon and Morocco The example of two Moroccan firms illustrates 10 percent of sales, which compares favorably the potential of innovation clusters. Such fi rms are with its competitors. R&D personnel represents becoming important role models for educated tech- 35 percent of its total employees, and all of them nical professionals in Morocco. are Moroccans. Hightech Payment Systems (HPS) was founded Nemotek, established in 2008 by the Caisse de in 1995 by a group of Moroccan consultants and Dépôt et de Gestion, is located at the new Technop- experts in monetics (electronic banking). More olis Park near Rabat. With its worldwide customer than 90 percent of its products are exported to portfolio, Nemotek is considered one of the lead- well over 100 clients in 60 different countries in ing manufacturers of wafer-level cameras and is a Europe (26 percent), the Middle East (36 percent), world leader in microcamera technologies. In fact, Asia, the United States, and Africa; it had rev- it was the fi rst company to buy a Micro Via Pack enues of more than US$30 million in 2010. The (MVP) license from Tesseraa and the fi rst to indus- domestic market represents less than 10 percent trialize it. Signifi cantly, it benefi ts from relatively of the total. With branches in Paris and Dubai, it high intellectual property protection compared to has become a global company with three regional Asia. R&D is conducted in partnership with the offices and two joint ventures in Bahrain (GPS) Moroccan Association for Science, Innovation and and in Mauritius (ICPS). The PowerCARD soft- Research, a public foundation specializing in nano ware currently operates 100 electronic fi nancial materials, microelectronics, and biotechnology. transaction sites, which manage a total of 300 Nemotek relies heavily on the Moroccan diaspora: financial institutions. Its human capital is the it has been able to attract numerous highly qualified vital key to its success: its labor force grew rap- nationals back to the country, turning the “brain idly from 179 in 2009 to 350 in 2011, most of drain” into a “brain gain.” It draws on other for- them under 35. They are recruited mostly from eign competencies by employing engineers from all the Ecoles d’Ingénieurs in Morocco. These uni- over the world. versity graduates have satisfactory performances MultiLane SAL is a Lebanese company that once they receive specifi c training related to mon- develops technology and products for the optical etics, which is not taught anywhere in the coun- communication market. MultiLane’s products sup- try. Returning Moroccans from the diaspora have port the global tier-1 suppliers of communication proved central to the company’s successful innova- infrastructure and enable leading semiconductor tive strategy. The company has collaborators from companies to offer communication equipment for 60 different countries. Its R&D budget reached the next-generation network. Customers include (continued next page) 218 JOBS FOR SHARED PROSPERITY BOX 7.6 From emerging role models to innovation clusters: The experience of firms in Lebanon and Morocco (continued) Broadcom, Cisco Systems, Google, and Intel Corpo- For 25 years, Fadi lived and worked outside Leba- ration. The company’s key asset is its 30 locally hired non. During this time, he started and sold three engineers. The team’s expertise was developed fol- high-tech ventures in the United States. In 2006, lowing intensive applied training to each of its team Fadi returned to Lebanon with the aim of creating members. MultiLane invests extensively in training, technology and products from Lebanon to serve the providing at least a year to each engineer. This model global market. MultiLane and its subsidiary dem- is rarely adopted in the industry, because operating onstrate how a highly trained workforce can attract companies usually prefer to attract expensive and multinational companies to invest and open R&D highly experienced staff (who may be easily lured centers. Fadi is currently developing Houmal Tech- away again). In 2009, the company offered an inten- nology Park to incubate companies that develop sive six-month course to the graduating class at the products and technology for the global market. Lebanese University in very-large-scale integration (VLSI) and system-on-a-chip design and then hired Sources: World Bank; Aridi 2011. a. Tessera Technologies, Inc., is a company that develops, licenses, and delivers many of the team under a subsidiary. MultiLane innovative miniaturization technologies and products for electronic devices, founder, Fadi Daou, is a repatriating entrepreneur. http://www.tessera.com/Pages/tessera.aspx. direct result of government-led initiatives. potential for job creation of comprehen- By contrast, qualitative interviews with suc- sive investment climate reforms. On the cessful entrepreneurs suggest the opposite: other hand, the ad hoc and sporadic nature companies often cited their ability to mini- of these successes underscores the need mize their interactions with government for broad-based and multifaceted regula- bureaucracies as one of the reasons for their tory reform that systematically addresses success. Companies that export software or underlying political-economy constraints. services that do not require relying on cus- An example is provided by the experience toms officials or trade regulations and that of the International University of Rabat, to operate in a relatively uncontested domestic date the only private university in Morocco. market (with few incumbents lobbying to It has demonstrated promising achieve- restrict competition among successful inno- ments in R&D, provision of internationally vators) prove successful despite government competitive education, creation of start-ups, intervention, rather than because of it. and private sector participation (box 7.7). In the same spirit, some countries have However, the university’s success largely created export-processing zones as a way to rests on its ability to operate as a private reduce bureaucracy. While this approach was autonomous entity, free from the regula- able to attract FDI in some countries such as tions that govern all other institutions of Tunisia, the dual system of export zones on higher education. the one hand and ordinary domestic enter- prises on the other undermines the ability How institutions can facilitate of domestic actors to benefit from the same innovation opportunities for growth. Some salient examples of successful incu- The study of successful enterprises in bation of innovation have benefited from MENA points toward the importance of the direct protection of high-ranking cham- institutions, such as diaspora networks, pions in the government. On the one hand, private universities, and venture capital these success cases attest to the enormous funds, that mitigate market failures like ALIGNING INCENTIVES TO INVEST, INNOVATE, AND GENERATE EMPLOYMENT 219 BOX 7.7 The International University of Rabat The International University of Rabat (UIR) is the bypass the civil service code and design its own fi rst private university in Morocco, led by a member terms of recruitment and management of talent, of the Moroccan diaspora, a former professor at the and, above all, diaspora members. Forty percent Polytechnic University of Nantes. Established in 2006 of the positions are reserved for faculty working in in the framework of a new law, the UIR is an inter- partner universities, which allows better integra- nationally oriented, R&D-driven university housed tion with world-class universities. The university is in the Technopolis of Rabat, an industrial area that able to pay internationally competitive salaries for a aspires to host high-tech industries. In 2012, the uni- few magnet department heads, who in turn attract versity began providing its fi rst undergraduate and talented younger researchers. Staff members are graduate courses in English and French, at high but expected to connect with clients in the local private internationally competitive fees. The aim is to attract sector and are rewarded on results (such as patents, mostly elite students from the whole African conti- research grants, and contracts with the private sec- nent who would normally study in Europe. UIR, tor). The R&D strategy is tailored to generating under contract with the government of Morocco, is “inexpensive innovations” for the domestic and a public-private partnership that can use academic African economy: (1) infrastructure development personnel from the public sector and acquire recogni- for transportation, tourism, and affordable hous- tion of its degrees and diplomas from the Ministry of ing; (2) renewable energy using local sources; and Higher Education. In addition, the university enjoys (3) local niches: railway, naval, automobile, and the patronage of a powerful segment of the local aerospace engineering. Recent successes include an economy: it is built on land donated by the king and R&D contract with Local Micro-Camera Export counts leading local financial institutions and compa- Company, the creation of a start-up producing pat- nies among its board and financers. ented solar and wind-fueled devices, and the tech- However, the institution’s autonomous status is nical assistance to the government-owned foreign the crucial element of its success. The school can currency exchange office. information asymmetries. Although these investing in early-growth Lebanese compa- institutions are typically not state led, the nies (see box 7.8). state still has an important role to play in creating an enabling environment (both de High-skilled diaspora networks jure and de facto) that facilitates the efficient Another strategy for stimulating innovation functioning of these institutions. is to encourage the transfer of knowledge by leveraging highly skilled diasporas, which Private universities with a high degree of the MENA region has in abundance (see autonomy World Bank 2011). Some of the examples Private universities, for example, have of successful firms illustrated in box 7.6 proved to be effective incubators of innova- involved diaspora returnees, who took tive ideas in MENA. Apart from the exam- advantage of local high-skilled labor and ple of the International University of Rabat, their knowledge of global markets to create a second case in point is the University of successful export-oriented businesses. But St. Joseph in Lebanon. The university was diasporas do not necessarily need to relocate able to generate a successful cluster of ser- permanently in the home country. Transfer vice industries that targeted export markets can occur through “diaspora institutions,” by creating a private technology park and which can be formal or informal networks launching a seed capital fund dedicated to of practice that domestic actors can tap 220 JOBS FOR SHARED PROSPERITY BOX 7.8 Lebanon’s Berytech Technology Park In Lebanon, the Berytech Technology Park, initiated technology companies, providing both the equity by the private University of St. Joseph, is a multi- capital and the guidance needed to succeed. The sector park focusing on services: advertising, busi- fund invests from US$100,000 to US$1.2 million in ness coaching, accounting, management consulting, any single investment—a range not generally served computer hardware and peripherals, and clinical by formal venture capital funds—and helps secure research. a Berytech launched a seed capital fund additional fi nancing if and when needed. Focus sec- (Berytech Fund)b dedicated to investing in early- tors include ICT, energy, agrofood (food process- growth Lebanese companies whose business is in ing), environment, health care and medicine, and information and communication technology (ICT) media and communications. in exchange for equity ownership. With over US$6 Source: Aridi 2011. million under management, Berytech Fund stimu- a. http://www.berytech. org/component/option,com_wrapper/Itemid,392/ lates high-tech growth by investing in early-stage lang,en/. b. http://www.berytechfund.org/about.php. into in various ways to solve business prob- Justifiable government interventions to lems. The production structure in MENA support innovation countries relies heavily on micro and small Even if incentives for innovation are aligned, enterprises: for these fi rms, the challenges of market failures and other constraints may access to the know-how and networks they prevent firms from innovating, especially need to upgrade production and to be com- given the relative scarcity of role models. For petitive in larger international markets can instance, fi rms may not have the capacity to often be formidable. In this context, dias- access global sources of knowledge easily or pora institutions can help reduce these costs to identify and connect with the actors that and act as important agents of transforma- can respond to their R&D needs. Institu- tion. These institutions play a facilitating tions dedicated to generating knowledge and role, matching diasporas with local entre- research, such as universities, may operate preneurs who have the potential to grow but in isolation from the private sector, because face constraints that require external exper- they lack incentives or opportunities to col- tise to resolve. laborate. Finally, firms may have trouble Diaspora engagements are about entrepre- obtaining risk-tolerant and forward-looking neurship and risk taking in expectation of fi nancing, such as venture capital. higher-than-usual returns. As such, they can However, the legacy of relations between and should not be mandated, administered, the state and the private sector described or directed by the state. However, they can earlier calls for strong caution toward any be nourished and supported through mecha- interventionist policy. If governments are nisms that incentivize domestic firms to part- committed to intervening, they should oper- ner with diaspora experts, and they require ate in a way that facilitates the process rather the same conditions needed by other civil than leading it, share the risk with the private society initiatives to thrive: freedom of asso- sector, and institute the appropriate gover- ciation and of information exchange across nance structures to prevent capture of ben- borders. An example is the nongovernmental efits by special interests. For this reason, the organization (NGO) Maroc Entrepreneurs, program designs will need to ensure that the which has become a successful platform for performance of any institution can be moni- recruitment of diaspora talent abroad within tored and that outcomes can be evaluated the Moroccan private sector (see box 7.9). ALIGNING INCENTIVES TO INVEST, INNOVATE, AND GENERATE EMPLOYMENT 221 BOX 7.9 Diaspora-led intermediation of talent and entrepreneurship The NGO Maroc Enterpreneurs was founded in London and Paris. The organization offers business France by seven self-starting students of business training and every year has a business plan competi- administration to foster entrepreneurship, particu- tion for graduating students. The authors of winning larly in Morocco, and to link fi rms looking to recruit projects are matched with experienced Moroccan talent in Morocco with high-skilled graduates. A businessmen as mentors. The government can facili- decade after its foundation, the organization is still tate these spontaneous initiatives by its openness to run mostly by volunteers but now counts a net- partnering with such organizations on specific activ- work of about 3,000 students and 7,000 Moroccan ities but should not direct them. graduates in economics and business, with offices in Source: Kuznetsov and Morgandi 2008. regularly. Selection criteria should be public innovation agents with domestic actors in and, where possible, the subject of consul- search of innovation. Competitive funding tation with relevant stakeholders; in addi- schemes have proven to be effective catalysts tion, programs should systematically publish of innovation. In the medium term, con- information on beneficiary firms and the tests that encourage collaborative innova- subsidies that they receive, as well as allow- tion efforts could become a very important ing independent access to data and surveys to source of funding for national R&D. They evaluate interventions. When feasible, impact should involve participation of international evaluations should be built in at the start of partners and rely on international peer any intervention. review. The main objective of such fund- ing schemes is to provide an opportunity Public-private venture capital funds for dynamic and entrepreneurial individu- Some governments have been able to jump- als and organizations to develop their own start venture capital for start-up enterprises. projects in collaboration with each other However, as these financing schemes can and with international players. The science be captured easily by vested interests, the and technology ministries of Mexico and participation of the private sector is a fi rst the Russian Federation have implemented important element for ensuring that risk such competitions. taking is directed toward investments where the potential for success is highest. External monitoring mechanisms are also impor- Linking universities to local knowledge tant. Highly skilled diaspora networks can producers and the private sector sometimes provide a check and balance on some of the more sophisticated institutions, Rather than intervening in targeted fi rms, such as venture capital funds, by acting as governments should pursue reforming the authoritative agents with specialized knowl- governance of universities (the main gen- edge on the selection of promising projects erator of talent pools) to help them become for government cofi nancing. 2 active participants in the innovation pro- cess. The following section describes the les- Competitive funding schemes with sons learned from the study of world-class diasporas research universities in developing countries A second example of the role that govern- and discusses their relevance for universities ments can play is facilitating matches of in MENA. 222 JOBS FOR SHARED PROSPERITY Research universities promote innovation exception of Morocco, Tunisia, and some in three ways: fi rst, by preparing talent for GCC states, most MENA countries seem innovation (teaching); second, by engaging to lack these elements (see Nour 2011, 409; in research; and third, by commercializing Arvanitis 2007, 36). Regarding university research that is valued by markets, through governance, an appropriate regulatory the creation of patents, studies, and techni- framework, strong and inspiring leadership, cal assistance. Because of demographic pres- and adequate management significantly sure and the expansion of tertiary education influence the ability of research universities in the Arab world, universities have focused to prosper (Altbach and Salmi 2011, 331). resources mainly on the fi rst mission (teach- Autonomy in a wide array of domains is also ing), less on the second (research), and not essential for (1) mobilizing supplementary at all on the third (creating commercial resources from nonpublic actors; (2) deter- value). mining attractive wage packages to attract Just as the region needs role models of the best staff; (3) boosting the international knowledge-based fi rms, it also needs exam- dimension of study programs; (4) using Eng- ples of new and effective research universi- lish as a teaching language to enhance the ties capable of producing knowledge val- possibilities of recruiting high-level academ- ued by markets. The paradox of the best ics; and (5) innovating in curriculum and research universities is that “everyone wants pedagogy (Altbach and Salmi 2011, 334). one, no one knows what it is, and no one Despite the reform processes mentioned knows how to get one” (Salmi 2009, 15) (see earlier, most of the public institutions of box 7.10). higher education in MENA are still unable T he alignment of talent, abundant to address most, or any, of these issues resources, and favorable governance is nec- autonomously. 3 To give but one example, essary to creating the enabling environment in Morocco public institutions of higher for innovation. At a minimum, this con- education can act on points (2) and (3) on vergence implies the existence of national their own initiative (no incentives are being science or innovation policies with links provided), while points (4) and (5) are still to higher education policy, clearly defi ned overlooked by the relevant authorities and roles for research universities, and coop- point (2) would require a systemwide reform eration with dynamic companies. With the that might prove difficult, if not outright BOX 7.10 Common characteristics of successful research universities The main fi ndings from a study of 11 world-class • With the exceptions of Japan and the United States, universities in nine countries are relevant for MENA they are overwhelmingly public institutions. (see Altbach and Salmi 2011). The study determined • They are cost-intensive as adequate staffi ng and that successful research universities share common infrastructure are both expensive and necessary characteristics that distinguish them from other for success. Therefore, they also need an appropri- institutions of higher education and therefore call ate, and sustained, budget. for a tailored policy response: • They can potentially generate signifi cant income through tuition fees that students are willing to • They are part of a differentiated academic system pay, through intellectual property and other inno- in which they are recognized as quality leaders vations that can be sold to the corporate sector, and supported accordingly. (continued next page) ALIGNING INCENTIVES TO INVEST, INNOVATE, AND GENERATE EMPLOYMENT 223 BOX 7.10 Common characteristics of successful research universities (continued) and—in some countries—through the manage- The study also defi nes a theoretical framework ment of an endowment generated by fund raising. presenting the key factors behind the success of • Their success is stronger when strong ties exist these premier universities (see figure B7.10.1). between them and nonuniversity research institutes.a Source: Salmi 2009. • Building a successful research university takes Note: World-class university. time—possibly several decades—and it is a complex a. Altbach and Salmi 2011, 24–25. b. Altbach and Salmi 2011. process that requires constant attention, adaptation, and fine-tuning.b FIGURE B7.10.1 Characteristics of a world-class university Concentration of talent Students Teaching staff Researchers Internationalization Research Graduates output Supportive World- class regulatory university framework Abundant Public budget resources Autonomy Favorable resources Endowment revenues Technology Academic freedom governance transfer Tuition fees Leadership team Research grants Strategic vision Culture of excellence impossible. For an example of how to that few of the highly educated—those most enhance university autonomy, see box 7.11. likely to create the greatest returns from entrepreneurship—are actually becoming entrepreneurs. A combination of constraints Sustaining low-skilled and lack of incentives may prevent them entrepreneurs and promoting from making that choice. In contrast, the the potential of the highly skilled same profi ling exercise showed that the less educated have a high propensity for entering Fostering entrepreneurship with a subsistence self-employment, that the pro- tailored approach ductivity of their enterprises tends to be low, The distribution and characteristics of entre- and that subsistence self-employment is still preneurs, discussed in chapter 1, suggests highly correlated with poverty. 224 JOBS FOR SHARED PROSPERITY BOX 7.11 Enhancing university autonomy: The case of Shanghai Jiao Tong University Despite a context different from MENA’s—that (4) the investment in social sciences to promote is, one in which private-public partnerships are the development of consulting services to the gov- facilitated by the government’s power over “private” ernment and local organizations. companies—the reform process of China’s Shanghai • Using research resources for talent development Jiao Tong University (SJTU) is a good illustration by increasing research funding and by promot- of concrete measures that enhance autonomy. To ing higher education and research standards at all strengthen its international reputation, the SJTU educational levels, including multiplying partner- embarked on a reform program that focused on four ships with industry. The latter take the form of goals: joint research undertakings, student internships in partner companies, and supervision of postgradu- • Strengthening international publications by ate students by experienced, active, engineers. requiring students of the graduate school in sci- • Promoting internationalization through bilingual ence and engineering to publish in a journal cov- (Chinese and English) courses and through the ered by the Science Citation Index, as well as by participation of its students in study tours, univer- providing incentives for such publications. sity exchanges, internships, and jointly supervised • Encouraging applied research and technology PhD programs. The university is actively engaged transfer through (1) the commercialization of in the development of dual degree programs and patents through the dedicated SJTU technology joint institutes with other institutions of higher transfer center; (2) the creation of an informa- learning around the world. tion platform on patents; (3) the authorization for staff to invest and gain a commensurate personal reward for any commercialized innovation; and Source: Liu, Wang, and Cheng 2011, 50–55. These facts suggest that two types of build essential skills and reduce the risk of entrepreneurship interventions may be useful failure. in MENA, if focused on two specific target groups. The first are interventions that facili- Improving the climate for start-ups tate entry of the highly skilled into entrepre- According to economic theory, individu- neurship. The second are interventions that als may engage in entrepreneurship if the assist the existing stock of low-skilled entre- endeavor is likely to be profitable enough to preneurs in upgrading their skills and that offset the opportunity cost of not working reduce constraints to investment and growth. in the wage sector, which for high-skilled individuals offers relatively high returns. Thus all the factors that increase the costs Facilitating entrepreneurship for more of setting up and running a business or that educated individuals affect the chances of a business to grow In addition to reforming the investment are likely to influence the entrepreneurship climate and improving access to credit for decision of the highly skilled population. start-ups, a number of well-targeted and Today in MENA most entrepreneurship implemented policies can help promote cases seem to follow a strong intergener- entrepreneurship among the most produc- ational pattern: 72 percent of the young tive segment of the active population, the (non-household-head)4 employers in Jordan high-skilled individuals. For instance, live in households whose head is either an entrepreneurship training (which remains employer or is self-employed (fi gure 7.1). relatively rare in the region) can help Similarly, more than three-fourths of ALIGNING INCENTIVES TO INVEST, INNOVATE, AND GENERATE EMPLOYMENT 225 FIGURE 7.1 Share of employed individuals whose household head is an entrepreneur, by work status, selected economies in MENA, 2006–10 100 % of head of household in entrepreneurship 75 50 25 0 Employer Employer Employer Employer Employer Wage earner Self-employed Wage earner Self-employed Wage earner Self-employed Wage earner Self-employed Wage earner Self-employed Jordan Egypt, Arab Rep., Tunisia Morocco West Bank and 2006 Gaza Head of household, self-employed Head of household, employer Source: Based on Arab Republic of Egypt’s Labor Market Panel Survey (LMPS) 2006, Jordan’s Labor Market Panel Survey (LMPS) 2010, Morocco’s Household and Youth Survey 2009, Tunisia’s Labor Force Survey 2010, and West Bank and Gaza’s Labor Force Survey 2008. Note: MENA = Middle East and North Africa. Individuals considered are all employed non-household-head people under 40 years of age. young employers and the self-employed Potential of entrepreneurship training in Morocco live in households headed by In addition to improving the business envi- entrepreneurs; these shares are much lower ronment and access to credit, specific train- for wage employees. 5 ing is another way to foster entrepreneur- The overall perception from opinion sur- ship among the highly educated. Many veys in MENA countries is that some of the interventions have focused on education core inputs to setting up a business—such and training, based on the evidence that as permits, credit, and, above all, training— some entrepreneurial traits and skills are are diffi cult to access. The constraints are strongly related to business setup and suc- perceived more strongly in non-GCC coun- cess 6 and that these traits can be taught. tries and decrease slightly as education and Some are implemented through school income increase (figure 7.2), but they remain curricula for relatively early intervention,7 high. High start-up and operating costs are while others cover those who are already in more likely to affect the decision of those the labor market. Skills of interest include who may not want to work informally or of basic business and managerial skills, fi nan- those who may have a higher profit thresh- cial literacy and capability, and vocational old, such as the highly educated group. and life skills. In addition to this main For instance, broadening access to profes- intervention, counseling and mentoring are sions, as discussed earlier in this chapter, in often provided.8 countries where these are still highly restric- According to a recent inventory of inter- tive and opaque is an important avenue for ventions for youth employment, only a small increasing entrepreneurship among the highly number of existing active labor market pro- educated. grams in MENA focus on entrepreneurship 226 JOBS FOR SHARED PROSPERITY FIGURE 7.2 Main constraints to starting a business in MENA, by level of education, 2011 90 80 70 60 Percent 50 40 30 20 10 0 Primary Secondary Tertiary Primary Secondary Tertiary Primary Secondary Tertiary Hard to obtain loans Difficult to obtain government No access to training for starting permits business MENA GCC countries MENA non-GCC countries Source: Gallup World Poll Survey 2011. See appendix for more information on the survey. Note: MENA = Middle East and North Africa; GCC = Gulf Corporation Council. promotion (table 7.2) (Angel-Urdinola, Sem- study also underscored that high wages in lali, and Brodmann 2010). Of the existing the formal public sector continue to act as a programs, only two have been rigorously disincentive to take risk: the impact evalu- evaluated; only one program out of those ation found that the training had no effect reviewed is found to target women; and on the participants’ preference for the pub- microfi nance programs are less common in lic sector as an employer of choice. MENA than in the rest of the world. An example from which many lessons can Importance of work experience in be drawn is the business plan competition for entrepreneurial success Tunisia, a combination of entrepreneurship The profile and history of entrepreneurs sug- training and customized coaching focused on gests that while tertiary-educated individuals graduating university students (see box 7.12). should be incentivized to set up firms, incen- The program was found to have positive tivizing the entry of first-time job seekers impacts on the acquisition of cognitive and may not be optimal. Only a few inexpe- noncognitive skills, as well as on the pro- rienced individuals are likely to enter into pensity for entering self-employment, which entrepreneurship. Most entrepreneurs have started from a very low base. In addition, accumulated experience and know-how (and the program confirmed the important probably also capital) before transitioning role that both the business climate and into self-employment. Figure 7.3 shows that the incentives in the labor market play in in the case of Egypt and Jordan, about 30–40 whether an individual enters into self- percent of the entrepreneurs ages 25–46 were employment. In fact, the Tunisian graduate already involved in entrepreneurship a decade youth who participated in the business plan earlier, either as heads of their business or as competition mentioned access to fi nance as unpaid family members. Another large share a top constraint to their entering into self- of the entrepreneurs (25–45 percent) previ- employment, and this perception did not ously worked as wage workers, especially in change over the course of the training. The the informal economy. Comparing employers ALIGNING INCENTIVES TO INVEST, INNOVATE, AND GENERATE EMPLOYMENT 227 TABLE 7.2 Programs promoting entrepreneurship in MENA Program name Economy Target group Component Injaz al Arab Jordan; Egypt, Arab Rep.; School students Private firms’ volunteers provide job skills and Tunisia; Lebanon; West (grade 7 to university) financial literacy training to students in public Bank and Gaza; Morocco schools Achievement Egypt, Arab Rep. Youth Private sector volunteers share experience and provide entrepreneurial training Textile merchandiser program Egypt, Arab Rep. Unemployed youth Provide textile merchandise (vocational training) skills training as well as life skills Positive youth initiative Egypt, Arab Rep. Unemployed youth Vocational training through apprenticeship and grants provided for business setup Microcredit youth lending program Egypt, Arab Rep. Youth Group-based microcredit Microcredit program Egypt, Arab Rep. Youth Microcredit and business training Entrepreneurship Training for West Bank and Gaza Women Entrepreneurial training and access to funding Women Artisans Know about your business Syrian Arab Republic Youth Business education Graduate entrepreneurship Tunisia Unemployed university Entrepreneurial and IT training program through IT graduates Youth microfinance initiative Yemen, Rep. Youth Entrepreneurship training and microfinance Turning theses into enterprises Tunisia University students Business training, coaching, and financial prize for business plan competition winners Microcredit Morocco Rural households Microfinance Source: World Bank Youth Employment Programs Inventory. Note: MENA = Middle East and North Africa; IT = information technology. BOX 7.12 Entrepreneurship training in Tunisia In 2009, Tunisia introduced an entrepreneurship among a sample of university students. Results show track into the applied undergraduate curriculum, that benefi ciaries of the pilot program were 61–88 which entailed training in business and entrepre- percent more likely on average to be self-employed neurship as well as coaching. Students could choose compared with graduates from the control group. the entrepreneurship track as an alternative to the The program was also effective in fostering business standard curriculum. As a graduation requirement, skills, expanding networks, and affecting entrepre- students had to prepare and submit a business plan neurial traits and behavioral skills, as well as rais- instead of a traditional undergraduate thesis. Of ing graduates’ sense of opportunities and optimism 1,800 eligible students, 1,702 applied in the first toward the future. Participants entering entrepre- year. Given capacity constraints, half the applicants neurship training placed a lower value on agreeable- were randomly selected for the newly established ness (the literature regards placing a lower value on entrepreneurship track, while the other half were agreeableness as an important quality for success in assigned to the standard curriculum. business professionals or managers), but the entre- The rigorous evaluation by Premand et al. (forth- preneurship track increased extroversion. This fi nd- coming) of this entrepreneurship program is the ing suggests that different behavioral skills may be fi rst attempt to take an experimental approach to shaped by different types of training or activities. quantifying the effectiveness of business training Source: Premand et al., forthcoming. 228 JOBS FOR SHARED PROSPERITY FIGURE 7.3 Transition matrix of self-employed and employers known about constraints to (largely infor- ages 25–46 in the Arab Republic of Egypt and Jordan, by work mal) subsistence entrepreneurs in MENA. A status, 1998–2006 and 1999–2010 number of programs have been implemented percent around the developing world, however, to promote low-scale entrepreneurship; and a. Self-employed b. Employers the barriers that they try to reduce usually 100 100 include the lack of generic business skills or experience and the lack of start-up or work- 75 75 ing capital.9 Percent Percent 50 50 Access to microcredit Expansion of access to credit can ben- 25 25 efit microenterprises. An impact evalu- ation of a microfi nance program in rural 0 0 Morocco (Crépon et al. 2011) showed only Egypt, Jordan Egypt, Jordan Arab Rep., 1999–2010 Arab Rep., 1999–2010 mild effects of credit expansion on start- 1998–2006 1998–2006 ups. In contrast, however, expansion of Unpaid family workers Self-employed and employers access to credit had important effects on Informal wage employees Formal sector existing entrepreneurs, who cut back their Inactive and unemployed spending on consumption and nondurable goods in order to finance the expansion Source: World Banks staff, using the Arab Republic of Egypt’s Labor Market Panel Survey (LMPS) of their businesses. For most subsistence 1998–2006 and Jordan’s Labor Market Panel Survey (LMPS) 2010. See the appendix for more infor- mation on these surveys. entrepreneurs, the main challenge may not be entry but business growth. The posi- in the top and in the bottom wealth quintiles tive effect of releasing credit constraints for Jordan and Morocco also shows that the is also documented by De Mel, McKenzie, high-earning employers stand out not only and Woodruff (2008a, 2008b), who show for having higher education but also for that providing cash or in-kind transfers being older than the low earners. This finding to small retail fi rms in Mexico had posi- could be related to the long time it takes for tive effects on profits, but the effects were an entrepreneur to reach success but perhaps higher for the fi rms that were more credit also to the fact that more successful entrepre- constrained. neurs have more experience. According to the evidence, then, govern- Tailored training in essential business skills ments should be cautious about promoting A second set of policies can help foster entrepreneurship as a short-term measure for skills in existing microenterprise owners. reducing the unemployment of first-time job Informal enterprise owners can benefit seekers. Instead, the government could help from training in basic accounting, literacy, young people acquire work experience early and numeracy but also from workshops for on and then encourage the better-performing technological upgrading of production and individuals to shift from wage employment IT skills. Such skills may also lead to bet- to entrepreneurship. That way, these new, ter labor market matching by facilitating highly productive entrepreneurs can create the search for wage-labor opportunities. the jobs that will absorb unemployment. However, the lack of a developed market for provision of these skills implies that appropriate design is crucial to incentiv- Improving the productivity of izing providers to offer only relevant and subsistence entrepreneurs work-time-compatible trainings. Demand- Most studies of the investment climate focus driven approaches, such as for example, on formal enterprises, and relatively less is the provision of training vouchers, have ALIGNING INCENTIVES TO INVEST, INNOVATE, AND GENERATE EMPLOYMENT 229 yielded promising results in other contexts, transportation networks are crucial for notably Kenya and Uganda (Van Adams improving access to markets where prod- 2001). uct demand is higher, as well as for access to the digital infrastructure that spreads Business environment, infrastructure, and information on prices and demand oppor- access to technology tunities. Transportation is also important A broader set of policy interventions is likely because it allows individuals to connect to have important spillovers on the welfare directly to markets and reduces dependence of this group of largely poor entrepreneurs. on intermediaries. First, improving the formal business envi- Finally, a number of organizational tech- ronment is crucial to enabling the transi- nologies can improve the efficiency and spe- tion of (subsistence) microentrepreneurship cialization of microentrepreneurs, allowing to wage employment, as many microfi rms them to benefit from economies of scale. are created just because wage jobs were For instance, agricultural cooperatives can lacking. In addition, regulations that dis- be an important vehicle for acquiring high courage formality may undermine employ- fi xed-cost goods such as machinery needed ment growth and productivity-enhancing to process and market products. The legal investments. framework needs to be in place to allow Second, many of the constraints that these entities to operate. Table 7.3 provides microentrepreneurs face relate to the examples of programs developed to address geographic location where the working particular constraints affecting small-scale poor live and operate. Infrastructure and entrepreneurs. TABLE 7.3 Programs for small-scale entrepreneurs in various developing countries Main constraints Types of intervention Basic component Examples Limited skills to start up a business Vocational training Skills training in particular areas of Uganda NUSAF Program, Malawi vocation, such as carpentry, tailoring, Apprenticeship Program, India Self Help and baking, for example Group Limited knowledge to operate a Business training Education on business planning, cost Tunisia business training and planning, business calculation and price setting, market business training for microcredit clients analysis, marketing, customer relations, (Ghana, Peru, Tanzania) and business practice, for example Limited knowledge to make use Financial training Education on the concept of Financial training for existing business of capital and smart investment inflation, interest rate, investment, owners (Bosnia and Herzegovina, decisions to grow a business profit maximization, portfolio, and Indonesia) accounting, for example General lack of career plans and Life skills training Promotion on positive life Social assistance programs in Chile and inability to manage skills, time, perspectives, general problem-solving Nicaragua and money skills in life, and sound decision making Limited access to credit to start, (Micro) credit, subsidies Provision of loans or grants for Microcredit in Bangladesh, India, invest, and expand businesses business operation Pakistan, and Tanzania, among others Risk averseness, high discount (Micro) insurance, Provision of institutional support to Microinsurance in Malawi, savings factor saving purchase insurance or open a savings account account for the future Limited experience and social Counseling, mentoring Hands-on advice for business Graduation model in Argentina and capital operation, application of knowledge, Chile, on-site visit and counseling in the guidance on financial products and Dominican Republic credit, and linkage to the existing market and network Source: Cho and Honorati 2012. Note: NUSAF = Northern Uganda Social Action Fund. 230 JOBS FOR SHARED PROSPERITY Global lessons on the design of is the core driver of the growth of labor entrepreneurship programs productivity. Reforms that can facilitate this process include reducing adjustment costs and The meta-analysis of a recently developed barriers to entry and exit. Facilitating fi rm global inventory of entrepreneurship promo- exit—while potentially increasing short-run tion programs around the world sheds light job losses—will be crucial for job creation in on some of the design features that could the long run as it ensures that resources are enable these programs to work better but reallocated toward their most productive uses also reveals some of their limitations as a and because it encourages entry by reducing policy tool.10 Programs in MENA are likely the costs of failure. It will also be important to yield better results if they combine entre- to create the conditions for incumbent fi rms preneurship interventions with opportunities to invest and grow, given MENA’s limited for broader access to credit—especially for dynamism. Enhancing transparency in public credit-constrained groups, which can include administration and increasing institutional existing entrepreneurs. Adding counseling accountability are important for ensuring improves the effectiveness of training pro- equal opportunities and reducing the scope grams, but the evidence is more mixed in the for corruption. case of counseling and fi nancing. The effi- Addressing structural constraints to the cacy of training programs has also been dem- provision of credit, currently limited for firms onstrated to rise with the skills of the partici- of most sizes, will allow fi rms to invest and pants: high-skilled workers learn faster. help them grow. Promoting trade would not Approaches specifically targeting women only alleviate demand constraints, which pre- may yield better results than generalized vent firms from scaling up, but also serve as a programs that also try to include women disciplining mechanism as firms would have (Attanasio, Kugler, and Meghir 2011; Field to innovate to become and remain globally et al. 2010; Field, Jayachandran, and Pande competitive. 2010).11 According to the inventory men- In spite of the complex political economy tioned earlier, these interventions are still of energy subsidies, removing them is likely rare in MENA countries. Credit for young to pay a triple dividend in job creation: it will people, however, has been particularly suc- reduce the relative costs of labor, it would cessful only in countries where youth unem- immediately incentivize investments in more ployment was not high, which is not the energy-efficient (and likely relatively more case across MENA economies.12 labor-intensive) production, and it would To conclude, caution remains warranted free up fiscal space that can be used to lower when applying any of these approaches, labor taxes and further reduce the (relative) given the dearth of evidence from within the cost of labor. region. In this context, establishing a strong This chapter also discussed the poten- evaluation system—a rigorous impact evalua- tial of innovation policies, in light of the tion would be best—will be essential. effort that many governments are making to actively upgrade the quality of their pro- duction systems. While the importance of Realigning the incentives for innovation is widely appreciated, innova- employment and productivity tion is typically led by the private sector. This chapter reviews a number of policy However, market failures in innovation may options that could unleash the repressed persist that would leave a role for govern- potential of the private sector in MENA to ments to play: appropriately aligning incen- create more employment, and more employ- tives for innovation is crucial. This goal may ment of better quality, along the whole spec- require facilitation of knowledge diffusion, trum of fi rms. for example, by creating forums where inno- Boosting labor demand requires catalyz- vating agents and fi rms could meet, as well ing the process of creative destruction, which as by enabling firms to trade freely across ALIGNING INCENTIVES TO INVEST, INNOVATE, AND GENERATE EMPLOYMENT 231 borders. Reforming university governance is 7. Organizations such as the Kauffman Foundation also likely to help realign the incentives for and Junior Achievement, for example, focus on creating positive and tangible spillovers to promoting entrepreneurship curricula as a part the private sector. of primary and secondary education, while a number of interventions including microcredit Similarly, increasing the rate of participa- and training programs target those who are tion in entrepreneurship among the current already in the labor force. and growing stock of highly educated indi- 8. The provision of counseling is particularly viduals will require the general improvement prevalent among programs in the Latin of the investment climate for this option to American and Caribbean region, such as become more attractive. However, at the Chile’s Solidario and Argentina’s Jefes y Jefas. margin, specific programs could release some 9. See the World Development Report 2013 on of the immediate skills-constraint to entry. jobs for a comprehensive review of the evi- In addition, it is important to address the dence (World Bank 2013). problems faced by the very large but unno- 10. Building on Cho and Honorati (2012), the ticed segment of the private sector—the metaanalysis focuses only on studies with rigorous impact evaluation and identifies 37 mostly informal microentrepreurs, who interventions around all regions, including cannot be absorbed into the wage market in the two from the MENA region. Since meta- the short term but who would benefit from analysis identifies the extent to which different a range of cross-sectoral policies to improve settings of studies—different design, outcomes, their productivity. and methods of analysis, among others—affect reported results, this is particularly useful for synthesizing multiple studies with variations in Notes multiple aspects like the studies on entrepre- neurship programs. In addition, World Bank 1. See World Bank (2002) for a discussion on (2013) provides a comprehensive review of the designing competition authorities. evidence. 2. An example is Fundación Chile, created as a 11. For a comparison, note that programs target- nonprofit corporation by the Chilean gov- ing only youth do not necessarily improve the ernment and the conglomerate ITT in 1976. effects on youth. Fundación Chile became the key player in 12. Whether the labor market is favorable to youth Chile in renewable resources. This foundation is measured as the ratio of the youth unem- is, for instance, widely credited with turning ployment rate over that of adults. Compared Chile (a country with no native salmon popu- to adults, if the youth unemployment rate is lation) into the world’s second-largest salmon high, the labor market is considered less favor- producer. able to youth. 3. Please refer, for example, to the 2009 Arab Knowledge Report classification system (see Mohammed bin Rashid Al Maktoum Foundation and UNDP 2009). References 4. Data limitation does not allow us to capture Abouleinein, S., H. El-Laithy, and H. Kheir- the profession of the parents of young entre- El-Din. 2009. “The Impact of Phasing Out preneurs who have left their household. Subsidies of Petroleum Energy Products in 5. Data limitations allow us to detect only the Egypt.” Working Paper 145, Egyptian Center work status of entrepreneurs who are not for Economic Studies, Cairo. household heads, thus, those who are rela- Altbach, P. G. 2004. “Globalization and the tively young. University: Myths and Realities in an Unequal 6. For example, Ciavarella et al. (2004) using World.” In The NEA 2005 Almanac of Higher data from the United States find a strong rela- Education, 63–74. Washington, DC: National tionship between some attributes of person- Education Association. ality (conscientiousness, emotional stability, Altbach, P., and J. Salmi. 2011. The Road to openness, agreeableness, and extroversion) and Academic Excellence: The Making of World- business survival. Crant (1996) also points to Class Research Universities. Directions in personality as a predictor of entrepreneurial Development series. Washington, DC: World intentions. Bank. 232 JOBS FOR SHARED PROSPERITY Analysis Mason. 2010. “Analysis of Economic Algeria, the Arabic Republic of Egypt, Jordan, Impact of Wireless Broadband in India.” Lebanon, Morocco, the Syrian Arab Republic, Discussion document. http://w w w.gsma and Tunisia. http://www.cuts-ccier.org/pdf/ .com/spectrum/wp-content/uploads/2012/03/ Competition_and_Consumer_Policies_in_ amindiaexecsummaryfinal.pdf. MENA_Countries.pdf. Angel-Urdinola, D., A. Semlali, and S. Brodmann. Czernich, N., O. Falk, T. Kretschmer, and 2010. “Non-public provision of Active Labor W. Ludger. 2009. “Broadband Infrastructure Market Programs in Arab-Mediterranean and Economic Growth.” CESifo Working Countries: An Inventory of Youth Programs.” Paper 2861, Center for Economic Studies Social Protection Discussion Paper 1005, Group, Munich, Germany. World Bank, Washington, DC. De Mel, S., D. McKenzie, and C. Woodruff. Aridi, A. 2011. “Case Studies in Emerging High- 2008a. “Are Women More Credit Constrained? Innovation Firms in MENA Countries.” Experimental Evidence on G ender and Working paper, World Bank, Washington, Microenterprise Returns.” American Economic DC. Journal: Applied Economics 1 (3): 1–32. Arvanitis, R. 2007. ESTIME: Towards Science and ———. 2008b. “Returns to Capital: Results Technology Evaluation in the Mediterranean from a Randomized Experiment.” Quarterly Countries: Final Report. IRD Project No. Journal of Economics 123 (4): 1329–72. INCO-CT-2004-510696. Paris: Institute for Fattouh, B., and L. El-Katiri. 2012. “Energy Development Research. Subsidies in the Arab World.” Arab Human Attanasio, O., A. D. Kugler, and C. Meghir. Development Report Research Papers Series, 2011. “Subsidizing Vocational Training for United Nations Development Programme. Disadvantaged Youth in Developing Countries: http://www.arab-hdr.org/publications/other/ Evidence from a Randomized Trial.” American ahdrps/ Energy%20Subsidies-Bassam%20 Economic Journal: Applied Economics 3 (3): Fattouh-Final.pdf. 188–220. Field, E., S. Jayachandran, and R. Pande. 2010. Cafiero, C. 2009. “Study on Supply and Demand “Do Traditional Institutions Constrain Prospects for the Major Syrian Agricultural Female Entrepreneurship? A Field Experiment Products.” Ministry of Agriculture and on Business Training in India.” American Agrarian Reform, National Agricultural Policy Economic Review 100 (2): 125–29. Center. Damascus, Syria. Field, E., R. Pande, J. Papp, and N. Rigol. 2010. Cho, Y., and M. Honorati. 2012. “Entrepreneurship “Term Structure of Debt and Entrepreneurial Prog ra ms in Developing Cou nt ries: A Behaviour: Experimental Evidence from Meta Analysis.” Working paper, Human Microfinance.” Working paper, Harvard Development Network Social Protection, World University, Cambridge, MA. http://scholar. Bank, Washington, DC. harvard.edu/files/repayment_default_dec19- Ciavarella, M., A. Buchholtz, C. Riordan, R. 1_0.pdf. Gatewood, and G. C. Stokes. 2004. “The Javorcik, B., and T. Harding. 2011. “Roll Out the Big Five and Venture Survival: Is There a Red Carpet and They Will Come: Investment Linkage?” Journal of Business Venturing Promotion and FDI Inflows.” Economic 19: 465–83. Journal of the Royal Economic Society 121 Crant, M. 1996. “The Proactive Personality Scale (557): 1445–76. as a Predictor of Entrepreneurial Intentions.” K at z , R . , S . Vat e rl au s , P. Z e n h äu s e r n , Journal of Small Business Management 34 (3): S. Suter, and P. Mahler. 2010. “The Impact 42–49. of Broadband on Jobs and the German Crépon, B., F. Devoto, E. Duflo, and W. Parienté. E conomy.” Inte reconomic s: Re vie w of 2011. “Impact of Microcredit in Rural Areas European Economic Policy 45 (1): 2. of Morocco: Evidence from a Randomized Kuznetsov, Y., and M. Morgandi. 2008 “Why Evaluation.” Working paper, Massachusetts Is Diaspora Potential So Elusive? Towards Institute of Technology, Cambridge, MA. a New Diaspora Agenda in North African C U TS. 2011. Understanding the State of Economies.” Background paper to Labor Domestic C ompetition an d C on sume r Migration from North Africa: Development Policies in Select MENA Countries. Report Impact, Challenges, and Policy Options of a “Scoping M ission” undertaken in (World Bank 2010). http://siteresources ALIGNING INCENTIVES TO INVEST, INNOVATE, AND GENERATE EMPLOYMENT 233 .w o r l d b a n k . o r g / I N T E M A / R e s o u r c e s / Communication for Development Report MIGRATIONREPORT.pdf. (IC4D), World Bank, Washington, DC. Liu, N., Q. Wang, and Y. Cheng, eds. 2011. Rossotto, C. 2011. “New ICT Sector Strategy Paths to a World Class University: Lessons to Support Inclusive Growth in MENA.” from Practices and Experiences . Global P r e s e n t a t i o n f o r M E N A’s R e g i o n a l Perspectives on Higher Education Series, Management Team Meeting, December 7. vol. 23. Rotterdam: Sense Publishers. https:// World Bank, Washington, DC. www.sensepublishers.com/media/743-paths- Saade, Maurice, Dyaa Abdou, and Nasredin to-a-world-class-university.pdf. El-Amin. 2011. “Agricultural Subsidies in McKinsey. 2010. “Fostering Economic and Social Syria: Enhancing Efficiency, Sustainability Benefits of ICT: The Global Information and Equity.” Unpublished draft paper, World Technology Report 20 09-2010.” World Bank, Washington, DC. Economic Forum, Geneva, Switzerland. http:// Salmi, J. 2009. The Challenge of Establishing w w w3.wefor u m.org /do c s / W E F_G I T R _ World- Class Universities . Directions in Report_2010.pdf. Development series Washington, DC: World Mohammed bin Rashid Al Maktoum Foundation Bank. and the United Nations Development Pogramme Silva J., V. Levin, and M. Morgandi. 2013. (UNDP). 2009. “Arab Knowledge Report 2009: Inclusion and Resilience: The Way Forward Towards Productive Intercommunication for for Social Safety Nets in the Middle East and Knowledge.” UNDP, New York. http://www North Africa. MENA Development Reports .mbrfoundation.ae/English/Documents/AKR- Series. Washington, DC: World Bank. 2009-En/AKR-English.pdf) Van Adams, A. 2001. “Assessment of the Jua Nasr, Sahar. 2010. Egyptian Women Workers and Kali Pilot Voucher Program.” World Bank, Entrepreneurs: Maximizing Opportunities Washington, DC. in the Economic Sphere. World. Washington, World Bank. 2002. Building Institutions for DC: World Bank. Markets: World Development Report 2002. Nour, Samia S. O. M. 2011. “National, Regional Washington, DC: World Bank. and Global Perspectives of Higher Education ———. 2008. Tunisia’s Global Integration: A and Science Policies in the Arab Region.” Second Generation of Reforms to Boost Minerva: A Review of Science, Learning and Growth and Employment . World Bank Policy 49 (4): 387–423. Country Study 45409. Washington, DC: O E C D. S e r v i c e s Tr a d e R e s t r i c t i v e n e s s World Bank. Index (database). Paris, Organisation for ———. 2009. From Privilege to Competition. Economic Co-operation and Development. Washington, DC: World Bank. ht t p: //w w w.oecd.org /tad /ser vicest rade / ———. 2011. Financial Access and Stability: towardsaservicestraderestrictivenessindexstri A Roadmap for the Middle East and North .htm. Africa. Washington, DC: World Bank. Premand, P., S. Brodmann, R. Grun, and ———. 2012. “Jordan Development Policy R. Almeida. Forthcoming. “Entrepreneurship Review.” Working paper, World Bank, Trai n i ng a nd S el f-E mploy ment a mong Washington, DC. Universit y Graduates: Evidence from a ———. 2013. World Development Report 2013: Randomized Trial in Tunisia.” Working paper, Jobs. Washington, DC: World Bank. World Bank, Washington, DC. Yout h E mploy m e nt P ro g ra m s I nve ntor y Qiang, C., and C. Rossotto. 2009. “Economic (database). World Bank, Washington, DC. Impacts of Broadband.” Information and http://www.youth-employment-inventory.org/. Regulatory Framework of the Labor Market: 8 Overcoming Segmentation Main findings • To remove segmentation in their labor markets, countries with the most restrictive regulations could consider reforming the regulation of wages and especially contracts. • A reform of the overly generous employment terms in the public sector to align risks and returns with the private sector would entail large fiscal gains and improve the allocation of talent to where it is most productive. • In the private sector, more lenient but enforceable contract regulation could be combined with unemployment insurance to protect workers’ incomes during periods of joblessness. • A more transparent and actuarially fair social insurance system could contain fiscal costs better than the current system, while increasing coverage. • Policies that encourage women to work go beyond immediate labor market regulation and need to target the underlying determinants of women’s access to economic opportunities. Segmentation in MENA’s labor minimum wage, wage agreements, and pub- markets lic sector wage scales. They will also involve decisions about the regulation of contracts: H ow can the countries of the Middle hiring and dismissal procedures and work- East and North Africa (MENA) ing hours, in both the private and the public overcome segmentation and achieve sectors. Finally, they will involve decisions a dynamic labor market, one that gives all about particular aspects of the social insur- groups an equal chance and allows the labor ance system. force to move to the highest-productivity These decisions need to rest on a broad sectors? While the solutions will have to social consensus involving all affected citi- be specific to each country in the region, zens. Chapter 5 suggests two fundamental given that conditions differ widely, they will preconditions for overcoming the labor mar- likely involve political decisions about the ket segmentation in MENA, which apply to 235 236 JOBS FOR SHARED PROSPERITY all countries in the region. First, policy mak- Strengthening basic labor market ers and citizens need to know what is hap- institutions pening in their labor market and how they A small group of countries—including the are affected individually. Access to reliable Arab Republic of Egypt and the countries and inclusive information therefore must be in the Gulf Cooperation Council (GCC)— ensured. This requirement is discussed in would benefit from expanding essential more detail in chapter 10. labor market institutions. In these countries, Second, it is important that citizens can the traditional social partners such as trade make their voices heard in an inclusive unions and employer representatives are not debate about political options. This means organized and diverse enough to hold a tri- not only strengthening the institutions that partite dialogue in which all stakeholders represent traditional social partners (trade have equal standing. Other social partners unions, employers, parliament, and the gov- are nonexistent, weak, or just emerging ernment) but especially establishing mecha- (Egypt). In line with this social landscape, nisms to represent the new social partners which favors a “laissez-faire” environment, (youth, women, the unemployed, and infor- minimum wages or collective wage agree- mal workers): their voices are essential for ments either do not exist or are not based on making socially optimal decisions on key a broad social consensus. elements of labor market regulation. Two essential institutions that merit Only an inclusive social dialogue can, strengthening in all countries of the region for example, set a minimum wage that are labor inspection and employment ser- protects workers without harming the job vices. Strengthening employment services entry chances of others. Only an inclusive is discussed in chapter 9. Evidence on labor social dialogue can negotiate collective inspection is currently available for Algeria wage agreements that balance inequality and Tunisia. Both countries have developed concerns with the entry chances of inexpe- these institutions with the official mandate rienced youth or design a social insurance of enforcing labor market regulations and system that covers the entire workforce. establishing a presence in different parts An inclusive social dialogue can go as far as of the territory. In Algeria, experts point involving social partners in the governance of out that labor inspectors officially inter- the social insurance system, as in Germany, vene only reactively, at the express request where a diverse group of employee and of a complainant (Musette and Mohamed- employer representatives self-administer Meziani 2011), similar to the model in the social insurance system. There, repre- northern European countries. In Tunisia, sentatives are elected by all those covered the mandate for labor inspection also cov- by social insurance. This form of gov- ers managing labor disputes and reconciling ernance can in principle enable smooth the workers’ and unions’ dialogue with the reforms, because all social partners have a employers.1 Tunisian labor market experts direct insight into potential strains on the rate poorly the performance of the national system. The Jordanian Social Insurance labor inspection institution, whose cover- System (see case study in chapter 10) and age is limited to large formal sector fi rms, the Tunisian pension funds also have clear with smaller and informal fi rms operating roles for both employers and employee under the radar. This situation contributes representatives in the governance of the to strengthening the informal sector and system, albeit with less direct insight and thus to segmenting the labor market. The access to information. insufficient quantity and quality of human Once these two important preconditions resources are frequently cited as the main for successful consensus building are met, the causes of low performance and, together recommendations will differ by country. The with the broad mandate and discretionary following sections explore some options. REGULATORY FRAME WORK OF THE LABOR MARKE T: OVERCOMING SEGMENTATION 237 powers of labor inspection institutions, Republic, and the Republic of Yemen—might have contributed to the reputation of these want to review their minimum wages care- institutions as “labor police” (Boughzala fully and adjust them downward to meet 2011). Addressing this negative perception market realities. A somewhat lower mini- and improving the capacity of labor inspec- mum wage, agreed to in an inclusive social tors would entail strengthening the labor dialogue and well enforced by all partners, inspection functions and removing exces- could be a substantial improvement over the sive mandates and discretionary powers status quo. Successful downward reform of from the legal codes. minimum wages is rare, and since it might come at high political costs, its pros and cons should be weighed carefully. The experience Overcoming segmentation in wages in the Netherlands offers some lessons in this The initial statistics presented in chapter regard (see box 8.1). 5 are consistent with weak enforcement of Among the countries in MENA whose minimum wages in several MENA coun- minimum wages merit revision, the Dutch tries, which can discourage hiring or divert example could be particularly relevant to it to the informal sector, thereby circum- Algeria, Jordan, and Morocco. The institu- venting the official minimum wage. 2 Over- tions in these countries are powerful enough all, a move to less restrictive (but enforced) to hold a tripartite social dialogue and could regulation would be better than regulation embark on a public debate along similar that invites circumvention and informal- lines. A public pact or accord between social ity and would prevent the majority of the partners for a reform of the minimum wage workforce from operating outside the regu- or its indexation would be credible. latory framework. In a small group of countries (Algeria, In view of this possibility, some countries— Tunisia, and, to a lesser extent, Jordan) including Algeria, the Islamic Republic of opening up the dialogue on collective Iran, Jordan, Morocco, the Syrian Arab wage agreements would likely result in a BOX 8.1 Wage reform in the Netherlands, 1980s In the Netherlands, social partners—trade unions In 1982, trade unions and employers signed the and employer representatives—normally negotiate historic Wassenaar Accord and committed themselves collective and minimum wages autonomously. How- to wage increases below infl ation and productivity ever, the state has traditionally set statutory wage growth. Indeed, real wages fell in the following years. policy and retained the right to intervene with wage The two main motivations for this were, fi rst, to get freezes in extreme cases. the Dutch economy back on a course of industrial During the late 1970s, the country suffered an profitability and, second, to retain autonomy as social unemployment crisis that needed a strong strategy partners and reduce government intervention. In for job growth. The government announced a freeze 1984, the indexation of public sector pay and social on public wages and on the minimum wage. Unions transfer payments was abandoned; wage formation suffered membership losses, organizing less than remained below price and productivity increases, 30 percent of Dutch workers in the early 1980s coincident with substantial employment increases. In (Ebbinghaus and Visser 1999), and ultimately vol- 1993, a new bipartite agreement pledged the continu- untarily agreed to a de facto wage freeze in exchange ation of a “responsible wage development.” for employers’ negotiating working time. Source: Visser and Hemerijck 1997. 238 JOBS FOR SHARED PROSPERITY downward adjustment for wage floors for Austrian workers’ capitalization fund bear graduates. Although these countries have features of this approach. relatively well-organized traditional social The other potentially high barrier to partners, the political dialogue does not yet fi ring—administrative charges—needs care- sufficiently include new partners. Conse- ful review country by country. The Tunisian quently, the collective wage agreements of labor code, for example, requires employers the past represent only formal “insiders” but that dismiss an employee for economic rea- are binding for youth and women. sons (such as productivity gains) to submit an application and receive the approval of a third party, the Commission des Licencie- Overcoming segmentation in labor ments (layoff commission). Changes to this contracts rule could involve replacing the discretionary Most countries in the region would benefit power of the commission with a published from greater flexibility in fi ring rules, com- list that spells out the criteria for the lay- bined with better security in incomes (the so- offs that require consultation with (but not called flexicurity policy). This process would approval of) the commission. All layoffs not involve the moderation of employment pro- listed would then be free from the obligation tection legislation and an increase in social for consultation. protection to buffer the cost of a transition Moving toward less stringent employ- between jobs. These changes would mean ment protection in hiring can also be a moderate but not excessive protection beneficial. Labor markets with relatively against dismissal and well-designed unem- liberal hiring policies, on the one hand ployment insurance and assistance. The (for example, policies that allow tempo- country-specific design and mix of these two rary and fi xed-term contracts), and protec- depend strongly on the political feasibility in tive fi ring policies that exclusively burden the country, and one nation’s recipe for suc- open-ended contracts, on the other hand, cess may fail in another. A few lessons from are segmented. New job market entrants international experience merit consideration, end up (and get stuck) in unprotected tem- however. porary contracts, while established insiders Strict employment protection usually benefit from open-ended contracts. Inspired manifests itself through either high sever- by the experience of southern Europe, nota- ance pay or high administrative burdens bly Italy and Spain, where open-ended con- on fi ring. For example, high severance pay tracts mandate high severance pay and such can make fi ring so costly that employers are clauses are not likely to be removed in the reluctant to make contract hires that include short run, recent empirical literature has severance pay and prefer to hire informally developed the model of a single open-ended or use more precarious temporary contracts. contract to overcome the dualism of fi xed- One way to ease the burden of severance pay term versus open-ended contracts. This type is to convert the one-off severance charge at of contract gradually accumulates the sever- termination into a severance savings account ance pay charge that the (old) open-ended or fund that accumulates over time and contract provided upon signing. Evidence receives contributions from both employers from Bentolila, Dolado, and Jimeno (2011) and employees. The contributions can be set and Costain, Jimeno, and Thomas (2010) at a small percentage of the current annual shows that this model could be beneficial if salary, rather than the last (and likely higher) fi ring costs as such cannot be changed. The annual salary at termination. The severance marginal increase in severance pay would account could then follow an employee from be minimal from a contract extension under job to job and be awarded upon retirement. this model, compared to a discrete switch Both the Colombian cesantía model and the from temporary to open-ended hiring. 3 REGULATORY FRAME WORK OF THE LABOR MARKE T: OVERCOMING SEGMENTATION 239 Overcoming segmentation in social need for a close (that is, actuarially fair) rela- insurance coverage tionship between the contribution of an indi- vidual and the expected benefits will further Perhaps the most critical task in address- undermine the financial sustainability of the ing the segmentation between labor market scheme, especially pensions. Redistribution insiders and outsiders is to develop effec- in the context of social insurance benefits can tive reform programs that rapidly increase be justified, but it should be transparent and access to social risk-management tools. explicitly directed at improving the social Indeed, both the coverage and the scope of position of targeted groups. In particular, social insurance need to be revised in most the use of (even implicit) subsidies should be MENA countries, where economic develop- restricted to helping individuals with low sav- ment has yet to be reflected in broad-scope ings capacity (informal workers in particular) social insurance programs. Most other enroll in low-threshold social insurance pro- middle-income, modern market economies grams that extend coverage beyond formal offer unemployment insurance (UI) and employment. This last step is especially hard, maternity benefits. The case for increasing as it will require a leap beyond the conven- MENA’s coverage from its current level of tional wisdom that individuals must have a 33 percent is based on the negative conse- formal job to participate in social insurance. quences of excessive exposure to uncovered social risks (old age, disability, death, and Promoting greater transparency work injury) and on the associated pro- and efficiency ductivity gains and improvements in social A review of the spending efficiency of the cohesion that would be achieved. social insurance system, including labor taxes and contributions, will help balance Designing incentive-compatible the economic efficiency and employment social insurance losses of contributions to social insur- Strategies for expanding social insurance in ance with the benefits of the protection MENA should start with eliminating gaps offered. Egypt, the Islamic Republic of Iran, in legal coverage and then adapting social Lebanon, and Tunisia have a “tax wedge”5 insurance rules (in particular for pensions) exceeding that of the Organisation for to create stronger incentives for individuals Economic Co-operation and Development and fi rms to participate and contribute over (OECD), and Morocco follows closely long career paths. The improved incentives behind. However, these countries do not should be designed with fi nancial sustain- achieve the high degree of social insurance ability in mind. Specific, possibly subsidized, coverage and services typically offered in coverage extension programs—aligned well OECD countries. An in-depth survey of the with the core design of a reformed social system, following the money spent, would insurance system— should complement help identify bottlenecks. Greater transpar- traditional interventions. Reforms should ency is needed in both collection and spend- also allow (voluntary) entry to those who ing, including parafiscal (that is, levies or fees either lack access to social insurance through other than taxes, but still going to the state; formal jobs or who cannot pay the full cost quasi-taxes) and “voluntary” contributions. of participation in social insurance.4 At the same time, social insurance contri- For example, the self-employed, the work- butions need to be aligned with the long-term ers in agricultural cooperatives, and the financial sustainability of social insurance employees of small firms are often unnec- systems to ensure that the desirable reduction essarily excluded from legal coverage in in high contribution rates corresponds with MENA. In the context of contributory sys- reforms on the benefit side. Achieving this tems, any expansion that does not address the balance is easier said than done. The same 240 JOBS FOR SHARED PROSPERITY political economy forces that contributed to acceleration of creative destruction and job the current equilibrium are likely to skew creation is really a refl ection of increased choices in favor of pursuing reform through flexibility in formal employment. increases in contribution rates, rather than U n e mploy m e nt i n s u r a n c e s av i n g s restricting early retirement, reducing income accounts (UISAs) may be a new institution replacement generosity, or inducing longer that is especially useful in reducing the bar- contributory paths through other measures. riers around formal employment in MENA. Unemployment insurance in a broader sense Designing unemployment insurance to may rely on two types of mechanisms: risk mitigate rigid employment practices pooling and individual savings of various Most MENA countries should consider kinds, both of which mean that the risks unemployment insurance as an alternative remain with the individual. Risk pooling is to legislating rigid employment protection its traditional core mechanism, whereby the (and severance pay). Effectively protecting contributions of the employed pay for the income during job transitions can sustain a benefits of those laid off. A limitation of risk more dynamic labor market with simultane- pooling is that it relies on the ability of insti- ously increased job creation and destruction tutions to unfailingly verify unemployment and more frequent transitions across the status. In the absence of such institutions, current insider-outsider divide. This report moral hazard (the risk that individuals will has shown how excessively rigid labor regu- claim unemployment benefits while infor- lations increase the segmentation between mally employed) could cause the collapse of insiders and outsiders. In a broader sense, the unemployment insurance system. Initially any regulation that increases the costs of dis- used in Latin America, unemployment insur- continuing unproductive employee-employer ance savings accounts address this challenge matches constrains the labor market’s effi- by combining mostly mandatory savings cient allocation (and reallocation) of produc- (“saving for a rainy day”) with a degree of tion workers and therefore also constrains risk pooling to make resources available dur- the expansion of economic activities and ing periods of job search. The UISA design employment. Paradoxically, the enemy is not allows an unemployment insurance mecha- the destruction of (unproductive) jobs per se nism to prevail even in a weak implementa- but rather that the social consequences of tion environment, as those who attempt to layoffs are borne solely by employees. abuse the system can “steal” only their own Unemployment insurance is a mechanism savings. Any unused savings should be paid by which fi rms and workers share the social out to the UISA “owner” upon retirement costs of increased labor market dynamism (Robalino, Vodopivec, and Bodor 2009). The and the associated frictional unemployment desirable incentive effect has been working (that is, the period between jobs when an in Chile (Reyes, van Ours, and Vodopivec individual is searching for employment, or 2010), and the experience of Jordan’s UISA transitioning from one job to another, often scheme (introduced in 2011 and soon to be in a context of imperfect or incomplete infor- available) could provide lessons for other mation) without restricting the forces of cre- MENA countries as well. ative destruction. In other words, it is a tool for simultaneously decreasing legislated Public versus private sector employment protections but increasing segmentation income protection for periods of frictional unemployment. A shift from legislated Gains from reforming public sector employment protections to unemployment employment insurance would then promote less dualistic There is evidence that the public sec- labor markets; in fact, the vision of accel- tor crowds out some private employment, erated net job creation through the parallel induces queuing for coveted public jobs, and REGULATORY FRAME WORK OF THE LABOR MARKE T: OVERCOMING SEGMENTATION 241 allows “double dipping.” Chapter 5 pre- employ ment agenc y recently began sented evidence of this phenomenon for five decentralizing budgeting and procure- diverse countries in the region, representing ment to local offices. the Maghreb (Algeria, Tunisia), Mashreq • For the career-based core of civil servants, (Egypt, Jordan), and the Gulf (United Arab revisit the incentives set by salaries and Emirates). employment terms. Essentially, this pro- Most countries in the region would ben- cess means maintaining merit in appoint- efit from reviewing the risks, returns, and ment and promotion. Current public extent of employment in the public sector. sector recruitment in MENA is modern Such a review would address the very heart and transparent on paper but in reality of employment arrangements in the public presents major limitations. Evidence from sector and prompt a debate on both the effec- Egypt and Morocco, for example, shows tiveness and the financial sustainability of that the majority of new magistrates public sector activity. entering the judiciary come from families where one parent already holds this posi- Practical steps toward change tion. Moreover, the disconnect between A recent review of the experience of public regulation and implementation needs to sector reform suggests the following practi- be overcome. In response to the meritoc- cal steps (see Manning et al. 2012): racy challenges in the region described in chapter 6, the public sector should take • Identify core and noncore public sector the lead in demonstrating transparent and activities and employment. Most coun- fair practices. To that end, the hierarchies tries conduct the former through civil ser- in command and control, which are far vants. This fi rst step is necessary for the steeper in the civil services throughout MENA region if it is to achieve a separa- MENA than in other parts of the world, tion of functions between the government should be flattened. Hierarchies can have (responsible for policy and legislation) a negative impact on civil service culture, and the private and nongovernmental as well as a detrimental effect on innova- organizations (NGO) sectors (as providers tion and creativity. of employment, health and environmental services, utilities, and the like). Given the significant size of the public sec- • Contract out noncore activities to pri- tor in most MENA countries, any reform vate and third-party providers. Noncore needs to recognize the importance of public activities rely on skills that do not need to workers as a political constituency. Public be nurtured within the public sector or on servants are citizens, they vote, and they are services that can be bought. This second unlikely to want to let go of privileges any step will help reduce crowding out of the more than other parts of society. In the Gulf private sector. The public sector can thus states, Jordan, and Lebanon, public servants be a legitimate source of revenue for the represent more than 30 percent of the labor private sector in areas where public fund- force—a very powerful and vocal political ing is justified. constituency. • Decentralize decision making to regional Inducing youth to aspire to jobs outside or local offices where possible. Decen- the public sector will require changes in other tralization helps reduce the workload sectors of the economy. The attractiveness of the central office and alleviates any of the public sector comes primarily from its bot tlenecks associated with central risk-return package but is also a function of approval and clearance. Empirically, it the unattractive terms offered by the infor- has been shown to foster innovation, as mal and small business sectors. The barriers problems detected locally can often be in the private sector are explored in detail in solved locally. For example, the Tunisian chapter 4. 242 JOBS FOR SHARED PROSPERITY Removing the segmentation limiting short-term incentive for fi rms that employ women’s participation in the labor force young women. Early results suggest that such incentives do work (see box 8.2). Policies that encourage women to work and increase their attractiveness to employers Encouraging women to participate go hand in hand. Labor force participation in lifelong learning reflects a combination of individual choices Active labor market policies (see chapter 9) and reactions to market conditions. Chapter 2 can address specific labor market problems discusses evidence that a share of the women and provide training, place individuals in who do not participate in the labor force actu- new jobs, and help the job seekers overcome ally would like to work. In this spirit, policies information problems by communicating that target the underlying determinants of their abilities to employers. Where these pol- female access to economic opportunities and icies focus on specific sectors or occupations, touch on more than the immediate labor mar- they can address the information problems ket regulations—such as time constraints, that arise from gendered networks. These access to inputs, and market and institutional measures could include providing informa- failures as priorities for intervention—can tion about wages and qualifications for a promote a better alignment between individ- given occupation and engaging experienced ual aspirations and actual opportunities for workers as mentors for younger workers of women. Since more than one factor may be the same (underrepresented) sex. In MENA, in play, a package of interventions is typically women are, however, severely underrepre- needed. sented as beneficiaries of active labor market policies, partly due to insufficient female tar- Reducing the mismatch between labor geting but maybe also to lack of flexibility market requirements and what women in program schedules. Many programs are learn in school conducted during working hours or require Although it is important to reduce the mis- full-time participation. It may be difficult match between labor market requirements for a young woman with children to partici- and what is learned in school for young men, pate in programs with such strict schedule this is particularly critical for women in requirements. Training programs should the region, because to a greater extent than attempt to provide flexible schedules. More- men, young women choose the humanities over, cultural restrictions sometimes require and social sciences for their higher edu- female teachers and classes for women only, cation. These disciplines are better suited although in practice the majority of pro- to the increasingly rare public sector jobs, grams are coeducational (Angel-Urdinola, but they are not oriented toward the skills Semlali, and Brodmann 2010). that the private sector is seeking. One of the most effi cient ways of acquiring work- Ensuring safety in travel to work and in relevant knowledge and professional skills the workplace for excelling in the job market is to apply Women in MENA countries report that the theories learned in class to real work set- lack of safe and reliable transportation con- ti ngs t h roug h on-t he -job t rai n i ng or stitutes a main constraint to their ability to internships. Equally important, policies work (World Bank 2011). This constraint is encouraging on-the-job training for women relatively inexpensive to address and could can also help change employers’ attitudes significantly increase female employment. toward female workers, challenging the Several countries around the world (Egypt, prevalent stereotype in MENA that female Japan, and Mexico, for example) provide workers are not as productive as male work- public transportation for women only so that ers. In Jordan, for instance, a pilot program women can travel safely. Moreover, the pri- provides employability skills training and a vate sector as well can undertake measures REGULATORY FRAME WORK OF THE LABOR MARKE T: OVERCOMING SEGMENTATION 243 BOX 8.2 Jordan NOW: The New Opportunities for Women pilot program Despite low employment rates, the majority of recent received 45 hours of instruction in team building, female graduates in Jordan want to work: 93 per- communications, presentations, business writing, cent say they plan to work, and 91 percent say they customer service, resume writing, interviewing, and would like to work outside the home after they are positive thinking. married. Yet only 17 percent of women aged 20–45 Just over half the trainees (62 percent) completed work in Jordan, compared with 77 percent of men. the courses. Unmarried women were more likely to This gap also holds among the more educated; with attend. Employment rates among graduates who community college graduates, it starts immediately received vouchers alone or vouchers plus training are on graduation and widens thereafter. The Jordan between 55 and 57 percent, compared with 17 and New Opportunities for Women (Jordan NOW) pilot 19 percent among those who received training alone program is rigorously evaluating the effectiveness of or received neither training nor vouchers. In all two policies to aid young female graduates in their groups, employment effects were higher for unmar- entry into the workforce with short-term wage sub- ried women. Financial empowerment (measured sidies and employability skills training. Short-term as the proportion of women who have their own wage subsidies give fi rms an incentive to risk hir- money and can decide how to use it) also increased ing young female graduates and an opportunity to significantly for all who received vouchers, training, overcome stereotypical perceptions of women as or both. Follow-up surveys will determine whether employees. The subsidies also give young women the employment effects of job vouchers are sustained more confidence in approaching prospective employ- and will explore the link between marriage and ers. The subsidy has a value equal to the minimum work, given the early fi ndings that married women wage for six months. Employability skills training are less likely to attend the training, less likely to use supplements the technical skills graduates learn in the vouchers, and less likely to be employed. community colleges. Students in the NOW pilot Source: World Bank 2011. to ease mobility constraints for women, such limit access to economic opportunities. Sev- as providing buses to take women from their eral property laws allow brothers to inherit homes directly to the workplace. In addition, land rather than sisters or require brothers to it is important to institute policies that pro- inherit double what a sister inherits. More- tect women at work and make the population over, laws that make the husband the head of aware of the legal consequences, in addition household can drastically reduce a woman’s to losing one’s job, of harassing females. ability to enter contracts, register property in her name, or join the workforce after she mar- Addressing legal bottlenecks ries, because in some countries the husband’s Labor regulation does not always take into permission might be needed for the woman account women’s key role in the household. to travel or work. Maghreb countries have In particular, legislative restrictions pertain- already made important progress in reform- ing to part-time work should be revisited, as ing these aspects of their legal systems, but they can discriminate against female work- much remains to be done in the region. ers who cannot consider full-time work Finally, maternity insurance needs to because household chores. Relaxing such be carefully crafted to minimize the risk of constraints would give women more oppor- disadvantaging women in the labor mar- tunities for paid employment. ket. Jordan, for example, recently abol- Laws that govern property rights and ished employer-paid maternity benefits and inheritance, especially land, are disadvanta- introduced maternity insurance paid by the geous for women in several countries and thus social security system to increase women’s 244 JOBS FOR SHARED PROSPERITY economic participation and protect their industry itself also creates jobs (mainly for rights and entitlements immediately after females) and allows parents, mainly young childbearing. Specifically, the law “will limit mothers, to be economically active. Child excuses and justifications presented to dis- care service can be provided directly by the miss the employment of women in cases of state (including local governments) or by the marriage, pregnancy or maternity leave. As private sector or NGOs, usually with pub- a matter of fact, maternity insurance shall lic subsidies and regulation. It could also be encourage employers to employ women and provided locally through so-called commu- as a result promote their role in the labor nity mothers who receive training and then market” (SSC 2011). Box 8.3 explores imple- care for children in their own homes. High- mentation challenges of this new law. quality early childhood development pro- grams have positive effects on the welfare of Providing affordable care for children future generations; they have been shown to Women spend more time on child care than prepare young children for success in school men. This greater demand on women’s time and eventually in life. Colombia’s Hogares constrains their economic opportunities. In Comunitarios program, which provides sub- many countries, improved child care oppor- sidies to designated homes operating as com- tunities for women have been the key to their munity child care centers, has significantly participation in the labor force. A developed increased mothers’ participation in the labor child care education industry could be eco- market as well as hours worked (World Bank nomically important because the child care 2011). Employers can also provide day care. BOX 8.3 Stakeholders’ perceptions of Jordan’s new social insurance law on opening up employment opportunities for women A forthcoming qualitative research study sought to recommendations with respect to the law and its capture perceptions of the ease or difficulty with implementation: which Jordan’s new social insurance law could be implemented. Approximately 141 women and men • Public awareness. Improve public awareness of in 24 focus groups, as well as 40 employers, were the law; clarify the provisions of the law and dis- interviewed in early 2012. Most men and women seminate any changes to employers, employees, who were aware of the law viewed the changes and the general public. positively and believed that, given the right circum- • Dialogue and consultation. Strengthen dialogue stances (such as the nature of the particular job and between the Social Security Cooperation, employ- changes in employers’ and family members’ mind- ers, and employees to clarify the rationale and sets), it could increase women’s employment. They provisions of the law; include employers, employ- thought that changes in the way maternity benefits ees, and civil society in consultations on any were paid were a positive step toward equality for review of the law and its implementation. women in the labor force. However, both men and • Implementation. Clarify the payment mechanism women participating in the focus groups suggested for maternity insurance during maternity leave that employers might not view the law favorably and facilitate processes for payment to reduce and expressed concern over employers’ compliance the burden on the recipient during this time (for with the law and the implications for hiring prac- example, direct bank deposits). tices. This perception was reflected in employers’ • Monitoring and safeguards. Implement a strong comments. Most were aware of the law, but they monitoring system and equitable and clear enforce- were largely divided on its usefulness to employ- ment of the law to facilitate implementation. ers, and many were uncertain how to implement its provisions. The participants made a number of Source: Brodmann et al. 2012. REGULATORY FRAME WORK OF THE LABOR MARKE T: OVERCOMING SEGMENTATION 245 In rural areas, employer-provided day care the primary household care provider is rein- is likely to work best in places where a large forced when paternity leave is not provided. number of employees are clustered. India is Although there is certainly nothing wrong experimenting with child care options for with the woman caring for the children, it rural women in the informal sector and in a is very important that children have oppor- public works program (box 8.4). tunities for substantial bonding with fathers as well, which paternity leave could allow. Redesigning parental leave policies to Starting off by providing both maternity and maintain incentives to hire women paternity leave and making paternity leave There are deeply rooted norms and beliefs mandatory in the medium term will likely about who does housework and provides be “transformative” by giving men incen- child care within the home. It is important to tives to take on more child care duties and start off with policies that work around these fostering more nuanced norms around care. norms rather than trying to change them. Moreover, paternity leave will likely decrease Parental leave policies can either reinforce employers’ reluctance to hire women, as men social norms around child care or try to work would also have the right to paid leave. A toward more nuanced social norms. Provid- factor that determines the effectiveness of ing maternity leave, for example, allows parental leave in opening additional labor a woman to take some time off to be with market options for women is the question of her children and also enhances her prospects who pays. When the costs are borne only by for participating in the economy. The expec- the employers, their incentive to hire women tation that a woman should at all times be is reduced. Public subsidies or equal amounts BOX 8.4 Expanding child care options in India In India, Mobile Creches (a nongovernmental orga- NGO, Mobile Creches also built a group of centers nization) is experimenting with models for provid- in remote rural areas in the states of Jharkhand and ing child care services for women employed in the Bihar following discussions with local women about rural informal sector and in public works programs. the child care constraints they faced in accessing Thirty-one day care facilities were set up around income-generating activities (specifically, yarn pro- New Delhi on public construction sites in partner- duction) in their communities. These centers follow ship with contractors, who provide child care facili- a community-based model, relying on employers as ties in addition to sharing other costs. These centers well as the broader community for their manage- responded to the expectation that women would ment, contribution of food materials, and training represent a large proportion of those taking up pub- and selection of teachers. lic works programs: this figure reached 80 percent at Similar efforts have been undertaken in the some sites. The day care programs include a nutri- Indian state of Gujarat by the Self-Employed tional component by providing meals during the day Women’s Association, a trade union for poor, self- and tracking the nutritional status of children over employed workers in the informal sector. It has set time; they have also integrated health services (for up day care centers for members’ children up to age example, immunizations and regular doctor visits). six. This service not only provides education and To provide access to child care for women work- nutrition for the children but also reduces the work- ing in the informal sector in resettlement colonies ing days women have to miss to take care of their in Delhi, Mobile Creches also helped create home- children—an especially important contribution, based centers and community-based programs by because few of them are salaried workers. identifying and training local women to provide these services. Together with Pradan, another Indian Source: World Bank 2011. 246 JOBS FOR SHARED PROSPERITY of maternal and paternal leave could help Through the use of the Internet and mobile address this issue. phones, markets can also be brought closer to women, who can more conveniently con- Increasing women’s capacity to start and nect with potential buyers and get informa- run their own businesses while increasing tion about prices. In India, a program run by their capacity to access credit an NGO, the Foundation of Occupational Female self-employment is compatible with Development, organized groups of women to prevailing social norms in the region, and market their products directly and increase policies that encourage and support female their profit margins by providing them with entrepreneurship could increase female par- access to cell phones and the Internet. Also in ticipation in the labor force. Training in entre- India, the Self-Employed Women’s Associa- preneurial skills, mentoring, business plan tion trade facilitation center created an online development, and business start-up assistance outlet so that women could sell textile and are all important measures, and such services handicraft products to improve quality and should increasingly target women. Several reduce delivery times (World Bank 2011). MENA countries are experimenting with “microcredit plus” models, combining credit with training in business skills and efforts to How much do labor regulations facilitate access to markets. Because women matter for job creation compared use credit less than men do, however, lenders to private sector regulations? have little or no information about women’s As chapter 5 showed, labor regulations potential repayment capacity and are unwill- matter for employment, especially when ing to extend them credit even if they are excessively restrictive. But private sector creditworthy. Women’s lower ownership of regulation also matters: a good investment assets in MENA also limits their ability to climate and regulations that foster compe- offer collateral. Microcredit schemes, in a tition in the product markets are good for variety of forms, are the most common way employment. of addressing the information deficit prob- According to a recent and growing empiri- lems for entrepreneurs. In group lending, cal literature, labor market institutions and individuals (who typically lack collateral) product market regulation interact: liber- band together to obtain a loan; although alization of the product markets has more the loans go to individuals, the liability for pronounced effects on employment in labor repayment falls on the entire group. Regular markets that have higher collective-bargaining group meetings allow lenders to overcome the coverage. However, the rigidity introduced information problem. In Egypt, for example, through collective bargaining also seems to post office branches are being used as out- retard upward trends. Real wages increase lets in some of the poorest 1,000 villages to less with higher competition in product mar- offer microenterprise loans to disadvantaged kets when the unions are powerful (Griffith, women who lack collateral. A randomized McCartney, and Harrison 2007). impact evaluation has been designed to assess The interaction between labor market and the effectiveness of this program. product market regulation is complex. Fiori et al. (2007) show that employment gains from Providing rural women with market access reducing barriers to entry in product markets through infrastructure upgrades and access are larger when labor market policies are to technology restrictive. In this sense, deregulation of prod- Upgrading and expanding rural roads can uct markets and labor markets can be “substi- increase the labor supply for women as well tutes.” That is, countries in which it is politi- as for men in rural areas, making it easier to cally difficult to deregulate the labor market access markets, expanding economic oppor- might achieve the same outcome by concen- tunities, and raising household income. trating on deregulating the product markets. REGULATORY FRAME WORK OF THE LABOR MARKE T: OVERCOMING SEGMENTATION 247 At the same time, liberalizing product regulation and target the underlying deter- markets can reinforce labor market deregula- minants of women’s access to economic tion but not necessarily vice versa. Fiori et al. opportunities, such as social norms, time (2007) find that product market deregulation constraints, access to inputs, and child care. has led, over time, to a decline in workers’ Since more than one factor may be at play, a bargaining power. In this sense, product and package of interventions is needed. labor market deregulation can be considered “complements.” Therefore, when assessing Notes the total employment effect of deregulat- ing product markets, one needs to consider 1. See http://www.cnudst.rnrt.tn/cgi-bin/wwwi32 both its direct effect and the induced effects .exe/[in=C%3a/sources/appli/jort.06/ affiche.in; http://www.cnudst.rnrt.tn/jortsrc/ on labor market policies and institutions. In 2010/2010f/jo0872010.pdf. particular, the recent literature indicates that 2. Breaking the minimum wage rule usually labor market deregulation does not affect the entails the denial of social insurance, because setting of product market regulation. This registration in the social insurance program fi nding suggests that deregulation and liber- could cause the employer’s violation of the alization policies are best phased in, starting minimum wage law to be discovered. with the private sector and product markets 3. The impact of single open-ended contracts on and then letting the labor market follow or equilibrium unemployment is debated. See move in parallel. Bentolila, Dolado, and Jimeno (2011) and Costain, Jimeno, and Thomas (2010). 4. For an in-depth discussion, see Gatti et al. (2012). Towards more dynamic 5. A tax wedge is the difference between the labor markets before-tax and the after-tax wages: that is, the income tax rate. In this report, this MENA countries would generally benefi t includes social contributions and labor taxes from moving toward less restrictive labor paid by the employer and is not the classical market regulation, which would be easier “tax wedge,” which also takes into account to enforce than the legal status quo. Overly taxes paid by the employee. restrictive regulations include, for example, high minimum or collective wage require- References ments in several countries, which people try to circumvent through informal work rela- Angel-Urdinola, D., A. Semlali, and S. Brodmann. 2010. “Non Public Provision of ALMPs in tionships. They also involve rigid dismissal AMCs: An Inventory of Youth Programs.” rules that reinforce labor market segmenta- Social Protection Discussion Paper 1005, tion among newcomers such as youth and World Bank, Washington, DC. women, who are less likely to obtain jobs Bentolila, S., J. Dolado, and J. J. Jimeno. 2011. with formal, protected contracts. Finally, “Reforming an I nsider- Outsider Labor a reform of employment terms in the pub- Market: The Spanish Experience.” CEPR lic sector is necessary and would result in Discussion Paper 8691, Centre for Economic large gains. Policy Research, London. More lenient dismissal rules and wage Boughzala, M. 2011. “Labor Market Regulations policies could be combined with unemploy- and Institutions: The Case of Tunisia.” Tunisia ment insurance and incentive-compatible background paper, World Bank, Washington, DC. social insurance to protect workers’ incomes Brodmann, S., I. Jillson, N. Hassan, J. Jude and also cover periods of joblessness. A more Sajdi, and A. Zidan. 2012. “Social Insurance transparent and efficient social insurance sys- Reform in Jordan: Awareness and Perceptions tem could contain fiscal costs better than the on Employment Opportunities for Women.” current system, even with higher coverage. Mimeo, World Bank, Washington, DC. Policies to encourage women to work Costain, J., J. F. Jimeno, and C. Thomas. 2010. should go beyond immediate labor market “Employment Fluctuations in a Dual Labor 248 JOBS FOR SHARED PROSPERITY Market.” Banco de España Working Papers Resources/3035863-1285601351606/PSM- 1013, Banco de España, Madrid. Approach.pdf. Ebbinghaus, B., and J. Visser. 1999. “When Musette, M., and M. Mohamed-Meziani. 2011. Institutions Matter: Union Growth and “Beyond the Insider- Outsider Structure Decline in Western Europe, 1950 –1995.” of the Labor Market.” Background paper European Sociologic al Re vie w 15 (2): commissioned for this report. 135–58. Reyes, H., J. C. van Ours, and M. Vodopivec. Fiori, G., G. Nicoletti, S. Scarpetta, and 2010. “Incentive Effects of Unemployment F. Schiantarelli. 2007. “Employment Outcomes Insurance Savings Accounts: Evidence from and the Interaction between Product and Chile.” IZA Discussion Paper 4681, Institute Labor Market Deregulation: A re T hey for the Study of Labor, Bonn, Germany. Substitutes or Complements?” IZA Discussion Robalino, D. A., M. Vodopivec, and A. Bodor. Paper 2770, Institute for the Study of Labor, 2009. “Savings for Unemployment in Good or Bonn, Germany. Bad Times: Options for Developing Countries.” Gatti, R., D. Angel-Urdinola, J. Silva, and A. IZA Discussion Paper 4516, Institute for the Bodor. 2012. Striving for Better Jobs: The Study of Labor, Bonn, Germany. Challenge of Informality in the Middle East SSC (Social Security Cooperation). 2011. News, and North Africa Region. Washington, DC: July 17. http://www.ssc.gov.jo/english/pages. World Bank. php?menu_id=&local_type=1&local_id= Griffith, R., G. McCartney, and R. Harrison. 24 0 & lo c a l _ de t a i l s = 3& lo c a l _ de t a i l s1= 2007. “Product Market Reforms, Labour &localsite_branchname=SSC. Market Institutions and Unemployment.” Visser, J., and A. Hemerijck. 1997. A Dutch Economic Journal 117 (March): C142–66. Miracle: Job Growth, Welfare Reforms and Manning, N., J. Blum, V. Srivastava, and L. Corporatism in the Netherlands. Amsterdam: Van Gelder. 2012. “Better Results from Amsterdam University Press. Public Sector Institutions.” World Bank, World Bank. 2011. World Development Report Wa sh i ng ton , D C . ht t p: //sit ere s ou rc e s 2012: Gender Equality and Development. .worldba n k.org / E X TG OVA N T IC OR R / Washington, DC: World Bank. Educational and Training Systems: Realigning Incentives 9 for Skills That Matter and Making Employability Count Main findings • The reform agenda for educational and training systems needs to build on timely information on education and labor market outcomes. Such information is essential not only to formulate policy but also to guide young people’s decision making and realistically shape their expectations. • To close the skills gaps equitably and efficiently, educational and training systems need to replace the current logic of selecting students with a logic of fostering learning. • A renewed public-private partnership in education and skills development can also provide incentives and information for employers to play a prominent role in education and training. • Transparent intermediation and skills certification mechanisms are needed to effectively match talent to opportuni- ties in the private sector. • Active labor market programs need to be refocused to provide second-chance options to the most vulnerable who face challenges in successfully integrating into the labor market. T he Road Not Traveled education and demand-driven programs and curricula. flagship report (World Bank 2008) Although the Arab Spring ushered in a presented a policy road map for educa- political climate favorable to progress in this tional and training systems in the Middle East long-term program of reform, it is obvious and North Africa (MENA): good governance that these reforms will not come easily or and incentives for accountability were the quickly. They will require strong and consis- trigger for the reforms needed in the region. tent leadership, the engagement of all stake- Results—i.e., relevant learning outcomes— holders, a focus on governance, sustained and rather than inputs should drive policy mak- predictable financing over the long term, and ing and lead to effective public accountability robust monitoring and evaluation (M&E). 249 250 JOBS FOR SHARED PROSPERITY This flagship report focuses on a specific • Closing the information and knowl- dimension of that comprehensive reform edge gaps. This requires monitoring and program advanced back in 2008: building evaluating the quality of education on employability and making it count in the the one hand and reforming assessment labor market. It examines specific issues and certification systems on the other. related to the transition from education Addressing knowledge and informa- to work. The central message is that skills tion failures, gaps, and asymmetries will gaps and mismatches reflect the inability of allow employers to better communicate educational and training systems to trans- what they require from the educational form heavy investments by young people and training system, and will enable and their families into tangible labor mar- educators to improve quality and to bet- ket outcomes. In addition, unmeritocratic ter assess and certify learning. Closing job-matching mechanisms exacerbate the these gaps will also lead families and stu- difficulties that young people encounter in dents to make more informed decisions acquiring the skills to be employable. Such and build realistic expectations for the mechanisms prevent young people from transition from education to work, and it cashing in their employability capital and will permit all stakeholders to have more successfully mastering the transition from systematic, accurate information about education to work. what education and training are accom- Instituting reforms that address these chal- plishing. This increased transparency can lenges and yield more responsive and innova- lead to more solid accountability across tive educational and training systems is urgent. all the institutions and actors involved. Realistically, however, such reforms, no matter • Replacing the “logic of selection” with how successful they become, make a necessary a “logic of learning.” The ultimate chal- but insufficient contribution to solving prob- lenge for education in the 21st century is lems related to the transition from education paradoxical: to design a race that every- to work in MENA. The lack of relevant skills body can win and in which there are among first-time job seekers and the absence multiple ways to win. This challenge will of a level playing field for accessing jobs cre- be met only when governments decide ate a low-level equilibrium trap, in which nei- to make maximizing opportunities for ther the private sector nor the educational and all citizens, particularly for youth, a training systems appear to have the capacity, top priority. Hence, educational and information, or incentives to do things differ- training systems—particularly at the ently. For this reason, it is essential to foster secondary level, including technical competition in the private sector and realign and vocational education and training incentives for hiring in the public sector. Fur- (TVET)—should become more inclusive thermore, financing mechanisms in the educa- and more focused on learning and less on tional and training systems will also need to selecting and exclusively rewarding the change in response to increased dynamism in academically able while leaving the rest the private sector resulting from reforms. behind (OECD 2012; MEC-IFIIE 2012). Policy priorities should thus shift toward preventing students from dropping out and Overcoming skills gaps, leaving school early and find alternative mismatches, and unmeritocratic training itineraries for everyone to acquire hiring practices relevant skills and competencies for a suc- From the specific perspective of educational cessful transition from education to work. and training systems, long-term policies for • Making employability count through a meeting the challenges in the transition from renewed partnership between the private education to work need to focus on three sector and the educational and training sys- fundamental goals: tems. A new, more effective public-private EDUCATIONAL AND TRAINING SYSTEMS 251 partnership (PPP) framework in education In addition, the region needs to find policy and skills development in MENA would solutions that provide young school leavers yield the right incentives and necessary with second-chance options in case the tran- information for employers to play a promi- sition from education to work fails and to nent role in education and training. A com- help those who are particularly vulnerable, mon language and incentives are needed such as low-skilled youth or women, integrate to correct the signaling failures from both into the labor market. Such second-chance sides. Public and private intermediation also options outside the educational and training has a role in providing employers access to a system include tightly targeted active labor larger pool of candidates and thus improv- market programs (ALMPs). Figure 9.1 sum- ing the efficiency of the job-matching pro- marizes these points. cess. Efficient private and public interme- This chapter addresses both long- and diation systems could also mitigate young medium-term policies as well as short-term people’s concerns that they will not succeed measures with high potential for innova- in the second transition from employabil- tion and relatively low implementation ity to employment because of factors they difficulty that are generally applicable regard as beyond their control. throughout MENA. A suggestive country- specific list of priority policies and measures Timely and publicly available data that based on inputs from education policy mak- inform policy making and guide stakeholders’ ers in MENA is presented in the annex to this decisions are important because they simul- chapter and reprised in chapter 10. The chap- taneously cut across all reform efforts and ter will make periodic reference to some of provide the “glue” that makes them cohere. these policy options and measures to identify These long-term policy areas respond to the which are already high on the priority list of employability and employment barriers iden- MENA countries. tified in chapter 6. Education and skills have low quality and relevance for three reasons: (1) the private sector and the educational and Closing the information and training systems operate in isolation; (2) the knowledge gaps public sector continues to be the main cli- Closing the information and knowledge gaps ent of the educational system; and (3) a logic related to skills supply and demand and to of selection prevails over a logic of learning learning outcomes will help set the right incen- in educational systems. In addition, hiring tives for producing relevant skills and enable practices are driven by privilege rather than both educational institutions and employ- competition. Learning rather than selection ers to communicate more effectively. Closing should be the objective of an inclusive edu- the data gap on skills is therefore crucial to cational and training system that provides closing the skills gap itself. Data are needed opportunities to all students and equips them to inform the dialogue and decision making with the high-quality, relevant skills for suc- of stakeholders on students’ choices in upper- cess in a dynamic labor market. Making secondary and tertiary education and, more employability count depends on the ability generally, on expectations—from employers of the private sector to signal which skills are but also from students and their families—of needed and contribute to providing them. Effi- the educational and training systems (Sonder- cient and transparent intermediation between gaard et al. 2011). labor supply and demand—through public or private employment services, for example— Improving quality through could mitigate young people’s concerns that monitoring and evaluation they will not succeed in the second transition from employability to employment because of A first step is to establish M&E systems factors they regard as beyond their control. (including national systems to assess student 252 JOBS FOR SHARED PROSPERITY FIGURE 9.1 Linking the barriers to the transition from education to work to long-term policies Barriers Results Conditions for change Selection of policy options • Increase transparency, Close the e.g., through program A logic of information and evaluation, data selection knowledge gaps dissemination, career prevails over the logic of learning Low quality and counseling, and relevance of skills Replace the logic employment services of selection with a • Provide opportunities for logic of learning every student; reform Private sector and educational and school-leaving Make employability examination system; focus training systems count through a operate in isolation on soft-skills;  paradigm renewed partnership shift for TVET with the private • Institutionalize sector The public sector is coordination/cooperation the main “client” Little importance with the private sector and Provide second- of the education of merit in access involve them in curricula chance options system to jobs design/financing/training outside the educational system • Provide ALMPs outside the educational system to Realign incentives facilitate labor market Meritocracy for public sector integration and support deficit hiring vulnerable youth and women performance, tracer studies, and more student testing. MENA students have to sit sophisticated classroom-based evaluation for numerous public examinations, and, instruments) and to use the resulting data to as explained in more detail in chapter 6, make policies and improve institutions. In the entire educational and training system addition to strengthening the measurement in some countries seems to revolve around of learning outcomes, a parallel effort is those examinations. Student achievement is needed to build the capacity to use the data therefore thoroughly examined in MENA to improve results in schools (box 9.1). The but rarely for the purpose of improving the availability of data and a more transpar- quality of education: rather, it is used only ent policy for its dissemination will lead to for academic selection and student placement more solid accountability mechanisms, more in upper-secondary and tertiary education. informed and participative stakeholders, The lack of evidence-based policy formula- better grounds for new partnerships among tion and implementation is one of the factors stakeholders, more realistic expectations that explain why many efforts to improve the and aspirations among students, and a bet- quality of education have not been very suc- ter sense that reform policies and measures cessful, why education is not at the forefront are worthwhile and have high potential to of public agendas, and why so little public achieve their goals.1 debate has taken place on how to improve National assessments that provide the quality and relevance of education. evidence for developing policies and improv- This limitation, together with the lack of a ing schools differ from public exams. Usually critical mass of human resources to conduct administered to samples of students, these education research, is hindering the region’s assessments are designed to establish base- capacity to transform its low-performance lines and measure learning progress. Most educational systems into high-performance educational systems in MENA lack sufficient systems that offer more dynamic support evidence on student performance, although for integrating Arab countries into the 21st- ironically the region has never lacked for century global and knowledge economy. EDUCATIONAL AND TRAINING SYSTEMS 253 BOX 9.1 The Arab Regional Agenda on Improving Education Quality and the Regional Network for Education Research Initiative The Arab Regional Agenda on Improving Education The Regional Network for Education Research Quality seeks to enhance learning outcomes for all Initiative aims to build national and regional capac- in the Arab world by improving the quality and rele- ity in MENA for using empirical evidence in the vance of the education services available. To achieve formulation, implementation, monitoring, and this goal, the organization will tie existing and new evaluation of education policies aimed at educa- partners and institutions (public and private) into a tional improvement. In support of that goal, the coherent, more effective regional network. Directed World Bank endowed the Jordan National Center by the Arab League Educational, Cultural, and for Human Resource Development with a three- Scientific Organization, located in Tunis, the orga- year Institutional Development Fund. Additional nization maintains five regional programs housed funds were provided through a partnership with at five centers across the region, each targeting a the Islamic Education, Scientific and Cultural crucial area for improving the quality of education: Organization. (1) evaluation of learning achievement; (2) profes- Through a series of workshops, seminars, online sional development for teachers; (3) early childhood discussions, and mentoring visits, the regional network development; (4) curriculum innovation and infor- has enhanced the capacity of educational research mation and communication technology in educa- institutions, researchers, and policy makers from seven tion; and (5) entrepreneurship education. MENA economies (Dubai, Jordan, Lebanon, Oman, T he organization will build capacit y and the Syrian Arab Republic, Tunisia, and the West strengthen institutions both regionally and in Bank and Gaza) to analyze, present, and discuss data individual countries. Its regional strategy is to and evidence related to student learning and perfor- create a strong knowledge base and foster collabora- mance and to understand the factors that determine tion through a network of experts, programs, and and explain differences in learning achievements. It specialized institutions. In countries, it will build has also strengthened the capacity of participating the capacity to use assessment and evaluation to countries to formulate evidence-based policies and improve learning outcomes, focusing on school and programs aimed at tackling student learning and classroom factors. performance issues. The regional network has also International collaboration is the key strategic reached other goals: principle of this regional agenda, under the assump- tion that collaboration is the most effective way • It was instrumental in bridging the gap between to get results within a country. Building regional education researchers and policy makers. During capacity will make it easier to develop the required several workshops, local researchers presented institutions and critical mass of experts and special- policy notes on selected education issues to pol- ists, which would take far longer and require greater icy makers; the researchers also validated their investment to develop inside each country indi- approach to policy analysis and the immediate vidually. Regional learning will accelerate because impact of their recommendations. countries will have multiple institutional venues for • A virtual network of researchers from seven exchanging experiences, accessing knowledge, and countries was established to facilitate and obtaining technical assistance. encourage the exchange of ideas, technical assis- The organization was officially launched in tance, and reports and related documents among Tunis in January 2012. A number of regional and participants. Establishing the virtual network international organizations have already joined this proved challenging owing to the participants’ innovative network: the World Economic Forum, lack of experience with this type of collabora- UNESCO–Beirut, the Islamic Development Bank, tion, but the identification of a “champion” team Microsoft, Intel, Injaz Al-Arab, Queen Rania (West Bank and Gaza) helped overcome this Teacher Academy, and the Arab Resource Network. obstacle. 254 JOBS FOR SHARED PROSPERITY The contemporary drive to set up national school performance. These data will then student assessment systems — and for be used by teachers to improve their per- increased participation in international test- formance, help decision makers in mak- ing programs such as the Programme for ing well-informed choices, and inform International Student Assessment (PISA) or the community on the outcomes of the the Trends in International Mathematics and resources allocated to schools. Science Study (TIMSS)—reflects the global • Building capacity of human resources consensus that, in addition to the assessment in the field of quality and its monitor- of learning, countries must invest in assess- ing and evaluation, and providing them ment for learning, because such investment is with the necessary competencies to design the foundation for meaningful and sustain- and develop assessment mechanisms and able improvement in the quality of education tools as well as advanced methods of data (Greeney and Kellaghan 2008, 2009). It is analysis. not enough to have the capacity to measure • Disseminating a culture of quality, evalu- learning outcomes; what really makes the ation, and accountability in education, difference is whether countries also have the where the data on the functioning and out- capacity to use information about learning comes of the educational system are made outcomes to improve quality and enhance publicly available to all stakeholders. equity, the capacity to communicate the • Establishing a joint Arab evaluation sys- information to the wider public, the capacity tem and regional research program on the to learn from the results, and the capacity to quality of education, developing bench- make policy choices based on the results.2 marks and regional databases through Although more and more countries in the the collaboration between the A rab region participate in international testing Observatory for Education and the min- programs—particularly TIMSS—and some istries of education and other specialized have set up their own national assessment institutions. systems, policy makers and practitioners Implementing these agreements could be still lack the capacity for analyzing the data. transformational for education quality in Until very recently, countries have gener- the MENA region. ally been reluctant to disseminate student assessment data systematically; however, momentum is also building toward greater Reforming assessment and regional collaboration in a field where all certification systems MENA countries face the same issues and A high-quality educational system requires challenges (as shown in box 9.1). a high-quality assessment and certification The turning point may have arrived in system. Such a system clearly sets the rules September 2010, when 18 MENA countries, of the game for all stakeholders—including represented by their ministers of education, employers and public and private educa- met in Doha and signed the Doha Declara- tion and training providers—and has the tion on Improving Quality of Education in capacity to produce accurate information the Arab World. 3 The declaration contains on the skills and competencies of graduates the following agreements, among others: to reduce asymmetric information in labor • Establishing quality national standards markets and improve mobility. for all dimensions of the educational sys- Over the past several decades, many coun- tem, particularly for teachers and other tries have worked on developing National education professionals. Qualifications Frameworks (NQFs) as a way • Establishing national monitoring and to ensure the high-quality recognition of evaluation systems based on objective learning at all levels (ILO 2010).4 NQFs are data and quantitative and qualitative indi- “frameworks in which all the qualifications cators to measure student learning and offered in a particular country (or region) are EDUCATIONAL AND TRAINING SYSTEMS 255 organized according to level …. NQFs show within the region reinforce the lack of politi- how different qualifications relate to each cal leadership, key stakeholders in relevant other [and] are thus designed to enhance sectors are not involved, and technical lead- horizontal and vertical mobility within a ership and expertise are lacking within country’s or region’s education and train- institutions. ing system” (King and Palmer 2010). NQFs Moreover, despite the promising potential have the potential to bridge the gap between and the worldwide trend favoring national learning and certification—that is, between frameworks and related transnational skills development as such and the certi- approaches to qualifications, regional frame- fied value of the acquired skills in the mar- works appear to take a very long time to set ket. They introduce more transparency and up (at least a decade in most cases), are dif- therefore more complete and nuanced infor- ficult to implement, and (more important) mation to learners, teachers, and employers are challenging to amend or refi ne (Young about what graduates know and are able to 2005). King and Palmer (2010) suggest that do. Because NQFs focus on defining learning there are “other ways to ensure quality and outcomes, they can become a critical driver skill recognition than an NQF.” To be sure, of competency-based curriculum reform. instead of the top-down approach of national This clarity also enables employers to under- qualifications, in some countries a bottom- stand specific qualifications more easily and up approach— one that is sector based helps improve communication and under- instead of system-wide—could work better. standing between the private sector and the In fact, both could be undertaken in parallel, educational and training systems. because setting up a national framework is Referencing national work to a regional presumably, by defi nition, a long-term pro- qualifications framework is one way to assess cess. Traditional approaches such as collabo- the qualifications acquired in a given country ration between the government and employer in the context of the skills-specific labor associations in a particular sector in setting demand in several other countries. A regional occupational standards, defining required framework also promotes awareness of the competencies, and identifying assessment quality and relevance of postbasic education. and certification mechanisms could also Internationally harmonized national frame- work well in MENA. This approach has works can be important instruments for the added value of bringing together, for a bridging the asymmetry of information much-needed institutionalized dialogue, sec- across national borders and fostering interna- tor employer associations, chambers of com- tional labor mobility. A good example is the merce, trade unions, public authorities, and European qualifications framework, which education and training providers (CEDEFOP some MENA countries in the Mediterranean 2009; Psilos and Gereffi 2011). are considering as a reference for their national systems (Leney 2009). A more integrated Short-term measures for MENA regional approach to qualifications may speed up and add value to national processes while This agenda described so far definitely heightening the transparency of qualifications focuses on long-term policies, involving a systems throughout the region. focus on results rather than on inputs, a National and regional frameworks clearly drive for transparency, and a regional and add value by enhancing the employability of international dimension for benchmarking graduates (especially in key sectors), increas- and collaboration. In this context, how- ing the mobility and portability of qualifica- ever, it seems useful to identify short-term tions, and improving the formal and legal policies and measures with high innovation visibility and transparency of credentials. Yet potential and relatively low implementa- these initiatives in MENA face important tion difficulty that MENA countries could political constraints. Conflicting agendas adopt. 5 256 JOBS FOR SHARED PROSPERITY Design, develop, and strengthen national autonomy of governmental bodies or agen- systems of student assessment cies in charge of nationwide evaluation of Some countries in the region, notably Jordan learning outcomes. Some MENA countries and Tunisia, have made very important prog- implemented similar reforms long ago— ress in developing national systems of student among others, Jordan, through its National assessment over the past few years. As dis- Council of Human Resource Development, cussed in box 9.1, additional initiatives such and Lebanon, through its Education Center as the Arab Regional Agenda on Improving for Research and Development. Education Quality and the Regional Net- work for Education Research Initiative are Invest in school and career counseling and highlighting the potential of international guidance collaboration as an incentive for more coun- Promoting career counseling and guidance tries to set up their assessment systems, functions within schools and universities make relevant use of the data collected, and should facilitate placing students in intern- take steps to disseminate the fi ndings more ships and jobs. International evidence shows transparently. that career counseling is particularly cost effective in helping students identify the best Refine the quality assurance function in the match for their talents in the current labor educational system market.6 Continuing to professionalize the Institutional changes that strengthen the career counseling function could be coupled separation between policy-setting func- with strengthening the counseling role of tions and evaluation and oversight func- teachers and trainers as part of their profes- tions are needed to ensure the institutional sional development (box 9.2). BOX 9.2 Career counseling and guidance: The example of Sweden Sweden has invested in career and educational Quality counseling requires trained counsel- guidance throughout its entire school system. This ors. In Sweden, counselors are prepared through investment has succeeded partly because of the a three-year university program leading to a BA development of high-quality education for counsel- degree. Participants in this program learn to sup- ors. The national curriculum states that counseling port individuals in exploring, articulating, and is compulsory and that “student guidance offi cers developing their interests and capabilities as well and vocational guidance staff should inform and as their knowledge about themselves in relation guide pupils prior to the next stage of their educa- to society, education, and working life. The edu- tion and vocational orientation.” The activities and cational requirements for counselors who serve in roles of the counselors differ, depending on the type vocational education programs vary; sometimes a of school, but all activities have the common goal of teacher provides counseling, and at other times a helping an individual obtain more knowledge about degree in counseling is required. his or her own capacity, increase self-confidence, In addition to counseling, career fairs are arranged and form personal goals within education and regularly in cooperation between local employers and working life. Information about current opportuni- upper-secondary schools. The fairs provide general ties and promising fields within the labor market is information about various educational tracks, pro- shared. Pupils receive counseling both individually fessions, and labor market opportunities and usu- and in group sessions in primary, secondary, and ally offer personal career guidance. Fairs are often upper-secondary schools. Counselors usually work an opportunity for students to connect directly within the schools, although some municipalities with employers and arrange for summer jobs or offer career guidance through an external guidance internships. center instead of within each single school. Source: International Program Office for Education and Training 2010. EDUCATIONAL AND TRAINING SYSTEMS 257 Improve labor market information and competencies that both go beyond and cut graduate tracking across the traditional divide between general The collection of data on numbers, skills, and vocational education (World Bank 2005; and wage evolution of graduates of TVET Griffin, McGaw, and Care 2012). In addition, institutions, universities, and certifi cation deferring student selection and specializa- programs should be improved to facilitate tion is becoming more common as second- career decision making and enhance efforts ary education systems expand and become to promote investment. This measure will more inclusive. The global trend over the past provide job market and career guidance for 50 years or so has been toward upgrading spe- young people, criteria for program creation cialized vocational programs to postsecond- or withdrawal for TVET providers, and ary and tertiary levels at the same time that accurate information for employers on the vocational and specialized training elements labor market results of different institutions are being incorporated into the secondary and programs in specific sectors. education curriculum to keep them relevant and to retain students (World Bank 2005). Replacing the logic of selection with a logic of learning Reforming the secondary-school leaving examination Increasing the avenues of success available in educational and training systems is a criti- Responding simultaneously and effectively cal goal in transforming a selection-oriented to growing student diversity and rapidly system into a learning-oriented one. Multi- changing labor market demands requires plying the number of viable options for stu- fi nding the right mix of cognitive, soft, and dents is in itself a powerful, inclusive policy, job-related skills. According to the evidence, with high potential for innovation and qual- routine skills are increasingly less valued as ity improvement in education. countries get richer (Autor, Levy, and Mur- nane 2003). The key policy question for the school curriculum then becomes, which cur- Providing opportunities for riculum prepares students best for an uncer- developing meaningful skills tain, rapidly changing future? The answer The first long-term policy option is to necessarily implies balancing the short-term design alternative training itineraries for demands of the labor market against the every student, so that each acquires relevant long-term employability potential of gradu- skills for a successful transition from edu- ates. The global trend is toward outcome- cation to work. Implementing this policy based and competency-based curriculum in secondary education and especially in reforms in postbasic education (CIDREE/ postbasic or upper-secondary education DVO 2008; CEDEFOP 2009): these cur- requires strengthening and rebalancing the ricula have the potential to create a more links between general and vocational tracks challenging, demanding, motivating, and and introducing new, blended training pro- inclusive learning environment for students. grams. It also requires greater emphasis on One of their key goals is to give students developing core knowledge and cognitive more choice and more autonomy as learners. skills, in part through reclassifying some They also imply clearer student expectations TVET schools as general education second- about the daily routines and activities in the ary schools and increasing the share of the school and about the teaching and learning common secondary curriculum required process. in TVET schools while redefi ning the very Outcome-based and competency-based nature of that curriculum.7 curriculum reforms have very important orga- T he dem a nds of t he 21st- cent u r y nizational implications for schools, requir- labor markets give priority to skills and ing sweeping transformations in the ways in 258 JOBS FOR SHARED PROSPERITY which schools are organized, all the way from in education currently relies to a great extent the timetable, the use of space, and the teach- on these Tawjihi-type examinations and that ing and learning materials to external part- the public regards them as a central institu- nerships. Finally, these curriculum reforms tion in society and widely believes them to be lead to a much more sophisticated assess- a fair, meritocratic instrument for allocating ment regime, one which, for a start, is clearly further educational opportunities. Students aligned with the competency approach. and other stakeholders participating in focus Such a regime would obviously enhance and groups in Jordan highlighted these points promote the relevance of the curriculum, quite effectively and convincingly. Most men- although it is also true that assessment will tioned that the Tawjihi exam plays a large become more complex, time consuming, and role in determining one’s future, and, given potentially more bureaucratic. Schools and that admissions policies across all universities education policy makers alike need to be and colleges in Jordan are directly based on aware of those risks and ensure that they take Tawjihi grades, students’ choices about their concrete actions to address them. postsecondary education are very limited.9 It Reform of the secondary-school leaving is not surprising then that initiatives to reform examination system is perhaps the most criti- examinations are raised continuously in pub- cally needed educational change in MENA. lic debate but resisted systematically by some Transforming that system from an exclusively stakeholders (the Arab Republic of Egypt is selective mechanism into one that promotes the most salient example). Consensus appears inclusion and learning could trigger dramatic extremely difficult to build, although the changes in postbasic education in the region. changes brought about by the Arab Spring For one thing, the competency-based curri- could provide a favorable political climate for cula now being developed and implemented complex reforms in the examination system.10 to some extent in most countries in the region provide a strong case for national public Investing more in disadvantaged examinations to reflect and prioritize the students and schools evaluation of competencies. The introduction of more aptitude-oriented and higher-order To level the playing field for all individuals cognitive elements into the examination and to ensure that the proportion of young instead of—or at least in addition to—rote people who are unskilled and excluded memorization questions would send a strong from further education is as low as possible, message to schools, teachers, and students greater investment is needed in disadvantaged that the priorities for learning outcomes are students and schools. Chapter 2 presents changing.8 In this way, examinations could strong evidence that a large share of young realize their considerable potential as instru- people in MENA are excluded from educa- ments for closing the huge gap in percep- tional and training systems at some point tions of education quality: many employers during their school lives and, because of lack and other stakeholders regard quality as the of skills, they cannot access quality employ- inculcation of 21st-century skills, achievable ment. More inclusive policies are needed in by all, whereas some influential stakeholders education and training to prevent the exclu- still regard it as the development of an elite of sion of those vulnerable and disadvantaged academically able pupils who can score 99 in because of early school leaving or school a Tawjihi-like examination (that is, a general failure and to identify conditions in which secondary examination also used to regulate second-chance programs effectively provide admission to higher education). routes back into education—and therefore Any attempt to reform public examina- employment—for these young people. Such tions in MENA needs to acknowledge from policies are needed throughout the region. the outset, however, that public confidence Some education policy makers in Morocco EDUCATIONAL AND TRAINING SYSTEMS 259 and the Republic of Yemen consider them a out, introducing soft or 21st century skills, top priority (table 9A.1). or introducing innovative student grouping A report from the Organisation for Eco- arrangements—always risk triggering the nomic Co-operation and Development resistance of some stakeholders. Those stake- (OECD) (2012) on the quality and equity holders include some parents and teachers of education emphasizes that governments who see reforms as a risky watering down of should invest more in disadvantaged students standards that lowers the bar for what counts and schools “to ensure that everyone gets a as serious, demanding, academic, traditional fair chance” and identifies a number of con- education. Once again, only the systematic crete measures in that direction. First, grade dissemination of data on learning outcomes repetition should be eliminated; this wide- and therefore a better-informed and evi- spread practice is costly and ineffective;11 dence-based public debate on education can yet its academic benefits are slight and short address and mitigate this risk. lived. Second, student tracking and stream- ing policies should be avoided, at least until A paradigm shift for TVET the end of compulsory education (usually the start of upper-secondary school), because Another major option for opening up ave- they harm students who are moved down to nues of success for young people is the mod- lower tracks without raising student perfor- ernization of TVET. MENA’s challenge is mance as a whole. Focus group participants in to raise the capability of the TVET system Jordan noted that early tracking there, which and provide students with the skills required is usually done based on teachers’ recom- in the knowledge economy while offering mendations and grades and not on personal inclusive alternatives to nonacademic indi- aspirations or desires, limits students’ choices viduals. Traditionally, TVET has played two too soon in their school lives. The best evi- roles in educational systems (ETF and World dence of the advantages and potential gains Bank 2005). First, it has been an instru- of delaying tracking comes from Poland, one ment for preventing or offsetting the risks of of the success stories of secondary and TVET social exclusion that result from high rates reforms in the 2000s (Jakubowski et al. 2010; of early school leaving and school failure. In World Bank 2011). Third, school choice this common, “low-prestige” role, TVET is has to be managed to avoid segregation; for an alternative for those who fail academi- example, financial incentives could encourage cally or cannot be accommodated by higher the best schools to take vulnerable and disad- education. Second, TVET is used as part vantaged students. Finally, a research review of active labor market policies to combat of the most successful inclusive policies and youth unemployment and, to a lesser extent, measures in European schools has identified retrain workers affected by economic innovation in student grouping arrangements restructuring. In the 21st century, a third and parental and community participation and increasingly important role for TVET is in schools as the two most powerful inter- to support the development of a knowledge- ventions for improving learning outcomes based economy by helping workers adjust to among vulnerable, disadvantaged, and at-risk accelerating technological change and con- students in general (MEC-IFIIE 2012). See tribute to the competitiveness of enterprises. also box 9.3 on financial and nonfinancial With the exception of a small number of approaches to overcoming the equity chal- schools (see the example in box 9.4), TVET lenge in tertiary education. systems in MENA have mainly provided a Reforms that enable learning to prevail second-choice, lower-quality alternative to over selection—eliminating grade repetition, general education, with limited opportuni- providing alternative curriculum options and ties for lifelong learning. The task for TVET pathways to prevent students from dropping today is to combine a remedial role with the 260 JOBS FOR SHARED PROSPERITY BOX 9.3 Mechanisms for overcoming the equity challenge in tertiary education According to recent research, the most effective poli- age to encourage them early on to think about attend- cies for promoting equity and increasing opportuni- ing university. In the United States, Texas provides ties for disadvantaged students in tertiary education automatic admission into the University of Texas, the are those that combine fi nancial aid with measures state’s flagship public university, for the top 5 percent to overcome nonfi nancial obstacles. Strong evidence of all high school graduates in the state. In a number shows that well-targeted and well-managed financial of countries, Brazil and India, for example, affi rma- aid can be instrumental in reducing financial barriers tive action programs help increase opportunities for to tertiary education. A combination of three methods tertiary education for students from underprivileged can help students from disadvantaged groups over- ethnic groups. A recent evaluation of the preferential come financial obstacles to tertiary education: (1) no admission program at UNICAMP, one of Brazil’s top tuition fees or low fees; (2) scholarships and grants; public universities, reveals that the Afro-Brazilian and (3) student loans. With the exception of the West students admitted under that program have the same Bank and Gaza, tuition in MENA is already low in completion rate as other students. The Ford Founda- most public tertiary educational institutions. One of tion has provided funding to a number of universities the main challenges for governments in the region is in Sub-Saharan Africa and Latin America in sup- to ensure that existing scholarship and grant schemes port of retention programs targeting underprivileged are effectively targeted and managed to support stu- groups at risk of dropping out. dents from the lowest quintiles and girls as a matter of Finally, Colombia offers a good example of the priority. Experience from the 60 or so countries with integration of fi nancial aid and nonfi nancial mea- student loan programs—mainly in North and South sures for promoting equity. The Colombian Student America and in Asia and the Pacific—indicate that, Loan Agency, the fi rst such institution established when interest rate subsidies are kept at a reasonable in the world (1951), has recently implemented a level and repayment mechanisms operate efficiently, “comprehensive strategy” that involves participat- these programs are one of the best ways of promot- ing tertiary educational institutions in a partnership ing financial sustainability and equity at the same in which loan benefi ciaries receive not only fi nan- time. The Australian income-contingent student loan cial aid to pay the cost of their studies and living system, in particular, illustrates the advantages of an expenses but also academic and psychological sup- approach that conciliates efficiency in loan collection port to increase completion rates. A recent impact and fairness in the distribution of the payment burden. evaluation has shown that the probability of drop- Many countries have successfully implemented ping out is four times less for beneficiaries of student outreach and bridging programs to secondary schools loans than for other students. whereby universities build partnerships with K–12 institutions and reach out to students at a very young Source: Salmi and Bassett, forthcoming. BOX 9.4 Don Bosco in Egypt The Don Bosco Institute, run by the Italian Sile- of workers who can actually perform the activities sian Brothers, is a successful example of a private, corresponding to their certifi cates, so that the cer- fee-paying model for TVET. Rather than produce tificates effectively guarantee quality. This outcome old-fashioned second-chance diplomas, the institute comes about not just because relevant curricula offers three- and five-year diploma courses that pro- are taught but because the institute’s trainers are vide a path to employment, decent pay, and career recruited based on their technical and pedagogical progression—the very elements lacking in most instruction and practice. public TVET institutions. The institute’s credibility and trustworthiness lie in its consistent production Source: UNDP 2010. EDUCATIONAL AND TRAINING SYSTEMS 261 provision of cutting-edge skills, sometimes pathways equivalent in certification. Prior- even within a single institution. ity quality improvements cover a wide range The first step in carrying out that para- of issues, such as curriculum development, digm shift is to strengthen demand for TVET teacher and trainer development, better institutions and programs by diversifying testing and outcome measurement, M&E, provision modes in upper-secondary and ter- updating standards for qualifications, tiary education. This step implies opening and accreditation mechanisms (King and up pathways from secondary-level TVET to Palmer 2010). tertiary education to ensure that significant Critical to improving the quality of proportions of TVET graduates have a real TVET are better qualifications and greater chance of taking those pathways (which, diversity in the mechanisms for delivering again, also calls for a reform of the examina- those qualifications. Initial and continu- tion system, as discussed). ing TVET qualifications, either delivered Thus, a lifelong learning approach for through formal education and training modernizing TVET will demand reform or by nonformal courses, should be well in the articulation between tertiary educa- articulated to help individuals market their tion and TVET. Such reforms could expand skills. The limited alternatives for deliver- TVET graduates’ options for higher educa- ing T VET services should be expanded tion by developing a system of quality com- by exploring and developing e-learning munity colleges, technical institutes, and and other options for expanding access to polytechnics; setting incentives to enlarge those services and improving their quality the private tertiary education sector; and (OECD 2009a). creating distance teaching universities A paradigm shift for TVET, both the and virtual campuses. Miller-Idriss and medium-to-long-term measures and the Hanauer (2011), however, point to a clear appropriate short-term changes (see below), political-economy constraint: “Higher edu- are a priority for the whole MENA region. cation throughout the region is challenged These measures include curriculum reform by a high degree of centralization, a lack (Lebanon), addressing and altering the of incentives for improvement, and lim- fi nancing of TVET (Egypt, Jordan, Oman, ited mechanisms for reform or for evaluat- and the West Bank and Gaza), and accred- ing reforms.” In such an environment, it is iting and incentivizing private providers even more difficult for tertiary institutions (Jordan, Saudi Arabia). to channel students’ strong demand for pro- grams aimed at public sector jobs toward Short-term measures that more widely employable degrees that bet- MENA countries could adopt ter respond to the demands of an emerging private sector. Many countries in the region Establish a two-step examination system still compel private institutions of higher Establishing a two-step examination sys- education to mirror the curricula of their tem, with a secondary graduation exam public counterparts. While private institu- followed by a university entrance exam, tions have greater freedom in staff man- would separate the criteria for leaving or agement, the curriculum restrictions none- graduating from secondary school from theless represent a major impediment to the criteria for entering university or other responding to any emerging market needs. institutions of higher education. The high A second step in modernizing TVET is stakes of the examination would be dimin- to encourage students to enroll and gradu- ished, and a greater proportion of students ate by improving the quality of vocational would have a better chance of succeeding training courses at the upper-secondary at graduation. level, including work-based training, and Placing more weight on secondary school by making the different upper-secondary grades, class rank, school attendance 262 JOBS FOR SHARED PROSPERITY records, and other indicators of student upper-secondary education would break work during the upper-secondary years the association of TVET tracks with failure would also reduce the pressure on the last or low ability and facilitate a more inclusive year of upper-secondary education and transition from lower- to upper-secondary minimize the advantages implicit in private education. tutoring. Even more important, it would diversify student assessment instruments Regulate private tutoring and, as a result, amplify the range of skills Among other measures, public school teach- that can be used to qualify for graduation ers should be prevented from privately tutor- and further education. Even countries where ing students in their schools. It is very likely the national examination has tradition- that families invest excessively in private ally performed a double function (gradu- tutoring because tutoring and cramming ation and university entrance) have moved do in fact help, owing to the nature of the to minimize the high-stakes and high-risk exams. If the exams were more rigorous and nature of the exam. For example, the French relevant in the sense of testing real prob- Baccalauréat involves testing some of the lem solving, then tutoring might not help as common subjects (French, history) one year much; or if it did help, it would actually add earlier, placing more emphasis on a dossier value to young people. of the student’s work in some modalities of the examination, and introducing a subject called travaux pratiques encadrés, which Prioritize and invest in programs to prevent consists of group work to develop soft skills students from dropping out and leaving such as teamwork, problem solving, and school early leadership. National, regional, and institutional pro- grams that prevent students from drop- ping out and leaving school early should Offer all general secondary and TVET be supported. In addition to regulatory students the opportunity to take the same changes to prevent uncertifi ed youngsters level of examination from leaving the school system, it is criti- A llowing all students to take univer- cal to promote innovation in schools and sity entrance exams has worked well in training centers oriented toward imple- advanced and high-performing educational menting inclusive practices (OECD 2012; systems. In France, for instance, diversi- MEC-IFIIE 2012). fication of the Baccalauréat examination led to the creation of a “technical” and a “vocational” Baccalauréat. In the 1980s, A renewed partnership between in just over a decade, the modalities of the educational and training systems examination increased from 4 to about 38. and the private sector The tremendous inclusive impact of this diversification was part of a policy goal to Making employability count depends on increase the share of Baccalauréat graduates the ability of the private sector to signal to 80 percent of the age group (Eckstein and which skills are needed in the labor market Noah 1996). and to help provide those skills. Effi cient and transparent intermediation between labor supply and demand, for example, Abolish the selective nature of the leaving through public or private employment exam and reduce the rigidity of tracking services, could help allay young people’s arrangements concerns that they will not succeed in the Abolishing the selective nature of the transition from employability to employ- ninth-grade leaving exam and reducing ment because of factors they regard as the rigidity of tracking arrangements in beyond their control. EDUCATIONAL AND TRAINING SYSTEMS 263 Communication between educational Chapter 6 presents evidence from MENA and training institutions and the on the difficulties and constraints that pre- private sector vent TVET governance bodies—notably TVET councils—from successfully involv- Involving the private sector in education and ing and engaging employers. A good part of training must shift from a focus on “consul- the reason for those failures is the marginal tation” to a focus on “engagement.” Such role that employers are given in such bodies. engagement can be facilitated and promoted Streamlining existing bodies to strengthen only if, as discussed, information gaps and their collective voice and to better articulate failures are addressed, the right incentives are employers’ views on skills is crucial. It is also in place—including fi nancial ones—and an necessary to reinforce employers’ role in a institutionalized dialogue takes hold among national governance agency that focuses on public education authorities, employers, and employment, skills, and qualifications and other key stakeholders. In other words, if reports to the central government (and, if TVET is to become a credible and respected applicable, to regional or local administra- option for students and their families, tions with some autonomy in education and employers must also perceive it as credible, training). Coordinating bodies in specific and this perception relies on much more than sectors can also help strengthen employers’ just consulting and informing them. Long- voice and promote their engagement. Special- term goals of this nature demand concrete ized sector entities like the Enterprise-Train- measures and interventions to promote and ers Partnerships in Egypt are a good example eventually institutionalize the private sector’s in that regard. El-Ashmawi (2011) points to engagement in education and training and, four core functions for these sectoral bodies, more specifically, in TVET. From the perspec- around which consensus is growing: (1) to tive of public education authorities, the long- provide labor market intelligence on a sec- term policy goal is to balance regulation with toral basis; (2) to support the development of innovation. This balance can be achieved by occupational standards; (3) to promote qual- reforming the governance and fi nancing of ity service provision in the sector; and (4) to TVET while creating the space, incentives, train teachers according to set standards.12 and enabling environment for innovation to flourish in both public and private provid- ers. Moreover, beyond TVET itself, a better- Expanding private provision of TVET functioning partnership between educational through incentives, partnerships, and and training systems and the private sector contracts should also help address the low-level equi- librium trap that appears to be behind the The second long-term policy option is to meritocracy deficit described in chapter 6. expand private provision of TVET through The first policy option to pursue is formal- tax incentives, public-private partner- izing and institutionalizing communication ships, and performance-based contracts. It and coordination channels between educa- entails providing equipment, training train- tional and training institutions and the pri- ers, and developing strong links among vate sector. Developing and enhancing these public and private TVET providers and channels—with an emphasis on the skills employers in curriculum development. An to be incorporated into the curriculum— example of such an initiative comes from call for governance reforms that bring key Egypt, where a TVET project funded by the stakeholders into the strategic decisions and European Union has developed close col- management of TVET policies and services. laboration among Siemens, the University Such reforms also include the development of of Heidelberg, and vocational training cen- accountability mechanisms and decentraliza- ters. Another example comes from Jordan, tion to ensure efficiency, relevance, and qual- where several private fi rms collaborate with ity in TVET service delivery. the Vocational Training Corporation to 264 JOBS FOR SHARED PROSPERITY establish model centers through PPPs (most with providers of education and training) notably in the tourism sector with fund- (El-Ashmawi 2011). Another very innovative ing from the U.S. Agency for International example (this one from Tunisia) is presented in Development). box 9.5. These pilots have certainly provided Also in Egypt, over the past few years a the foundation for a more ambitious approach number of international donors, in cooperation to systemwide reform. The governance reforms with the government, have invested in pilot suggested above are a necessary condition for reform programs to improve the attractiveness moving from innovative pilots to systematic of TVET by creating more powerful links with reform. the private sector and therefore more possibili- ties for employment. The most important of Efficient and transparent these is the Mubarak-Kohl Initiative, which intermediation is currently being implemented in a number of schools following the dual system13 and The predominance of informal networks for the Alternance education initiative introduced job search and hiring limits the efficiency by the European Commission–funded TVET of the job-matching process and introduces Reform Program (with its Enterprise-Trainers inequities, given that access to employment Partnerships, in which businesses partner depends strongly on personal connections. BOX 9.5 Tunisian Association for Communication and Technology Following the events of the Arab Spring, a Tuni- ICT fundamentals, ICT “environments,” languages sian business association formed to ensure progress (English and French), and communication. This cur- on the offshoring agenda: the Tunisian Associa- riculum is complemented by four months of on-the- tion for Communication and Technology (TACT, job training, six weeks of which cover a fi fth module l’Association Tunisienne pour la Communication et (with training in fi nance, computer science, project la Technologie). TACT promotes three objectives: management, or another topic). The 10-month pro- to make Tunisia known for its offshoring services; gram allows a trainee to pass a series of certifi ca- to operate in a think-tank capacity as a source of tions: one or more IT certifi cations (in Java, .net, ideas on offshoring for the government; and to help and so on) and in language (the TOEIC or TOESL). international companies locate offshore depart- TACT functions as a charitable trust, offering ments in Tunisia. all managerial and administrative services related The TACT pilot program is a public-private part- to this program free of charge. When candidates nership involving four ministries of the government are accepted into the program, the government of Tunisia (technology, economic development and pays 3,000 Tunisian dinars (TD) (US$1,985) for foreign investment, education and vocational train- tuition and the cost of certifi cation exams. Micro- ing, and higher education and research), public and soft intends to contribute to the program by paying private engineering schools, and the private sector. for the certification programs; in addition, thanks to Training is carried out by EsprIT, a private univer- the government’s recognition of the program, par- sity located in Tunis. ticipants retain the Amal (Arabic for hope) monthly Two hundred unemployed university graduates allowance of US$133 paid to support job seekers with a background in information and communica- over the 10 months of training. Candidates also tion technology (ICT) were selected to take part. An receive a free laptop (valued at US$462) that they estimated 30,000 ICT graduates are unemployed, of may keep. whom 17,000 are newly graduated. ICT graduates TACT has established this program on a results rapidly become “offshore-ready” through training basis. It accepts the full tuition of TD 3,000 from that lasts 10 months and includes a 6-month cur- the government only when students are placed in riculum with four 180-hour units of instruction in positions in the offshoring sector. EDUCATIONAL AND TRAINING SYSTEMS 265 The evidence on employment barriers pre- and unemployment. Repeated temporary sented in chapter 6 calls for more formal placements often lead to a permanent job, job-matching systems and mechanisms to and consequently it is in the interest of pub- permit more equitable, merit-based access to lic employment services to handle short-term jobs. The public sector has a role to play in work. providing formal and open intermediation For MENA countries, a closer partnership between job seekers and employers, as well as with private service providers, training and in enhancing the skills of unemployed work- educational institutions, employers, and com- ers and adjusting them to the needs of the pri- munities might be useful for ensuring that pro- vate sector through selected ALMPs (see, for grams respond to market conditions. Various example, Angel-Urdinola, Semlali, and Brod- methods can achieve this goal, including exten- mann 2010; Angel-Urdinola, Kuddo, and sive outreach by local office staff, job fairs, an Semlali 2013; Wazzan and Zovighian 2013). aggressive marketing campaign, and regular Although most MENA countries provide labor demand surveys (World Bank 2010). some form of intermediation through public Given that private recruitment agencies employment services, several challenges and are still at an early stage of development in shortcomings reduce their efficiency and ser- MENA, governments need to ensure proper vice provision. Public employment services regulation so that all rights are enforced, thus in MENA focus too strongly on expensive enhancing social acceptance for this form training and employment incentives rather of work.14 Private employment agencies are than on providing more cost-effective assis- often equated with temporary work agencies tance with job searches, including counsel- that facilitate short-term contracts and labor. ing. The last not only achieve employment They do, however, also perform job-match- results similar to other interventions at a ing services, particularly for employers with significantly lower cost but also address medium- and higher-skill requirements (like, the information asymmetries between job for example, headhunter services). seekers and prospective employers, thereby To improve the effectiveness of delivering reducing the importance of privilege in public employment services, Angel-Urdinola, access to jobs. Kuddo, and Semlali (2013) propose the fol- Public employment services assist prospec- lowing reform options: tive employers by posting vacancies, con- • Develop PPPs. Public employment services ducting database searches for job profiles, or in the region will need to develop partner- prescreening job applicants and matchmaking. ships with the private sector to deliver train- Most often, however, the approach of pub- ing and employment services to the stock lic employment offices is rather passive; they of unemployed (notably youth), promote expect employers to post vacancies and do not participation in internships and on-the-job actively build links with the local employer training, and develop demand-driven pro- community or market their services. Both pub- grams tailored to the needs of the private lic employment services and employers use a sector. These partnerships should assure the variety of other recruitment channels, includ- delivery of concrete results. One common ing personal contacts, advertising, applicant mechanism for doing so is to introduce initiative, and references from schools, con- result-based contracts, whereby providers sultants, and existing employees. In addition of employment services are remunerated to offering incentives to employers to increase based on their demonstrated capacity to vacancy notifications, the employment service connect beneficiaries to available intern- can register vacancies advertised elsewhere, ships and jobs. and it need not limit itself to longer-term jobs. • Promote entrepreneurship, on-the-job train- In many instances, temporary or part-time ing, and life skills training among job seekers. employment contracts can be regarded as an Service provision in the region should find a intermediate state between full employment better balance between (costly) interventions 266 JOBS FOR SHARED PROSPERITY that are cumbersome to scale up (notably (LMIS) (see box 9.6), and promote a cul- wage subsidies, vocational training, and ture of program evaluation and auditing. entrepreneurship promotion) and more cost- • Strengthen institutional and regulatory efficient interventions, such as labor market frameworks. Countries in the region need intermediation, the provision of soft-skills to improve institutional capacity to deliver training, and employment counseling ser- employment services by reducing fragmen- vices, and entrepreneurship awareness pro- tation in the system. Institutional coordi- grams that encourage experimentation and nation should be promoted across relevant learning among university graduates. agencies as well as between central and • Improve governance and accountabil- local agencies (given that the needs of the ity. Most countries in the region need to unemployed vary across localities). develop a clear framework (and allocate resources) to monitor and evaluate the efficiency and effectiveness of existing Short-term measures that MENA employment programs. For example, they countries could adopt could develop clear legal frameworks for M&E, develop results-based (not output- • Design and implement a communication based) labor market information systems strategy. A communications strategy could BOX 9.6 Introduction of a labor market information system in the United Arab Emirates A labor market information system (LMIS) is a from survey data (mainly a labor force survey, last set of institutional arrangements, processes, and administered in 2009). The data are then put into tools for the collection, integration, analysis, policy a data “warehouse” at the Ministry of Labor, and formulation, and dissemination of labor market the LMIS Unit uses business intelligence software information. Such a system assists in minimizing to analyze and report on the data. Custom reports the information gaps that lead to mismatches and are generated and provided to senior government distortions and signal information on supply and officials and other stakeholders. The cornerstone of demand for skills to the various stakeholders that the LMIS is a set of 23 key labor market indicators formulate and implement human resource develop- developed specifically to respond to policy priorities ment, employment policies and programs, and pri- in the United Arab Emirates. The policy priorities vate sector decisions. An LMIS also allows policy include “emiratization,” fl exibility, demographics, makers to (1) monitor developments in the labor protection, and productivity. In addition, the LMIS market; (2) monitor the underlying shifts in the is governed through a multiagency steering commit- workforce, the skills dynamics in an economy, and tee that includes the major providers (and users) of how educational and training systems can respond; data. Memoranda of understanding are established (3) develop targeted and evidence-based policy between the Ministry of Labor and other agencies to interventions; and (4) assess the impact of policies ensure timely sharing of information. The dissemi- on the labor market. nation of labor market information is done through The World Bank has been supporting some a communications strategy that recognizes the vari- MENA countries in the development of LMIS. In ous stakeholders involved and provides information the United Arab Emirates, the LMIS collects infor- relevant to each group. In the medium term, the mation on all aspects of the labor market, including ministry is looking to strengthen the job-matching unemployment and employment, vacancies, types functions by matching nationals’ applications with of employment, wage data, productivity, and the labor applications for foreign labor. In addition, the like. Data are collected either from administrative ministry intends to develop occupational projections records (such as foreign labor registries, social insur- (by occupation, skills, industry, and the like) to aid ance registries, civil service records, and so forth) or in signaling across supply and demand. EDUCATIONAL AND TRAINING SYSTEMS 267 explain the merits of TVET as a lifelong beyond that, Morocco has adopted a process learning subsector and contribute to remov- of delegated management, whereby the voca- ing the stigma associated with TVET paths tional training minister signs a contract or in MENA. Such strategies must explicitly memorandum with an industry group that target employers in addition to students, then manages a public training center and, families, and public opinion. In Egypt, for in so doing, defines the skills to be deliv- example, ad hoc awareness campaigns that ered, designs the training program, includes were not conducted in the framework of a internships and job placements in the pro- communications strategy triggered mixed gram, and involves professional staff from results at best. In 2007, the Industrial the industry in evaluation. Morocco has four Training Council partnered with the Fed- such centers in three sectors (fashion, aero- eration of Egyptian Industries under the nautics, and automobile). sponsorship of the Ministry of Finance to • Achieve sustainable, effective, and equi- commission a large media campaign to table fi nancing of TVET. Public funding promote a positive attitude toward indus- should be continued for TVET systems, trial employment. The campaign was partly through increasing the number complemented by pre-employment train- and amount of scholarships as an addi- ing for youth to increase the likelihood of tional incentive to create demand. While matching vacancies in private firms as well the promotion of that new partnership as by some monitoring and evaluation. between the educational and training sys- Acceptance of vocational training among tem and the private sector mostly involves young people was less than 2 percent in governance-related reforms, alternative the precampaign survey. As the campaign fi nancing incentives and mechanisms will rolled on, 10 percent of inquiries (from also be needed so that the educational and among 1,100,000 callers) expressed a training system can respond to a more willingness to be employed on production dynamic and engaged private sector. Thus, lines. Of these, 110,000 were interviewed, suffi cient and predictable fi nancing will and 87,000 (8 percent) became employed be needed to create demand for TVET, (UNDP 2010). In Jordan, a communica- to attract higher-caliber students, to sus- tion strategy is being developed through a tain training programs and qualifications consultative approach with all stakehold- consistent with local and regional labor ers, relying heavily on market research for market trends, and to ensure high-quality perceptions and attitudes that can help tai- practical training in private companies. lor the awareness campaign. • Provide incentives for accreditation of • Reform, relicense, and empower sec- private institutions. Accreditation of tor skills councils to increase employ- training institutions needs to be encour- ers’ engagement and investment in skills aged through policies such as tax incen- development. In the United Kingdom, for tives to the private sector for engaging instance, the Leitch Review of Skills pro- accredited organizations. posed to deliver more economically valu- • Encourage early intervention. Early in an able skills by allowing public funding for unemployment spell, employment services vocational qualifications only when the should intervene to provide job search content had been approved by the sector assistance and counseling through high- skills councils (Leitch 2006). contact density between job seekers and • Promote business engagement in the class- employment counselors. room and in the management of TVET • Reorient public employment services training centers. Private sector representa- toward the private sector labor mar- tives should have the opportunity to partici- ket. In some countries in MENA, public pate in classroom settings (for example, as employment services focus on job match- visiting lecturers or adjunct teachers). Going ing in the public sector. These efforts 268 JOBS FOR SHARED PROSPERITY and resources should be directed toward wage subsidies or public works programs the productive private sector. Innovative can help them gain work experience and financial and career incentives for job obtain temporary income support. In dis- seekers could be explored to encourage cussions with education policy makers from them to give up waiting for a public sec- MENA, providing second-chance options tor job and actively search in the private outside the educational system through sector. The use of employment subsidies ALMPs is high on the list of medium-to- could be explored for helping bridge the long-term measures in Jordan, Oman, and gap between market wages and job seek- Saudi Arabia, whereas Jordan, Tunisia, and ers’ reservation wages. the West Bank and Gaza prioritize investing • Create more opportunities to support in entrepreneurship and coaching as a short- job seekers. Job-matching services, career term measure (see table 9A.1 in the annex). guidance, and counseling could be scaled up, and services could be provided to a Provide comprehensive ALMPs larger and more diverse group of job seek- ers. Innovative ways to better target youth A package of services for young job seekers and women in job matching and training with medium-to-high skills, including job could be explored to ensure their equita- counseling, matching services, and train- ble access to the programs. Women and ing in soft skills, should be provided. Most youth form the bulk of MENA’s unem- ALMPs targeting young people in MENA ployed and inactive; yet publicly provided focus only on providing job-related skills, services do not target these groups sys- and very few provide on-the-job training, tematically. Such targeting, coupled with let alone individualized case management a means test, would ensure that programs (Angel-Urdinola, Semlali, and Brodmann reach those most in need. 2010). International evidence shows that • Regulate and coordinate with the private comprehensive programs that target youth sector. If the private sector finances, man- through PPPs are successful in increasing ages, and implements employment services, employability. Several countries in Latin the primary role of public agencies would be America have moved from in-classroom to ensure proper regulation and policy coor- training to a package including workplace dination among public and private provid- training and additional services such as ers, which balances security with flexibility. counseling, mentoring, and job search and placement assistance, as well as training in soft and life skills.15 A well-known pro- A role for active labor market gram is the Chilean government’s Chile policies Joven program, which offers comprehensive Young fi rst-time job seekers who have diffi- employer- and demand-driven training pro- culty in mastering the challenging transition grams to unemployed youth. This program’s to work as well as particularly vulnerable success led to customized replications of the groups, such as low-skilled youth or women, model in several other countries. Impact need special services outside the educational evaluations in Argentina, Chile, and the and training system. The following policy Dominican Republic indicate that program options focus on ALMPs—active labor participants display a 10–21 percent greater market policies—and highlight some of likelihood of finding employment than simi- the same priorities as identified above: the lar young people who did not attend the importance of counseling, matching supply program. Earnings among those who find and demand, and strengthening soft skills employment after the program are also in training curricula. In addition to training higher than among young people who fi nd or entrepreneurship support for those with employment but did not participate in the medium or high levels of skills, temporary program (World Bank 2006). EDUCATIONAL AND TRAINING SYSTEMS 269 The evaluations identified the following to hire young workers without experience, key success factors: (1) PPPs and training particularly females, because employers programs that offer practical experience generally have even less experience work- linked directly to an internship with a pri- ing with women and might regard them as vate employer previously identified by the less committed to staying employed. Wage training institution; (2) training that pro- subsidies to private fi rms sometimes encour- vides youth not only with technical skills but age employers to risk hiring new entrants, also with soft and life skills; (3) a program and several evaluations from Europe show that provides flexible schedule alternatives, a large beneficial impact on employment such as evenings and weekends, to ensure (OECD 2005). Evidence suggests that that youths who work or care for children wage subsidies can work to the advantage during regular business hours can attend; of women and youth (Galasso, Ravallion, and (4) the regular monitoring of success- and Salvia 2002). Subsidies are particularly ful programs so that they can be improved effective when combined with other pro- continuously (Angel-Urdinola and Semlali grams, such as on-the-job training, counsel- 2010). ing, and job search assistance (Kluve 2006). See examples from Jordan and Turkey in box 9.7. Invest in entrepreneurship training There are a few caveats. Although wage and coaching subsidies have seemed cost-effective in Many aspiring entrepreneurs in MENA countries such as Argentina and Poland have had little formal education or training (Cunningham, Wuermli, and Sanchez- when they enter the entrepreneurial labor Puerta 2010), in other countries they have force. The knowledge and skills that many often shown a negative cost-benefit bal- currently use in their businesses have likely ance because of deadweight loss, substitu- been acquired from prior experience or tion, and displacement effects (Calmfors from relatives in family enterprises. Chap- 1994). To be effective, wage subsidies have ter 7 showed the importance of prior work to be targeted well and be limited in dura- experience to transition successfully into tion; employers’ obligations also have to be entrepreneurship; however, aspiring entre- enforced. It is important to ensure that the preneurs still need to strengthen business subsidy goes to those employers who would know-how (such as numeracy and fi nancial not have hired young workers in the absence management skills), as well as to facilitate of the additional fi nancial incentive. access to credit and appropriate technology. Paying for training and other services, as Provide labor-intensive public works well as fi nding and selecting quality train- and training in job-specific skills ing, can be problematic. Governments can bridge the gap through targeting prospective By providing market-relevant skills and gen- entrepreneurs early and linking them with erating local infrastructure, public works relevant sources of knowledge and support can relax some of the constraints to market (see box 7.12 in chapter 7 on Tunisia’s busi- access that rural workers face and generate ness plan competition). effects that last beyond the program’s dura- tion. Aside from temporarily employing unskilled and semiskilled workers at low Subsidize wages and social security wages on labor-intensive activities to cre- contributions for women and young ate new community infrastructure, public people works programs can also serve as a tempo- Wage subsidies can be a means for improv- rary safety net, ensuring income in times of ing opportunities and the employability of shock (such as natural disasters or macro- youth and women. Employers are reluctant economic crises). For example, the Egypt 270 JOBS FOR SHARED PROSPERITY BOX 9.7 Wage subsidy programs in Jordan and Turkey Jordan’s training program to increase one-quarter fewer women in the training group participation of women in the workforce indicated that they are severely mentally depressed The World Bank’s Jordan NOW (New Work compared with the control group. However, there Opportunities for Women) pilot combined two are indications that the challenges faced by young interventions aimed at reducing the barriers to entry female entrants in the Jordanian labor market are into the labor market for young women: a short- far more serious: on the whole, soft-skills training term wage subsidy and training in employability had no sustained impact on employment. Neither skills. The soft-skills training was designed and did it affect any other dimensions of employment: implemented jointly with a local Jordanian nongov- salary, weekly hours worked, labor force participa- ernmental organization in the fall of 2010. From a tion, and total time employed. group of 1,347 women who had recently graduated from community colleges in Jordan, about 600 girls Subsidizing social security contributions for were randomly selected to participate in the nine- youth and women in Turkey day-long (45-hour) training course. Following dis- In Turkey, a subsidy designed for new youth and cussions and feedback from private sector fi rms on women hires between July 2008 and June 2009 con- the lack of soft skills among young entrants into the sisted of 100 percent of employers’ social security labor market, the course was designed to cover a contributions (at the legal minimum wage) in the first series of basic employability skills. These included year, with a 20-percentage-point decline per year in effective communication and business writing (for the subsidy in the following four years. An evaluation example, making a presentation and writing busi- of a similar program piloted in selected regions of ness reports and different types of correspondence), Turkey between January 2004 and December 2005 team-building and teamwork skills (such as explor- found that employment among youth and female ing characteristics of a successful team and how to workers in provinces that benefited from subsidies work in different roles within a team), time man- grew by 1.7 percent per month, whereas the growth agement, positive thinking and how to use this rate was 1.1 percent per month in provinces that did skill in business situations, excellence in providing not benefit from the subsidy (in other words, subsi- customer service, and résumé and interviewing dies contributed to a 63.0 percent increase in employ- skills. Sessions were based on active participation ment growth rate in beneficiary provinces). Based and cooperative learning rather than lectures, with on the results of this pilot experience, the effects games, visual learning experiences, group exercises, of a similar program implemented nationally could and active demonstrations used to teach and illus- have contributed to a net creation of approximately trate concepts. 163,000–235,000 new jobs. Unfortunately, Turkey A year after the soft-skills training courses was hard hit by the financial crisis, which under- ended, the World Bank conducted a follow-up mined the program’s impact. Programs like this, survey to assess the impact of the interventions. while attractive for promoting employment among Countless numbers of participants gave strongly targeted groups, could be expensive. Estimates positive feedback about the skills training: that from the Turkish Treasury indicate that generating the experience opened their minds and gave them one extra job for those benefiting from the program a much better grounding for getting a foothold in could cost US$12,000–17,000. the labor market. The results of the survey back up this feedback to some extent: training helped Sources: Jordan NOW baseline, midline, and endline surveys; staff calculations; boost these recent graduates’ self-esteem—about Angel-Urdinola and Semlali 2010. EDUCATIONAL AND TRAINING SYSTEMS 271 Social Fund for Development created jobs funding (Del Ninno, Subbarao, and Milazzo and provided new community infrastruc- 2009). ture and services as part of the government’s safety net and poverty reduction strate- Invest in early childhood development gies. The Republic of Yemen Social Fund and home-based work for women for Development has implemented a labor- intensive work program since 2006 that “Workfare” programs need not always serves as a poverty alleviation mechanism involve public works. Instead of providing in urban areas for many unskilled workers new infrastructure, for example, the work arriving from rural areas (Del Ninno, Sub- can supply an underprovided public service, barao, and Milazzo 2009). For an example such as child care. This type of workfare is from India, see box 9.8. not necessarily temporary and not reserved Although public works seem to have a pos- to the unskilled, as an example from South itive impact on employment outcomes only Africa shows (box 9.10). in the short run, they have the potential to This type of workfare can be financed generate effects that endure beyond the life through the demand side, because there are of the program by developing infrastructure service users. For example, a new pilot pro- and providing market-relevant skills (Dar gram in Tunisia (Bedaya) explores the poten- and Tzannatos 1999; Betcherman, Olivas, tial for combining a program such as the one and Dar 2004). Infrastructure development in South Africa with child care vouchers for may unleash productive capacity, lower the poor working women. The vouchers, a ben- cost of doing business for micro and small efit from the Ministry of Social Affairs, could enterprises, and raise their access to wider be used to pay for the newly established child markets and customers. Box 9.9 explores the care centers. The centers can redeem the ser- links between infrastructure development vice vouchers at the ministry against cash and employment in MENA. revenue. While this program actively creates A review of public works projects in sev- new jobs, especially for qualified women (as eral countries over the past 20 years shows child minders), it is not workfare in the tra- that a number of design features are critical ditional sense. The child care managers act to their effectiveness: they need to have clear like entrepreneurs and need to attract enough objectives, they should create valuable pub- clients or service users to make their centers lic goods, and they should have predictable profitable. BOX 9.8 An employment guarantee scheme to build infrastructure in rural Maharashtra India has a long history of using employment guaran- at piece rates. The scheme has succeeded in targeting tee schemes to provide income relief to the poor and the deserving segment of the population and building build needed infrastructure. Introduced in the 1970s, rural infrastructure. In 2004, the Indian government the employment guarantee scheme of the Indian state enacted the National Rural Employment Guarantee of Maharashtra is one of the largest public works Act, which guarantees 100 days of employment in programs in the developing world. It guarantees public works to rural people and is projected to cost employment to all adults older than 18 years of age 1–2 percent of gross domestic product. who are willing to perform unskilled manual work Source: OECD 2009b. 272 JOBS FOR SHARED PROSPERITY BOX 9.9 The impact of infrastructure investments on employment Investment in infrastructure could be a natural Lebanon would fi nd it considerably easier to fi nance “early gains” measure for creating jobs in MENA the investment expenditure. Spending on construc- (see also the discussion in chapter 10). In the short tion of roads and bridges would generate more than run, every US$1 billion invested in infrastruc- twice as many direct jobs as the same amount of ture has the potential to generate around 110,000 spending in any other sector because the work is infrastructure-related jobs in the oil-importing more labor intensive. To reflect the variation across countries (OIC), 26,000 jobs in the economies of sectors, fi gure B9.9.2 shows the estimated cost of the Gulf Cooperation Council, and 49,000 jobs in creating one direct job in various sectors in Egypt the developing oil-exporting countries (OEC) (see in 2009. Sectors also differ in their propensity to figure B9.9.1). The region could therefore generate generate indirect jobs. It depends on the extent to 2.0 million direct jobs and 2.5 million infrastruc- which the sector requires inputs from other sectors ture-related jobs just by filling current domestic to produce its output. infrastructure gaps. FIGURE B9.9.2 Cost of creating a direct job in FIGURE B9.9.1 Estimated jobs creation through road and bridge construction relative to other infrastructure investment per billion U.S. dollars sectors in the Arab Republic of Egypt, 2009 of investment in MENA, 2009 45 1,200,000 40 35 1,000,000 30 Percent Number of jobs 800,000 25 20 600,000 15 10 400,000 5 200,000 0 ge ns n gs es vic d tio er an vic tio in wa 0 es uc ild n s ort sta er se str Bu s tio p tu n nt s tu n nt s d ity ity s te ob te ob on uc /b uc /b an ica ran ric ric re To ial re ial str bs str bs po ct j po al j rc er un T ct ct fra jo fra jo he t at re Ele To Ele in rect in tal W Ot Di Di m m co GCC countries Oil-importing countries Oil-exporting countries Apart from quick wins, the long-term employ- Note: GCC = Gulf Cooperation Council; bn = billion. ment effect of infrastructure investment could be significant. Estache et al. (2012) fi nd that infrastruc- ture investment resulting in one percentage point of The potential of such programs varies by sec- additional growth could create 9 million additional tor and country. Because of per capita income jobs in MENA in the course of 10 years, or a little differences, US$1 billion in infrastructure invest- less than 1 million jobs per year. ment would generate more than six times as many jobs in a sector in Djibouti as in Lebanon, but Source: Estache et al. 2012. EDUCATIONAL AND TRAINING SYSTEMS 273 BOX 9.10 Supporting early childhood development and home care services in South Africa A public works program in South Africa set out corresponding to accreditation of teacher aides and to provide temporary jobs, skills, and early child- kindergarten teachers; (2) direct and immediate hood accreditation to 19,800 young unemployed creation of work opportunities in targeted early (mainly female) persons over five years. The goal childhood development sites in very poor areas; was to combat unemployment while investing in the (3) on-the-job training and certification for early child- future generation through early childhood develop- hood development support staff, such as vegetable and ment and home-based care programs. Youth are legume gardeners, cooks, and administrators; and hired at a minimum wage, receive training, accredi- (4) short-term, three-month employment opportuni- tation, and designated jobs providing services for ties in auxiliary tasks for 3,000 unemployed parents the very young in their communities. Training and through existing schools and local authorities. employment include various levels: (1) qualifications Source: Antonopoulos and Kijong 2011. Skills as drivers of success in the in the “right” areas and reduces the mis- labor market match between what the market needs and what the educational and training system Building employability and making it count produces. Promoting meritocracy reinforces in the labor market are essential condi- the conditions for reform identified here: clos- tions for efficient and equitable labor mar- ing stakeholders’ information and knowledge ket outcomes. Young people and families gaps; valuing learning and problem solving in in MENA invest heavily in education and more inclusive educational and training insti- training; yet the majority of youth cannot tutions; involving employers in education and reap the appropriate individual returns from realigning incentives for public sector hiring; such investments. Several constraining fac- and improving the efficiency of job matching tors intervene, including the low quality while also providing second-chance options and relevance of their skills and the limited through active labor market programs. importance of merit in gaining access to a Citizens’ demand for agency and dignity job. From a social perspective, returns to in MENA underlines the importance of investments in education are low in MENA, involving civil society in crafting and imple- partly because the number of good employ- menting policy solutions. The policy options ment opportunities in the private sector is outlined in this chapter are certainly not limited and partly because so many young new. Most will take a long time to imple- people choose to queue for jobs in the pub- ment properly and bring about measurable lic sector. High youth unemployment—both results; yet the fi rst step forward is the most involuntary and voluntary—is a result of crucial one to take. But, to be sure, success this low-productivity equilibrium. will require more than technocratic solu- Meritocracy in access to education and tions. It will require a society that is ready hiring, the availability of multiple pathways for change, that is ready to provide real in education, and the provision of second- opportunities for future generations, and chance options are key elements in develop- that has leaders in the public and private ing a productive workforce. A meritocratic sector alike who are prepared to make a real society signals market demands more clearly commitment to building employability and to educational and training systems. As a making it count in the labor market. result, it creates demand for the “right” skills 274 JOBS FOR SHARED PROSPERITY Annex TABLE 9A.1 Selected short- and medium- to long-term educational policies and measures for MENA Early gains/short-term measures Medium-term to long-term measures Egypt, Arab 1. Integrate databases into a national education management Establish and develop a national system of student Rep. information system assessment 2. Update curriculum standards to enable a more ambitious Address the financing gap for technical and vocational curriculum reform in secondary education and vocational education and training (TVET) education and training (VET) Transform TVET from a second-rate option into a life-long 3. Complete the National Qualifications Framework (NQF) learning (LLL) system Jordan 1. Reform the examination system (Tawjihi), mainly by Formalize and institutionalize communication and establishing a two-step examination system, with a coordination channels between educational and training secondary graduation exam, followed by a university institutions and the private sector entrance exam, and by switching the focus of the exam to Provide second-chance options outside the education system higher-order skills and competencies through active labor market program (ALMPs) 2. Complete the modernization of the curriculum 3. Design alternative training itineraries for every student, so that each acquires relevant skills for a successful transition from education to work 4. Design and establish tracer studies for both technical and vocational education and training (TVET) and tertiary education graduates 5. Regulate private tutoring 6. Expand private provision of TVET through tax incentives, public-private partnerships, and performance-based contracts 7. Invest in early childhood development and home-based work specifically targeted to women 8. Invest in entrepreneurship training and coaching Lebanon 1. Open new channels between TVET and tertiary education Formalize and institutionalize communication and and reform the university entrance examination coordination channels between educational and training 2. Complete the NQF in partnership between the public and institutions and the private sector the private sector Develop a reformed national system of student assessment 3. Strengthen counseling and guidance services in all schools 4. Reform the curricula of TVET Morocco 1. Reform the examination system (Baccalauréat) and Formalize and institutionalize communication and evaluation of student performance as a whole so that the coordination channels between educational and training focus is on skills and competencies institutions and the private sector 2. Create a Labor Market Information System Establish and develop a national system of student 3. Systematically carry out tracer studies for graduates assessment West Bank and 1. Reform the secondary-school leaving and university Establish and develop a national system of student Gaza entrance examination (Tawjihi). assessment 2. Develop the Education Management Information System and the Labor Market Information System 3. Invest in counseling and guidance services 4. Achieve sustainable financing for TVET 5. Invest in entrepreneurship training and coaching Tunisia 1. Refine the quality assurance function in the education Create a comprehensive information system that is used as system, strengthening the separation between policy- the key policy tool for labor market–related decision making setting functions and evaluation and oversight ones 2. Reform the examination system (Baccalauréat) and evaluation of student performance, as a whole, so that the focus is on skills and competencies 3. Invest in entrepreneurship training and coaching 4. Strengthen and professionalize counseling and guidance services, particularly in VET 5. Regulate private tutoring (continued next page) EDUCATIONAL AND TRAINING SYSTEMS 275 TABLE 9A.1 Short-and-medium-to long-term educational policies and measures for MENA (continued) Early gains/ short-term measures Medium-term to long-term measures Oman 1. Create a labor market information system Improve the collection of data on numbers, skills, and wage 2. Systematically carry out tracer studies for graduates evolution of graduates of secondary and TVET institutions, 3. Continue the process of establishing a professionalized universities, and certification programs to facilitate career service of counseling and guidance for all schools decision making and to enhance efforts to promote investment 4. Ensure public funding for VET Design alternative training itineraries for every student so that 5. Design and implement a communication strategy that each acquires relevant skills for a successful transition from explains the merits of TVET as a life-long learning subsector education to work and contributes to removing the stigma associated with TVET Transform VET from a second-rate option to an LLL system 6. Invest in early childhood development and home-based Provide second-chance programs through ALMPs work specifically targeted to women 7. Invest in entrepreneurship training and coaching Saudi Arabia 1. Create a labor market information system Formalize and institutionalize communication and 2. Systematically carry out tracer studies for graduates coordination channels between education and training 3. Provide incentives for the accreditation of private institutions and the private sector institutions in TVET Transform TVET from a second-rate option into an LLL system Provide second chance options outside the education system through ALMPs Iraq 1. Establish a two-step examination system with a secondary Strengthen and rebalance the links between general and graduation exam followed by a university entrance exam vocational tracks and introduce new, blended training 2. Place more weight on secondary school grades, school itineraries attendance records, and other indicators of student work Continue progress toward a competency-based curriculum during the upper secondary years reform 3. Develop the national system of student assessment 4. Finalize a long-term education strategy that is based on quality data. Yemen, Rep. 1. Improve the collection of data on numbers, skills, and wage Formalize and institutionalize communication and evolution of graduates of secondary and TVET institutions, coordination channels between education and training universities, and certification programs to facilitate career institutions and the private sector decision making and to enhance efforts to promote Transform TVET from a second rate option into an LLL system investment Continue progress toward a competency-based curriculum 2. Prioritize and invest in programs to prevent vulnerable reform students from dropping out and leaving school early Source: Feedback collected by the report team from Ministry of Education/Higher Education officials on reform priorities (World Bank Course on Strategic Choices in Education Reform, Muscat, Oman, June 18, 2012). Note: MENA = Middle East and North Africa. Notes ~pagePK:2865066~piPK:2865079~theSit ePK:6130003,00.html. 1. Table 9A.1 shows that creating management 4. Many countries have created or are in the pro- information systems and integrating existing cess of creating NQFs. East and South Asian data or conducting new surveys is high on the countries such as Australia, Hong Kong SAR list of priorities in many MENA countries. (China), Malaysia, the Philippines, Singapore, 2. Concrete examples from selected countries on and Sri Lanka have NQFs that the International how to use assessment data to design better pol- Labour Organization deems “established” (ILO icies can be found, for instance, in Greeney and 2010); the same goes for many members of Kellaghan (2009) and in World Bank (2011). the European Union (such as France, Ireland, 3. The full text of the Doha Declaration can Romania, the United Kingdom, and others) and be found in the following link: http://web for a good number of countries in other regions. .worldbank.org/WBSITE/EXTERNAL/ In MENA, Egypt, Jordan, Lebanon, and Tunisia COUNTRIES/MENAEXT/EXTGC/0,,co have established NQFs (Leney 2009). ntentMDK:22711414~menuPK:6130053 276 JOBS FOR SHARED PROSPERITY 5. See the annex to this chapter for details on 13. A dual system combines apprenticeships in suggestive country-specific short-term poli- a company with vocational education at a cies and measures. vocational school. 6. For example, evidence from a United States– 14. International instruments that provide guide- wide sample of high school students suggests lines for regulating private employment services that guidance and counseling services improved include Convention 181 of the International the quality of students’ educational and occupa- Labour Organization as well as the accompa- tional decisions and educational performance nying Recommendation no. 188. (Lapan, Gysbers, and Sun 1997). For young 15. For example, a life skills training to improve women, such counseling may have positive employability in the Dominican Republic, externalities on the community. For example, a called Juventud y Empleo, integrates life skills study in India found that women who received education to help young people become more three years of career and recruiting services had effective employees and citizens. Among the higher employment rates than women who topics taught are self-esteem, teamwork, had not received the service; among girls ages communication skills, work organization and 5 to 15, the service increased the probability of service skills, job search skills, and knowl- school attendance (Jensen 2010). edge related to risky behaviors (reproductive 7. The need to design alternative education and health, drug use, violence, and others) (World training itineraries for every student is high in Bank 2006). the list of education policy priorities in Jordan and Oman, among other countries (table 9A.1). References 8. Well-known and highly influential national public examination systems such as the Angel-Urdinola, D, A. Kuddo, and A. Semlali. French Baccaluréat, the German Abitur, or 2013. Building Effective Employment Services the British General Certificate of Secondary for Unemployed Youth in the Middle East and Education have evolved in that direction and North Africa. Washington, DC: World Bank. have indeed been used as a tool for imple- Angel-Urdinola, D., and A. Semlali. 2010. “Labor menting curriculum reform in those countries Markets and School-to-Work Transition in (Eckstein and Noah 1996). Egypt: Diagnostics, Constraints, and Policy 9. A young university graduate claimed: “We are Framework.” MPRA Paper 27674, Munich forced into disciplines, it is never a choice.” Personal RePEc Archive, Munich, Germany. And a mother said: “When we have a Tawjihi http://mpra.ub.uni-muenchen.de/27674 student at home, for us it is a ‘horror’ because Angel-Urdinola, D., A. Semlali, and S. Brodmann. this will determine their future.” 2010. “Non-Public Provision of Active Labor 10. In consultation with officials from ministries Market Programs in Arab-Mediterranean of education from a large number of MENA Countries: An Inventory of Youth Programs.” countries, the need to reform the exami- Social Protection Discussion Paper 1005, nation system, such as the Tawjihi or the World Bank, Washington, DC. Baccalauréat, was among the top policy pri- Antonopoulos, R., and K. Kijong. 2011. “Public orities in Iraq, Jordan, Morocco, Tunisia, and Job- Creation Programs: T he E conomic the West Bank and Gaza. The issue has also Benefits of Investing in Social Care: Case been a recurrent focus of policy and public Studies in South Africa and the United States.” attention in Egypt (table 9A.1). Levy Economics Institute Working Paper 11. For instance, the OECD report shows that in 671, Levy Economics Institute, Bard College, Belgium, the Netherlands, Portugal, and Spain, Blithewood, NY. the direct costs of grade repetition account for Autor, D. H., F. Levy, and R. J. Murnane. 2003. nearly 10 percent of the annual spending on “The Skill Content of Recent Technological primary and secondary education. Change: An Empirical Exploration.” Quarterly 12. Formalizing and institutionalizing commu- Journal of Economics 118 (4): 1279–1334. nication and coordination channels between Betcherman, G., K. Olivas, and A. Dar. 2004. education and training institutions and the “Impacts of Active Labor Market Programs: private sector is a key medium- to long- New Evidence from Evaluations with Particular term policy priority, for example, in Jordan, Attention to Developing and Transition Lebanon, Morocco, Saudi Arabia, and the Countries.” Social Protection Discussion Paper Republic of Yemen (table 9A.1). 0402, World Bank, Washington, DC. EDUCATIONAL AND TRAINING SYSTEMS 277 Calmfors, L. 1994. “Active Labor Market Policy Galasso, E., M. Ravallion, and A. Salvia. 2002. and Unemployment: A Framework for the “Assisting the Transition from Workfare to Analysis of Crucial Design Features.” OECD Work: A Randomized Experiment.” Policy Economic Studies 22 (1): 7–47. Research Working Paper 2149, World Bank, CEDEFOP (European Centre for the Development Washington, DC. of Vocational Training). 2009. The Dynamics Greeney, V., and T. Kellaghan. 2008. Assessing of Qualifications: Defining and Renewing National Achievement Levels in Education. Occupational and Educational Standards. Washington, DC: World Bank. Luxembourg and Thessaloniki: European ———. 2009. Using the Results of a National Centre for the Development of Vocational Assessment of Education Achievement . Training. Washington, DC: World Bank. CIDREE/DVO (Consortium of Institutions for Griffi n, P., B. McGaw, and E. Care, eds. 2012. Development and Research in Education Assessment and Teaching of 21st Century in Europe / Department for Educational Skills. New York, NY: Springer. http://atc21s Development). 2008. A Toolkit for the .org/index.php/resources/white-papers. European Citizen: The Implementation ILO (International Labour Office). 2010. The of Ke y C ompete nc es — C halle nges an d Implementation and Impact of National Opportunities. Brussels: CIDREE. Qualifications Frameworks: Report of a Study Cunningham, W., A. Wuermli, and M.-L . in 16 Countries. Geneva: ILO. Sanchez-Puerta. 2010. “Active Labor Market International Program Office for Education and Policies for Youth: A Framework to Guide Training. 2010. Education, Work, Guidance Youth Employment Interventions.” World in Sweden. Stockholm: Euroguidance Sweden. Bank Employment Policy Primer 16. World Jakubowski, M., H. A. Patrinos, E. E. Porta, Bank, Washington, DC. http://siteresources and J. Wisniewski. 2010. “The Impact of the .wo rl d b a n k . o r g / I N T L M / 2145 78 - 1999 Education Reform in Poland.” Policy 1103128720951/22795057/EPPNoteNo16_ Research Working Paper 5263, World Bank, Eng.pdf. Washington, DC. Dar, A., and Z . Tzannatos. 1999. “Active Jensen, R. T. 2010. “Economic Opportunities Labor Market Programs: A Review of the and Gender Differences in Human Capital: Evidence from Evaluations.” Social Protection Experimental Evidence for India.” NBER Di s c u s sion Pap er 99 01, World B a n k , Working Paper 16021, National Bureau of Washington, DC. Economic Research, Cambridge, MA. Del Ninno, C., K. Subbarao, and A. Milazzo. King, K., and R. Palmer. 2010. Planning for 2009. “How to Make Public Works Work: A Technical and Vocational Skills Development: Review of the Experiences.” Social Protection F un d ame nt al s of E duc ation Pl anning . Paper 0905, World Bank, Washington, DC. Paris: International Institute for Educational Eckstein, M., and H. Noah. 1996. Secondary Planning. School Examinations. New Haven, CT: Yale Kluve, J. 2006. “The Effectiveness of European University Press. Active Labor Market Policy.” IZA Discussion El-Ashmawi, A. 2011. “T V ET in Egypt.” Paper 2018, Institute for the Study of Labor, Background paper prepared for the MENA Bonn, Germany. Regional Flagship, World Bank, Washington, Lapan, R. T., N. C. Gysbers, and Y. Sun. 1997. DC. “The Impact of More Fully Implemented Estache A., E. Ianchovichina, R. Bacon, and Guidance Program on the School Experiences I . S a lomon. 2012 . Inf ra st r uc t ure an d of H ig h S chool Students: A Statewide Employment Creation in the Middle East Evaluation Study.” Journal of Counseling and and North Africa. Directions in Development Development 75 (4): 292–302. Series. Washington, DC: World Bank. Leitch, L. 2006. Prosperity for All in the Global ETF (European Training Foundation) and World Economy: World Class Skills. Leitch Review Bank. 2005. Reforming Technical Vocational of Skills. Norwich: United Kingdom: Her Education and Training in the Middle East and Majesty’s Stationery Office. North Africa: Experiences and Challenges. Leney, T. 2009. Qualifications That Count: Luxembourg: Office for Official Publications Strengthening the Recognition of Qualifications of the European Communities. http://www in the Mediterranean Region. Luxembourg and .edis.sk/ekes/TA7005520ENC_002.pdf. Torino: European Training Foundation. 278 JOBS FOR SHARED PROSPERITY MEC-IFIIE (Ministerio de Educación–Instituto 2011. Skills, Not Just Diplomas: Managing de Formación del Profesorado, Investigación Education for Results in Eastern e Innovación Educativa). 2012. Actuaciones Europe and Central Asia . Directions in de éxito en las escuelas europeas. Madrid: Development S eries. Wash i ng ton , DC : Ministerio de Educación. World Bank. M iller-Idriss, C ., and E . Hanauer. 2011. UNDP (United Nations Development Programme). “Transnational Higher Education: Offshore 2010. Egypt Human Development Report. Campuses in the Middle East.” Comparative New York: UNDP. Education 47 (2): 181–207. Wa z z a n , M . , a nd D. Z ov ig h i a n . 2 013. OECD (Organisation for Economic Co-operation “Public Employment Services in Syria.” In and Development). 20 05. Employ ment Building Effective Employment Services for Outlook 2005. Paris: OECD. Unemployed Youth in the Middle East and ———. 2009a. Working Out Change: Systemic North Africa, by D. Angel-Urdinola, A. Kuddo, Innovation in Vocational Education and and A. Semlali. Washington, DC: World Bank. Training. Paris: OECD. World Bank. 2005. Expanding Opportunities ———. 2009b. The Role of Employment and and Building C ompete ncies for Young Labor Markets in the Fight against Poverty. People: A Ne w Age nd a for Second ar y Paris: OECD. Education. Directions in Development Series. ———. 2012. Equity and Quality in Education: Washington, DC: World Bank. Supporting Disadvantaged Students and ———. 2006. World Development Report 2007: Schools. Paris: OECD. Development and the Next Generation . Psilos, P., and G. Gereffi. 2011. “Workforce Washington, DC: World Bank. Development in the Local Economy: Linking ———. 2008. The Road Not Traveled: Education Skills and Capabilities to Upgrading.” In Skills Reform in the Middle East and North Africa. for Upgrading: Workforce Development and Washington, DC: World Bank. Global Value Chains in Developing Countries, ———. 2010. “Turkey’s Active Labor Market ed. G. Gereffi, K. Fernandez-Stark, and P. Psilos. Prog ra m s.” Pol ic y note , World B a n k , Chapel Hill, NC: Center on Globalization, Washington, DC. Governance and Competitiveness, Duke ———. 2011. Europe 2020: Poland: Fueling University. http://www.cggc.duke.edu/pdfs/ Growth and Competitiveness in Poland Skills-for-Upgrading-Workforce-Development- through Employment, Skills and Innovation. and-GVC-in-Developing-Countries_FullBook Washington, DC: World Bank. .pdf. Young, M. 2005. “National Qualifications Salmi, J., and R. M. Bassett. Forthcoming. Frameworks: Their Feasibility for Effective Opportunities for All: A Global Study on Implementation in Developing Countries.” Equity in Tertiary Education. Washington, S k i l l s Wo r k i n g P a p e r 2 2 , I n F o c u s DC: World Bank. Prog ra m me on Sk i l ls , K nowledge a nd Sondergaard, L., and M. Murthi, with D. Abu- Employability, International Labour Offi ce, Ghaida, C . Bodewig, and J. Rutkowski. Geneva. Spotlight 1 An Overview of Policy Options and Their Sequencing The chapters in part 3 (7, 8, and 9) have Table S1.1 summarizes the policy options presented an array of recommended policy discussed in part 3 grouped by main con- responses for opening pathways to more jobs straints in each of the areas. Table S1.2 in MENA. They covered the private sector provides a list of indicators across countries business environment, labor market institu- that could be considered as a proxy of the tions, and education and skills. Although extent of particular constraints in each econ- there are regional trends, ultimately each omy where data are available. Of course, this country needs its own policy response, which table cannot fully demonstrate the complexity will vary with the degree and number of of each constraint, but it might help prioritize constraints. possible policy options. TABLE S1.1 Policy options by context Policy area Constraints Priority policy actions Business Discretion and corruption • Substantive simplification of business regulations to reduce room for administrative environment in business environment discretion and corruption in countries with poor rankings in individual indicators • Make information on regulations, procedures, decision criteria, and anonymized enterprise data fully public • Increase accountability of public administration and sanction abuses High entry barriers and • Reduce discretion of licensing process and reduce costs of entry in product markets through low competition unnecessary licenses • Reduce state monopolies in product markets (for example, regulations prohibiting de jure or de facto private or foreign presence in specific sectors) • Make competition authorities fully autonomous from executive • Simplify regulations for microenterprises to encourage registration and to enable credit access Low access to credit for • Develop private credit registries and reform collateral regimes to include small borrowers firms • Increase bank competition (including through privatizations) • Revise regulations to enable entry of microfinance institutions and nonbank financial institutions to reach microfirms and low-income consumers Labor market Distortionary subsidies • Remove energy subsidies and subsidies to crops with low labor intensity • Develop targeted social safety nets to counter price shocks in low-income households and to assist displaced workers • Provide technical and financial assistance to subsidy-dependent producers and industries to transform production technology Labor market, public • Reform labor code to ease dismissal procedures, costs, and design unemployment sector, and social insurance systems insurance • Develop in-depth assessment of financial flows in social insurance, especially in pension system • Public sector: outsource noncore activities, decentralize decision making, and, most important, align the risks and returns of the wage package with the private sector Skills Skills constraints • Start with basic skills and ability measurement (standardized tests) • Radically reform TVET for higher quality • Reform secondary school exit exams together with university admission • Foster strong link between private sector and education in TVET • Establish quality assurance system, especially in HE Note: TVET= Technical and Vocational Education and Training; HE = higher education. 279 TABLE S1.2 Overview of constraints by country and policy area Ranking on constraints in the business environment Ranking on constraints in the labor market Ranking on skills constraints Ease of Regulation % aged % of % of firms reporting doing quality/ Control of Getting 15+ with loan Energy subsidy employment in Constraining labor skills as a constraint to business implementation corruption credit in past year as a % of GDP public sector market regulations business growth (1) (2) (3) (4) (5) (6) (7) (8) (9) Egypt, Arab High, Moderate, High, Moderate, Very high, 6.9% in 2009 High, High, High, Rep. 110 47th percentile 22nd 78 3.65 27% of (rigid hiring and (ICA: 50.1) percentile employment firing laws) Jordan Moderate, Moderate, Moderate, High and High, 6.0% in 2011 High, Moderate Moderate (ICA: 32.6) 96 57th percentile 50th worsening, 4.47 34% of percentile 150 employment Lebanon High, Moderate, High, Moderate, Low, 0.1% in 2009 Low, 16% of Moderate High (ICA: 55.5) 109 54th percentile 22nd 78 11.27 employment percentile Morocco Moderate, Moderate, Moderate, Moderate, High, 3.9% in 2009 Low, 11% of High Moderate (ICA: 30.9) 94 50th percentile 40th 98 4.3 employment (severance payments: percentile 26 weeks of salary) Syrian Arab High, High, Very high, Very high, Low, — Moderate, 28% Moderate High (61% employers find Republic 134 19th percentile 4th percentile 174 13.4 of employment (considered 8th most skills obstacle to growth; over WGI constraining among 50% have difficulty recruiting 19 by employers ICA) the skills they want—3rd highest) (ICA) Tunisia Low, Moderate, Moderate, Moderate, Very high, 2.5% in 2011 Moderate, Low/High (very Moderate 46 53rd percentile 42nd 98 3.18 22% of flexible for first 3 years percentile employment of contract; very rigid after 4 years) United Arab Low, Moderate–low, Low, Moderate, — — Very high for UAE High for nationals, High for nationals (lack of Emirates 33 62nd percentile 72nd 78 nationals (92% of low for expatriates specialization in private sector percentile employment) required fields) Notes: Rankings: red = very high; yellow = moderate; green = low. (The notes below show the data sources and threshold levels of the business environment, labor market, and skills constraints in MENA countries.) GDP = gross domestic product; ICA = investment climate assessment; — = not available; WGI = World Governance Indicators; UAE = United Arab Emirates. Column 1: Ease of doing business rankings. Source: Doing Business 2012 (World Bank 2011). World average ranking for middle-income countries: 100.9. Thresholds are set as follows: 1–49 = low; 50–100 = moderate; >100 = high. Column 2: Regulation quality indicator. Source: World Governance Indicators (2010 data). Thresholds: bottom 33 percentiles = high; 33–66 percentile = moderate; 67 + percentile = low constraint. Column 3: Control of corruption indicator. Source: World Governance Indicators (2010 data). Thresholds: bottom 33 percentiles = high; 33–66 percentile = moderate; 67 + percentile = low constraint. Column 4: Getting credit ranking. Source: Doing Business 2012 (World Bank 2011). This indicator assesses the institutions, regulations, and financial infrastructure in place to enable credit. Thresholds for constraints are set according to global country rankings: 1–49 = low; 50–100 = moderate; >100 = high. Column 5: Percentage aged 15+ with loan in the past year. Global Findex (2011). World average: 9.05%. Thresholds are set as follows: 0–50% of world average = high constraint; 50–100% of world average = medium constraint; above world average = low constraint. Column 6: Energy subsidy as % of GDP. Source for all countries but Jordan: Silva, Levin, and Morgandi (2013). Data are mostly for 2009. For Jordan, data are from IMF (2012). Thresholds: <=1% = low; <=3% = moderate; >3% = high. Column 7: Percentage of employment in public sector. Lebanon Employer-Employee Survey (2010). Public sector includes only employees in public administration. Thresholds: under 20% = low; 20–25% = moderate; over 25% = high. Column 8: Labor market regulations. Authors’ estimates conducted as part of this study based legislation analysis (see chapter 5). Column 9: Skill constraints. Percentage of firms reporting that skills of workforce are a constraint to their business, from Investment Climate Assessment surveys (2005–10). AN OVERVIEW OF POLICY OPTIONS AND THEIR SEQUENCING 281 References Silva, Joana, Victoria L evin, and Matteo Morgandi. 2013. “Inclusion and Resilience: Global Findex (Global Financial Inclusion The Way Forward for Social Safety Nets in Dat aba s e). World B a n k , Wa sh i ng ton , MENA.” MENA Knowledge and Learning D C . ht t p: //d at atopi c s .worldb a n k .or g / Quick Notes Series No. 71 (September 2012), financialinclusion/. Washington, DC. IMF (International Monetary Fund). 2012. World Bank. 2011. Doing Business 2012: Doing “Jordan: Selected Issues.” IMF Country Business in a More Transparent World. Report 12/120 (May 2012), Washington, DC. Washington, DC: World Bank. I nvest ment Cli mate A ssessment Su r veys. World Governance Indicators (database). World World Bank, Washington, DC . https:// Bank, Washington, DC. http://info.worldbank w w w.w b g i n v e s t m e n t c l i m a t e . o r g / .org/governance/wgi/. research-and-diagnostics/. The Road Ahead PART 4 I know that I am only dust under the steamroller of fate, but I have to do it. It is my duty as a human being. If I don’t speak up, I will stay all my life crouching, kneeling, crooked. I want to raise my head, express myself. — Abdelaziz Belkhodja, The Return of the Elephant 283 The Political Economy of Inclusion 10 Main findings • The political-economy equilibrium in the region has diluted the implementation of technical solutions that could remove the barriers to job creation, many of which have been widely known. • In many cases, a system of rents accruing to ruling coalitions has provided incentives to maintain barriers to entry in the private sector and withhold access to finance. At the same time, the limited space for voice and pluralism and restricted access to data have reinforced the status quo. • Moving towards a more inclusive development model is essential to sustained and quality job creation. • The aftermath of the Arab Spring presents opportunities and challenges. Newly elected governments face pres- sure to deliver results, but might do so with inefficient policies such as scaling up subsidies and public sector jobs without significantly changing the “rules of the game.” • Comparable experiences with regime transition elsewhere indicate that shifting to a model of broader-based growth is feasible. • Several tested approaches exist to produce measurable gains in job creation in the short run and build credibility, without compromising the feasibility of the acutely needed structural reforms. • Investing in data to promote openness and transparency and leveraging the role of new social forces for a more inclusive social dialogue can help build a shared vision for reform. A complex legacy After the colonial period, a rapid metamor- phosis in the region’s political landscape D uring the colonial period (mid- took place. Proto-democracies, military nineteenth to the early twentieth rule, and monarchies old and new prolif- century), France and Great Brit- erated and experienced numerous political ain influenced the diverse political institu- upheavals, including civil and international tions in the Middle East and North Africa wars, foreign invasions, coups d’état , and (MENA), largely through their spheres of revolutions, both peaceful and bloody. influence and the control exercised in their Despite the wide variety of political forms, colonies, mandate states, and protectorates. the longevity of individual political regimes 285 286 JOBS FOR SHARED PROSPERITY in most countries is remarkable. How can incentive to diversify their economies or open such political longevity within countries be up their political systems, they face signifi- reconciled with the diversity of political sys- cant challenges in job creation because of the tems in countries in the region? What are the capital-intensive nature of extractive indus- implications for economic outcomes? tries and other challenges documented in the The framework introduced by North et al. vast literature on resource-rich economies. (2009) studies different types of social order The second group comprises countries that and how these orders affect the appropriation are resource poor. There, elites benefit from and distribution of economic rents (that is, rents derived from their apex positions in special privileges) (see box 10.1). In this inter- organizations that control certain sectors of pretive framework, countries in MENA gen- the economy. These economies are more diver- erally align with the “rentier state model,” as sified than those with natural resources and defined by the literature specific to the region tend to rely on a larger dominant coalition. (Beblawi 1990). A characteristic of rentier states is their Within this model, two typologies of ability to secure consensus by sharing part countries emerge, based on their main source of these rents, especially through subsidies of rents (Luciani 1990). The first group of and public employment. The high degree of countries is characterized by easy access to interventionism required to keep this order rents accruing directly to the state, most com- in place strongly affects the dynamics of the monly from natural resources. As these states private sector. In fact, several studies sug- do not need to collect revenues from taxation, gest that the combination of oil rents, remit- the pressure from citizen-driven accountabil- tances, and foreign aid lessens the pressure ity is attenuated. While these states have little for market reforms.1 BOX 10.1 Social orders and economic management The framework advanced by North, Wallis, and the ability of nonelites to form new organizations Weingast (2009) and extended by North et al. that could compete with elite organizations and (2009) provides an interesting lens through which drive down elite rents. to view political structures around the world. The This model of social order—which North et al. authors map societies to different types of social (2009) term a “limited-access order”—differs from a order, defined as the means by which a society is model in which citizens can form new organizations, organized to maintain coordination and cooperation with little restriction. In these open-access-order among its constituent individuals. Many countries systems, impersonal institutions constrain the gov- solve the problem of coordination (and control of ernment from using the state’s coercive capacity— coercive capacity) by creating a dominant coalition. military or police—illegitimately (that is, to intimidate Members of the coalition (the elites of the society) political opposition parties or to retain power, for receive special privileges or “rents” as a reward for example), and the government seeks control by sup- their membership. These elites normally control porting a broad array of economic and social organi- vast patron-client networks and military, political, zations. To obtain and sustain an open-access order, economic, or social organizations, from which they a large number of individuals must have the right to derive rents. To protect and maintain the value of form political, economic, and social organizations. elite rents, the dominant coalition (usually through The state guarantees legal protection to these orga- state organizations) provides third-party enforce- nizations, as well as the rights of forming new ones ment against any shirking or rebellion in elite- within certain minimum requirements. Moreover, controlled organizations that might adversely affect the institutions that govern the formation of any new elites’ rents. The dominant coalition must also limit organization are guided by impersonal rules. THE POLITICAL ECONOMY OF INCLUSION 287 The Arab Spring may contribute to altering the Syrian Arab Republic, Tunisia, and the this trajectory by opening up political systems, Republic of Yemen), several governments which would enable social and economic sys- embraced policies to increase the role of the tems to become more open and inclusive and private sector and reduce the fi scal losses. experience more economic growth, poverty In North Africa, the fi rst wave of market- reduction, and creative destruction. oriented intervention in the 1990s was Historically, the political-economy equi- intended to attract foreign direct invest- librium in MENA yielded policies that have ment in the energy sector to increase reve- led to several unproductive outcomes: nue generation and forestall the taxation of citizens in a period of fiscal constraint. The • Highly uneven regulatory playing fields second wave corresponded to internation- for fi rms, including discretionary access ally backed structural adjustment programs to credit aimed at reducing government losses from • Significant labor market rigidities state-owned enterprises, mainly through • Access to jobs based on personal networks privatizations. rather than on objective signals of ability According to the (still-emerging) litera- • Generally limited access to information ture on this period, economic liberalization and civil liberties was part of a long-term strategy of power All of these outcomes were—and have preservation. Rulers used economic reforms the potential to remain—significant barri- to consolidate power, reconfigure ruling ers to employment growth in MENA. Their elites, and get buy-in from the upper-middle secondary legacy is an unorganized citizenry classes by allowing access to greater con- that may at first struggle to use new liber- sumption (Dillman 2001; Heydemann 2007; ties, such as those acquired in post–Arab and Kienle 2001). Among others, Kaufmann Spring nations. This chapter reviews the (2011) claims that the gains from privatiza- consequences of these policies from a polit- tion were captured by elites, as observed in ical-economy perspective and examines the Eastern Europe following the collapse of the alternatives for change. Soviet Union. At the same time, MENA gov- ernments proved unable to break away from their traditional sources of political support in Revisiting the political economy the public sector. Unlike their Asian counter- of private sector dynamics parts, MENA economies enjoyed only a lim- This section revisits key features about the ited political space for imposing the reforms interactions between the state and the pri- that could effectively increase private sector– vate sector—the experience of privatiza- led development and reduce rent seeking (see tion, enforcement of regulation, and limited Esfahani 1994, for an example from the Arab access to credit—and explicitly links them Republic of Egypt). Privatization in all coun- to some of the political-economy incentives tries occurred at a very slow pace and in a that prevail in rentier states as well as to the piecemeal fashion owing to its unpopularity types of exclusion and social divides that among workers and among those concerned this model perpetuates. about political stability (Posusney 2003). Many retrospective analyses have argued that one consequence of these reforms and of Elite capture of public benefits the way in which they were implemented was Following decolonization and independence, to reinforce inequities in labor and product most of the countries in MENA adopted the markets. The earlier regime of industrializa- (then common) development model of state- tion allowed governments to provide rents led industrialization. As rents from natural to a broader base of citizens, for instance, resources were depleted in the region’s rela- by subsidizing loss-making enterprises and tively resource-poor countries (for instance, granting monopolies. Instead, the new phase 288 JOBS FOR SHARED PROSPERITY of privatization and overall downsizing of implementation of regulations suggests that the public sector increased the polarization discretionary enforcement has continued of winners and losers in the reform pro- to dominate, limiting the positive effects of cess. As noted in earlier chapters, weak or reforms (see chapter 4). absent competition policies and an unfavor- For example, the discretionary enforce- able investment climate limited the positive ment of complex procedures is an impor- impact of reforms and prevented economic tant up-front fi xed cost for fi rms desiring to restructuring from yielding increased invest- enter a new sector or to grow. It dispropor- ment or the entry of new actors in liberalized tionately affects young and emerging enter- markets. prises while favoring incumbent firms, which Liberalization did have some successes. in turn limits competition and creates rents The history of liberalizing wireless tele- for incumbents. Recent anecdotal evidence com utilities in MENA in the early 2000s offers a glimpse of how regulations were used (described in box 7.5) illustrates the poten- to control rent distribution in Tunisia before tial of liberalization for harnessing invest- the Arab Spring. Cassarino (2011) reported ment and creating jobs. Interviews with that discretionary enforcement of regula- sector experts shed light on the specific politi- tions and overzealous tax enforcement were cal economy of this success story. At the time mechanisms for maintaining “the threshold of reform, mobile telecommunication had a beyond which autonomous private initiatives small consumer base and represented a small could not prosper without the prior approval revenue source for state-owned incumbent of the ruling party.” Interviews with key firms, which lacked sufficient incentives to stakeholders from the labor movement and in invest in this new sector. For this reason, economic think tanks in Egypt and Tunisia, opening up competition faced lower resis- conducted as background for this report, tance than in other sectors. However, to date, confirm this assessment. public telecoms have resisted the liberaliza- Rent seeking is pervasive in democra- tion of international voice communication, cies and nondemocracies alike, and it is not which has historically represented a large unique to MENA. Depending on the mecha- source of their revenue in a market where nisms for accountability, rent seeking can domestic communication has dropped in cost result in different types of economic distor- due to competition from mobile providers. tions (Fisman and Gatti 2006). For example, Experience in other regions has shown that Keefer (2007) argues that rent seeking in in a competitive environment, international China was kept at bay by the institutional- communication would see a dramatic reduc- ization of rules within the Communist Party tion in profit margins for all providers, at the cadres to ensure that elites, while benefit- expense of incumbent firms. ing from rent seeking, were accountable for delivering their end of the bargain, namely, economic growth. Many, including Keefer Enforcement of regulations and (2007), suggest that the lack of accountabil- discretion ity mechanisms among rent-seeking elites Starting in the 1990s, many countries in in MENA led to stagnation in the private MENA have made significant strides in sector in the region (Olsen 2000). Although reforming business regulation. As a result, rent seeking was entrenched, rent-seeking numerous measurable aspects of their elites were never certain that they could business environments have substantially appropriate rents indefi nitely. In response to improved, and some M ENA countries that uncertainty, incumbent fi rms probably have ranked among the top reformers in invested at a relatively lower rate and skewed the world. Improvement in these indica- their investment choices toward activi- tors signals positive changes, but qualitative ties with low value added and fast returns and quantitative evidence on the de facto and away from longer-term investments THE POLITICAL ECONOMY OF INCLUSION 289 with higher returns, such as investments improving credit access could be an endoge- in innovation. Interestingly, some countries nous outcome of the political-economy equi- used export processing zones to capture the librium described earlier. growth potential from export industries driven by foreign direct investment, with- out losing their grip on domestic economies Organized labor: Partners in and and control over domestic rent distribution. challengers of the status quo According to case studies in Jordan and Traditionally, unionization has been low Tunisia, the special policy environment (vir- in MENA. The data and literature suggest tually free of red tape and at times anchored that union representation has been largely in international agreements) in export pro- limited to workers in the public sector and cessing zones greatly facilitated their export in privatized public enterprises, but it has performance, because these zones offered rarely been present in the rest of the private greater stability to foreign investors and sector. Forteza and Rama (2006) estimated allowed exports to thrive. that about 16 percent of the workforce in MENA countries was unionized during Discretionary access to credit 1970–99, a percentage well below that of industrial countries (37 percent) and slightly With the sole exception of Sub-Saharan below Latin America (18 percent). Cherkaoui Africa, firms in MENA, small and large, and Ali (2007) estimated that unions cov- have the lowest access to credit in the world.2 ered about 5 percent of the workforce in Just as important, banks in the region have Morocco. According to survey data, in 2006, highly concentrated portfolios focused on about 22 percent of Egypt’s labor force was large enterprises. For example, countries not unionized; in 2009 in Jordan, 6 percent of part of the Gulf Cooperation Council (GCC) workers in the public sector and 11 percent have the highest ratio of top 20 loan expo- in the private formal sector were unionized. sures to total equity (World Bank 2011). Since independence, labor unions have Standard and Poor data showed that in 2006 been a unique social force within the archi- in Egypt, for example, more than 50.0 per- tecture of Arab political regimes. The rela- cent of loans were held by state-controlled tionship between organized labor and central banks; using data from the Central Bank of governments followed a rather similar pat- Egypt, the World Bank (2006) concluded tern across countries in the region, as orga- that 0.5 percent of total borrowers received nized labor acted as an explicit partner in more than 50.0 percent of total credit given state-led industrialization and modernization to the private sector (see also Keefer 2007). (table 10.1). As an Egyptian scholar and labor Access to credit can be skewed and lim- activist defi ned it during a series of qualita- ited for many reasons, including the lack of tive interviews conducted for this survey, properly developed accounting and auditing systems in fi rms, high rates of tax evasion, The rule of the unions is, usually, to limited development of legislation that would provide a direct and regulated way to allow fi rms to collateralize their assets, and solve the workers’ problems. This was the very limited depth of nontraditional not the case in Egypt, since workers sources of fi nancing, such as fi nancial mar- had no representation in these struc- kets (World Bank 2011). At the same time, tures. Unions were only a bureaucratic the strong concentration of credit could be figure to stabilize the political and eco- seen as an instrument for limiting entry and nomic regime, after the revolution of rent seeking to an elite group, particularly 1952. The way workers used to solve in countries where banks were traditionally their problems was by getting around state owned. Under this interpretation, the [them], or in [the] worst case scenario limited development of policy instruments for by quitting their jobs. 290 JOBS FOR SHARED PROSPERITY TABLE 10.1 Historical structure of unions in Despite these controls, labor unions rep- selected countries in MENA resented the main and sometimes the only Country Structure Affiliation (managed) outlet for popular participation. The history of labor market reforms in Egypt Morocco Three unions Loose party affiliation and Morocco during the decade of liberaliza- Jordan Single union Not affiliated tion in the 1970s illustrates the leading role Egypt, Arab Rep.a Single union Party affiliated Tunisiaa Single union Party affiliated that unions played in shaping labor legislation Algeria Single union Party affiliated and the governments’ reluctance to embrace Syrian Arab Single union Assimilated to single sweeping reforms, even when under pres- Republic party sure from domestic and international inves- Iraqa Single union Assimilated to single tors. During these reforms, unions obtained party important concessions such as legal space for Sources: Gobe 2008; Paczynska 2006; Posusney 1997; Posusney and Cook public expression, the right to strike (granted 2002; Zakaria 2006. Note: MENA = Middle East and North Africa. for the fi rst time), and only mild changes to a. Describes the setup before the revolution in the Arab Republic of Egypt severance conditions for the existing stock and Tunisia and regime change in Iraq. of workers (Paczynska 2007; Zakaria 2006). More recently, unions played an important For instance, in Tunisia, the state- role in Tunisia and Egypt in the months pre- sanctioned tripartite organizations covered ceding the Arab Spring revolutions in 2010 all sectors of the economy: white- and blue- and 2011.3 collar workers were organized through the Union Générale des Travailleurs Tunisiens, the business class through the Union Limits on civil society and access Tunisienne de l’Industrie, and rural work- to information ers through the Union Nationale des Agri- Freedom of association, freedom of speech, culteurs Tunisiens. Similarly, since Nasser’s and other civil liberties are key pillars of time, Egypt had followed a corporatist struc- accountability in a society. Without data ture, exemplified by the fact that the Minis- and information, however, civil society’s try of Manpower also served as head of the ability to hold governments and institutions national trade union’s confederation. Jordan accountable is severely limited. In 2008, maintained a single official union, while MENA ranked second to last in “freedom Morocco’s model stands as unique with three of association” and freedom of the press, official unions, each linked to one of the main and non-GCC countries in the region scored traditional political parties, often competing among the world’s lowest on citizens’ rights for workers’ support (Posusney 2003). of physical integrity4 (figure 10.1). The special role that unions played in the In addition, citizens’ lack of access to rel- state architecture granted them a somewhat evant data could remain a major roadblock more liberal treatment than that reserved for to inclusive development in MENA. Access other civil society organizations, which were is limited partly by explicit legislative provi- often repressed. In many countries, labor sions and partly by limited capacity (which is unions claimed a special legitimacy, because itself a product of a prolonged lack of access their birth often predated independence. Yet, to data). following independence, unions were recon- Table 10.2 illustrates how nearly all coun- stituted and progressively subjected to gov- tries in the region either do not produce or ernment control in one form or another. In rarely share statistical data in a way that can exchange, their members were guaranteed be used for independent analysis or policy what could be seen as a privileged distribu- making. Even data sharing between minis- tion of government rents, through higher sal- tries and the national statistics offices is often aries, nonwage benefits, and protective labor limited to a few specific datasets in most legislation (Gobe 2007). MENA countries. THE POLITICAL ECONOMY OF INCLUSION 291 FIGURE 10.1 Civic freedom indicators for world regions and OECD countries, various years a. Lack of freedom of association, b. Lack of press freedom, regional average, 2008 regional averages, 1993–2008 80 MENA non-GCC 70 MENA GCC 60 South Asia 50 South East Asia 40 30 Sub-Saharan Africa 20 Central Asia 10 World 0 1993 1996 1999 2002 2005 2008 Europe and Central Asia Year Latin America and MENA GCC countries MENA non-GCC countries the Caribbean Europe and Central Asia Central Asia OECD OECD countries South Asia 0 1 2 3 4 5 6 7 8 9 Southeast Asia c. Physical integrity rights index, 1981–2008 1.00 0.90 Physical integrity rights index 0.80 0.70 0.60 0.50 0.40 0.30 0.20 0.10 0 1981 1987 1993 1999 2005 2011 Year Central Asia Europe and Central Asia Latin America and the Caribbean MENA GCC MENA non- GCC OECD South Asia Sub-Saharan Africa Source: Based on CIRI Human Rights Dataset and Reporters without Borders 2012. Note: OECD = Organisation for Economic Co-operation and Development; MENA = Middle East and North Africa; GCC = Gulf Cooperation Council. Physical integrity index is calculated as a two-period moving average. 292 JOBS FOR SHARED PROSPERITY TABLE 10.2 Access to and quality of survey data in MENA Survey data access MENA countries Economies with no survey data for studying living Algeria, Bahrain, Kuwait, Libya, a Saudi Arabia, Oman, and Qatar standards or that conducted surveys but have not released the data even in aggregate form Economies that have systematically collected nationally Egypt, Arab Rep.; Iran, Islamic Rep.; Jordan; Lebanon; representative household data (at least since the 1980s in Morocco; Syrian Arab Republic; Tunisia; and the case of the Arab Republic of Egypt, the Islamic Republic United Arab Emirates of Iran, Jordan, Morocco, and Tunisia). Data access policies vary strongly. Tunisia published the anonymized microdata from its labor force survey online following the revolution. Access to microdata in the Islamic Republic of Iran is officially permitted but in practice is ad hoc. The remaining countries share data only in exceptional cases, or provide only subsets of data. Economies that relatively recently started regular Djibouti; Iraq; West Bank and Gaza; and Yemen, Rep. collection of microdata and grant access relatively freely. Iraq and West Bank and Gaza publish the data and the methodology on the Internet. Source: Scott and Yemtsov 2010. Note: MENA = Middle East and North Africa. a. Libya recently conducted a household budget survey and turned to international agencies for help in the analysis of its results. FIGURE 10.2 Number of publications on labor markets with data opinions about political decisions. In fact, from six world regions in the EconLit Directory, 2010 there is evidence that microdata on MENA are used relatively little in key research areas 400 such as labor markets and social protection. 350 As shown in figure 10.2, MENA has the least 300 Number of hits publications in these areas worldwide (Angel- 250 Urdinola, Hilger, and Ivins 2011). 200 150 100 Consequences of limited data for social 50 dialogue 0 Throug h lack of access to data, many citi- sia cif d be d lA d Af aran Af and Pa an ib an ra an hA ic an sia ca a ric ia ar ca nt pe rth st h zens and social partners in MENA have i Sa r ut As No Ea e C eri Ce ro So b- e Eu st th Am dl been unable to make their voices heard in Su Ea id tin M the public debate and in the related politi- La Source: EconLit, 2010 data. cal decisions that affect their interests. As a Note: Data capture the number of hits for “labor markets” and “region” on Econlit (database of the result, political consensus often rests on the American Economic Association). exclusion of those without a voice. In labor market regulations, for example, Moreover, technical limitations sometimes choices regarding the regulation of work prevent countries from sharing data with hours, types of contracts and their protection, users,5 as the dissemination of data requires wages and their elements, and social insur- proper documentation of the microdata, cat- ance and its contributory base affect many aloguing, platforms, and the technical skills people and need to be discussed and settled in to respond to criticism and questions, which a public and legitimate debate. A well-studied countries lack. In fact, when accurate data, example in the empirical literature is the in raw or capably processed form, rarely dynamics of wage bargaining, in which trade reach citizens, they cannot develop informed unions allow only employed members to vote THE POLITICAL ECONOMY OF INCLUSION 293 on wage agreements (a common practice in of rents and economic benefits will likely con- most countries). The unemployed, who would tinue to pressure governments. The persistent usually benefit more from a wage reduction risk is that these outsiders want to become that could lead to an increase in jobs, can- the system’s new insiders without changing not vote. Consequently, such negotiations the rules that excluded them in the first place: rarely result in stagnating or lower wages, in other words, they seek to become the new even if this outcome would improve the over- elites. all labor market (see, for example, Benassy Prima facie evidence reveals the extent 1995; Gaertner 1981). This division between of these pressures on governments since the insiders (employed, voting union members) beginning of the Arab Spring. For example, and outsiders (unemployed, nonvoting) is several countries have resorted to bulk hiring particularly insidious if the unemployed are of groups of job seekers queuing for public largely first-time entrants into the job market. employment in specific sectors (Morocco, Some youth activists in MENA have already Tunisia) or have created permanent posts for recognized and voiced this particular concern all temporary employees (Egypt). In Tunisia, (Cheikhrouhou 2012). there has been discussion of lowering the standards and moving away from merit as the main criterion for hiring certain categories of Challenges and opportunities public workers. The government in Morocco in the post–Arab Spring recently incorporated a wave of unemployed Chapters 7 through 9 highlighted the impor- graduates, setting aside the traditional entry tance of moving towards broad-based growth examination because of concerns over social as an essential precondition to sustained job unrest. Egypt implemented only the popu- creation. Will the Arab Spring be followed by lar part of a legislated pension reform (an opportunities to untangle the web of privilege increase in the minimum pension) without that has so far hindered this process? An in- addressing less popular elements essen- depth analysis and interpretation of the Arab tial to ensuring fiscal balance. As the Arab Spring are beyond the scope of this work, but Spring unfolded, nearly all governments in the following section documents both the the region offered important wage increases opportunities and the risks inherent in foster- for public sector workers, often coupled—in ing a more inclusive development model. resource-rich countries—with one-off cash Post–Arab Spring governments in MENA transfers to all households or to households face important changes. In some countries, of public employees (table 10.3). Although democratic elections transformed the politi- these measures clearly responded to rising cal landscape. Citizens are embracing new consumer prices, they did not ultimately tar- liberties, forming associations and parties, get the most needy. and developing alliances of parties. At the In the midst of a slump in gross domestic same time, new (and existing) governments product and fiscal revenue (figure 10.3), such are under tremendous pressure to deliver measures are fiscally unsustainable, and con- results in a context of high institutional tinue to preserve rules and incentives that keep uncertainty. For instance, in Tunisia, the current expectations and attitudes in place. number of labor strikes in both public and For example, based on rigorous evidence private enterprises has risen steadily since from a randomized impact evaluation, an the Arab Spring.6 entrepreneurship training program in Tunisia There is a risk that governments will improved participants’ move from wage to respond to this heightened pressure with pop- self-employment but did not change the young ulist measures that alleviate political pressure participants’ aspirations for a public sec- but are costly and do not constitute a step tor job (Premand et al., forthcoming). If the toward more inclusive growth. Those who are terms and conditions of public sector employ- not benefiting from the current redistribution ment remain the same, young graduates will 294 JOBS FOR SHARED PROSPERITY TABLE 10.3 Extraordinary policy measures implemented during the Arab Spring in MENA, 2011–12 Special salary Special financing in increases in public food subsidies/price Bulk expansion of sector controls Special tax cuts One-off Transfers public sector hiring GCC Bahrain Y Y Y Y Y Kuwait Y Y Y Oman Y Y Y Y Saudi Arabia Y Y Y United Arab Emirates Y Qatar Y Developing oil exporters Algeria Y Y Y Iran, Islamic Rep. Y Syrian Arab Republic Y Y Yemen, Rep. Y Y Y Y Y Oil importers Jordan Y Y Y Y Lebanon Y Egypt, Arab Rep. Y Y Y Y Tunisia Y Y Y Y Y Morocco Y Y Y Y Source: World Bank 2012b. Note: MENA = Middle East and North Africa; GCC = Gulf Cooperation Council; Y = yes; [blank] = no. FIGURE 10.3 Growth in GDP forecast for non-GCC countries continue to have an incentive to queue (see in MENA, 2000–13 chapter 2). Similarly, interviews with leaders of the emerging labor movement conducted 10 for this report in Egypt and Tunisia reveal 8 that their members often aspire to a more inclusive but still public sector-centered social 6 contract, with no acknowledgment of the fis- cal unsustainability of this model. Extending 4 public employment would indeed respond to 2 these demands, but it would not reduce seg- mentation in labor markets or promote more 0 inclusive growth. –2 –4 Enabling a broad reform process: Examples of successful –6 transitions –8 Part 2 of the report explained in detail the 2000–08 2009 2010 2011 2012 2013 Year many reforms that would need to take place Algeria Egypt, Arab Rep. Iran, Islamic Rep. to bring countries to a higher-productivity Jordan Lebanon Morocco path. This section highlights reforms from Syrian Arab Republic Tunisia Yemen, Rep. outside the region that have at least in part moved toward a more inclusive growth Source: World Bank 2011. model, with a particular focus on the reform Note: MENA = Middle East and North Africa; GCC = Gulf Cooperation Council; GDP = gross domestic product. process. THE POLITICAL ECONOMY OF INCLUSION 295 Including broad-based change debates around their reform topics. A popu- in the political agenda lar and relevant combination of agenda items is crucial for scoring an election victory. One Prototypical reform processes usually start relatively recent example is the landslide win by bringing the reform topic into public of the Turkish Justice and Development Party debate and onto the political agenda, fol- (AKP) under Tayyip Erdog ˘ an in the early lowed by actions to generate buy-in among 2000s, backed by a broad reformist agenda stakeholders and build a consensus of (box 10.2). approval. Citizen discontent is another frequently This dynamic can be heightened in elec- used channel for bringing items into the toral periods in which political parties must political debate, independent of a country’s compete with different agendas and lead BOX 10.2 Structural reforms under broad-based politics: Turkey’s Justice and Development Party in the 2000s Turkey’s fast and equitable growth performance yet solidarity-oriented rhetoric attracted wide seg- over the past decade is often attributed to the bold ments of Turkish society that had felt excluded by the reforms and prudent policies implemented by the country’s top-down modernization experiment led country’s ruling Justice and Development Party, by secular elites. Building on this popular momen- democratically elected for a third consecutive term tum, AKP leaders emphasized social justice and in a landslide victory in June 2011. The party’s inclusive growth. At the local level, AKP-run munici- success in balancing the requisites of macrostabil- palities implemented social aid programs, at times ity and inclusive growth under its leader, Recep relying on the party’s extensive grassroots organiza- Tayyip Erdog ˘ an, contains important insights for tion. This strategy was complemented at the national governments in MENA, as the gains from bringing level by increased spending on health and education, together diverse collective interests behind a reform- high agricultural subsidies, and improvements in the ist agenda were significant. quality of public services. The traditional support base of the AKP com- A broad electoral coalition also entails policy com- prised the small- and medium-sized entrepreneurs promises, as illustrated by Turkey’s enduring labor in the Anatolian heartland. Yet the party’s com- market challenges. The few changes include a 2003 mitment to macrostability, market-driven develop- law that provided a legal context for temporary con- ment, and the democratization of Turkey also drew tracts, the introduction of a modest unemployment endorsement from the well-organized, Istanbul- insurance scheme, and, more recently, targeted incen- based big businesses. This enhanced support from tives provided to employers’ social security premiums employers was instrumental in consolidating struc- and new programs for vocational training as part of tural reforms initiated in the aftermath of Turkey’s the government’s policy mix to counter the global cri- fi nancial crisis in 2000–01, including particularly sis of 2008–09. On the whole, however, the Erdog ˇ an challenging reforms in banking and public fi nancial government avoided dramatic shifts in labor market management. arrangements that could upset its social partners. As Beyond the big cities, the AKP government has a result, Turkey’s labor regulations remain relatively promoted expansion of the economy’s centers of rigid, with strict employment protections guaranteed dynamism by establishing regional development under high severance payments. Minimum wages are agencies, stepping up regional investment supports, also set consistently above regional averages. In addi- and maintaining close relations with attendant busi- tion, despite a mass privatization drive, the number ness organizations. Benefi ciaries were an array of of public employees increased through new recruits, smaller but rising Anatolian fi rms. and public salaries were raised in recent years. Cur- Particularly relevant in this regard is the AKP’s rently, the government is fi nalizing a new employ- broad public appeal. Erdog ˇ an’s socially conservative ment strategy, which has been put together after wide (continued next page) 296 JOBS FOR SHARED PROSPERITY BOX 10.2 Structural reforms under broad-based politics: Turkey’s Justice and Development Party in the 2000s (continued) consultations with related stakeholders and which women, are primarily the result of structural con- will bring further modifications to the system. straints such as the country’s industrial profile An important lesson to be drawn from the AKP and accelerated rural-urban migration in the past case is that a broad policy base does not neces- decade. Given the delicate balance of collective inter- sarily ensure far-reaching reforms in all areas. It ests upon which the AKP’s electoral coalition rests, does, however, provide a necessary platform for the party opted not to address these issues through social deliberation, from which sustainable solu- radical, top-down reforms. Instead, it chose a more tions might follow. Turkey’s remaining labor market careful approach, involving repeated consultations woes, including low activity rates, a high incidence and a series of policy experiments. of informality, and a low participation rate among Source: Guven 2012. form of governance. By way of illustration, consolidating challenging structural reforms, Turkish citizens experienced growing discon- especially in banking and public financial tent with the patronage politics of the rul- management, following Turkey’s financial ing ANAP Party in the 1980s. The nascent crisis in 2000–01. debate was directed against state economic Building buy-in also requires identifying enterprises, public banks, and public sector and mobilizing segments of the population contracts, and citizens urged the government that would benefit from the proposed reform to make this a part of its reform agenda. but either do not know it or lack the means Further, factors external to a government’s to voice approval. This subject has been well current scope of action can be strong push researched in the context of labor market factors and can shape both public debate reform. For example, data from the Organ- and policy agendas. Many, including Duval isation for Economic Co-operation and and Elmeskov (2005) and Høj et al. (2006), Development show that employees with lim- suggest that big economic crises are associ- ited skills, the unemployed, and youth have ated with higher overall reforms. Crises often higher rates of approval for increasing con- increase the costs of avoiding reform; for tractual flexibility in the labor market, sug- example, unaddressed labor market rigidities gesting that labor market reform would be can slow labor market adjustments to a crisis more likely if outsiders represented the major- (Forteza and Rama 2006). ity of the labor force (see Bentolila, Dolado, and Jimeno 2011; Dolado, Jansen, and Jimeno 2002). This effect can combine with Building stakeholder buy-in the aforementioned external “crisis” effect. based on shared benefits of Saint-Paul (2002) found that reforms occur change at a “tipping point” when unemployment is Building stakeholder buy-in on reform rising, not just when it is high, because the requires creating a strong narrative that employed—the labor market insiders—are in stresses the importance of aggregate social danger of losing their jobs. It is important to gains. To return to the example of Turkey note that these findings apply only when out- in the early 2000s, the AKP managed to siders have the information to understand the package its reformist agenda in a way that consequences of specific policy designs and attracted support from interest groups are equally represented. that did not traditionally support the party Once stakeholders’ buy-in is secured, (box 10.3). That support was crucial to actual approval of the reform package will THE POLITICAL ECONOMY OF INCLUSION 297 BOX 10.3 Lessons from Turkey’s reform in the 1980s During the 1980s, Turkey experimented with a vari- Empowered technocrats ety of reforms, particularly fiscal and trade reforms. In retrospect, the failures of macropolicy in the The variation in economic and political outcomes latter 1980s can be traced to three forces. First, a provides important lessons on what to do and what broad consultative process was not institutional- not to do. ized to cultivate popular support. Second, the top bureaucrats lacked autonomy from the politicians Balance of achievements and hence were unable to counteract pressures that Turkey succeeded in the structural transformation led to expansion of the fi scal defi cit. Third, there of its economy: the share of output for export rose were problems of coordination and confl ict within from 5 percent in 1979 to 23 percent in 1989, and the bureaucracy itself. The highly centralized policy real output roughly doubled. Turkey’s financial apparatus created in the early 1980s proved useful markets opened internationally, in stark contrast in initiating and sustaining reform through its early to the situation before 1980, and have since devel- stages. Yet insularity and lack of institutionalized oped depth and sophistication. Even in areas where links with interest groups increasingly turned out the program was judged a failure as of the end of to be a disadvantage for coordinating policy and 1991—for example, in reducing fi scal deficits, infl a- managing distributional confl icts under conditions tion, income inequality, and the size of ineffi cient of fully competitive politics. Political management public enterprises—the transformation of trade and of a reform process requires building coalitions with finance fundamentally altered the context of the interest groups—developing each part of the reform problems and changed the options available to the package so that it contributes to the overall objec- government. tives of the program and at the same time satisfi es the relevant groups enough for them to support it, Unbroken privileges even if other parts are not to their liking. W hile the reforms of the early 1980s greatly reduced the importance of rent seeking, particularly Speed of reform with regard to foreign trade, patronage politics by A large portion of the reforms were initiated at two the ruling ANAP Party became more widespread critical times when the government had a mandate in the latter half of the decade. As individuals or for dramatic action: the 1980 balance-of-payments fi rms with direct access to the government prolif- crisis and the postdemocratization honeymoon erated, the favors they managed to obtain led to of 1983–84. Making reforms rapidly was impor- growing resentment in the wider populace. Many tant at these junctures. Later in the 1980s, as the people became disaffected by the ANAP’s arbitrary democratic process became more routine, reforms distribution of rents. Traditional sources of patron- succeeded when they were carefully prepared in age politics included state economic enterprises, consultation with the groups affected. Both crash public banks, and public sector contracts. The programs and gradualism were thus appropriate, extrabudgetary funds created new avenues for rent each under different circumstances. redistribution. Source: Adapted from Onis and Webb 1992. depend on how the reform affects the inter- workers, rather than the unemployed or ests of different constituencies. When los- informal workers, who are likely to benefit ers from the reform are more organized and from more flexible labor markets but are not vocal, even reforms that bring aggregate usually organized. gains can be blocked. This outcome often Policy makers adopt a variety of strate- occurs when labor market reforms affect the gies to protect the rights of insiders while benefits of unionized (and therefore vocal) engendering change: (1) packaging reform so 298 JOBS FOR SHARED PROSPERITY that some of its elements satisfy the relevant build the credibility of the new government groups enough for them to go along, even if and mitigate the future cost of reforms. Given other parts of the package are not to their the pressure that governments receive to liking; (2) focusing on strengthening nascent repeat “more of the same,” it is also impor- sectors or areas where incumbents are not tant that these transitory measures also be already entrenched (such as highly innovative compatible with the new midterm agenda. export sectors) to build a new constituency Figure 10.4 illustrates this transition process for change and create jobs without directly toward reform. confronting vested interests; and (3) grandfa- What follows is a nonexhaustive illustra- thering—in other words, applying new rules tion of potential short-term actions that could to new entrants and securing agreement from facilitate the structural reforms highlighted the beneficiaries of the old rules, whose privi- in the previous section: (1) leveraging the leges remain intact. Turkey’s experience in new social forces for change; (2) improving the 1980s illustrates the benefits of packag- access to data and information and freeing ing reforms (trade and macro) to make them restrictions on political dialogue to allow a more politically acceptable. Export subsidies constructive and broad-based debate to take in the form of low-interest credit and tax place; and (3) scoring early wins that generate rebates played an important role in building credibility and mitigate the cost of reforms, a pro-export coalition, which could in turn without compromising midterm objectives. buy support for more conventional instru- These short-term policy actions include both ments such as exchange rate devaluation and general approaches, which are conditions for import liberalization (box 10.3). change in all areas of the jobs agenda dis- cussed in the report, and “low-hanging fruit” for each sector. Figure 10.5 provides a dia- Reforming for broad-based grammatic summary. employment growth in a post–Arab Spring MENA Some of the necessary reforms for making Leveraging new social forces for the private sector, the labor market, and the a renewed and inclusive social skills systems move toward higher job cre- dialogue ation will be costly to many groups in the A new political landscape is shaping the short run, even though such reforms promise region, affecting both established social growth and benefits in the future. Behavioral partners (trade unions, employers, legisla- evidence shows that people strongly prefer tive bodies, and government) and new social avoiding losses to acquiring gains (Tversky partners (civil society, youth, women, the and Kahneman 1991). It is therefore impor- unemployed, and informal workers). These tant to accompany changes to the status quo new groups have not been part of traditional with other measures that facilitate the tran- tripartite negotiations on issues such as sition to a new state: that is, measures that wages and labor market policies. Similarly, the private sector in a number of countries is still unable to organize in independent asso- FIGURE 10.4 Lessons from successful transitions: Steps in ciations and raise funding from members enabling a broad reform process (World Bank 2009). Having all stakeholders at the table will Disrupting Transitioning Consolidating contribute to building a shared view of • Include reforms in the political • Build credibility about • Implement reforms labor markets and barriers to private sector agenda the ability to carry out • Build institutions to • Enable public debate to come to a reforms sustain change development. A number of measures could shared view of solutions • Mitigate the cost of facilitate this inclusive process. Examples reform without compromising the future include reforming regulations on the rights of civil society organizations, workers, and THE POLITICAL ECONOMY OF INCLUSION 299 FIGURE 10.5 Short-term policies and reform objectives Disrupting the low equilibrium Generating credible interventions with early Reform objectives measurable gains • Making access Increase competition, to credit more reduce privilege democratic • Youth employment Leverage new social Rethinking programs, Early forces for a renewed and the social Childhood Rethink the social contract inclusive social dialogue contract Development interventions • Giving voice to service clients Improve governance in skills (like parents systems and students) employers to organize; opening up the dia- This kind of feedback and learning can logue on reforms to nontraditional actors also be provided in joint international sur- (representative employer associations, non- vey and assessment initiatives, such as union representations, and so on); adopting those under way in the education sector new channels for large-scale communica- (the Programme for International Student tion and consultation; engaging organized Assessment, or PISA, and the Trends in Inter- labor and civil society as agents with con- national Mathematics and Science Study, cerns beyond the short term, something or TIMMS). The Russian experience shows that political parties often cannot afford to that “breaking the monopoly” of the statis- do; and building national capacity for pro- tics office can be a driving force in improving viding and analyzing information so that data quality and access (box 10.4). outsiders can offer substantive proposals Without access to trusted data about the and eventually build shared views on how state of the economy, however, the ability of to foster employment. civil society and media to fulfill this function is extremely limited. Scientists need to be able to publish research results using data, with- Improving access to data and out direct or indirect censorship or pressure, information and journalists need to be able to access both Even in the political economy of MENA, statistical data and scientific results, interpret diffi cult reforms can be achieved by using them, and present them to the wider public. data and engaging in a national dialogue to Therefore, it is key that governments, espe- arrive at a shared view of the issues and a cially their statistics offices, allow the public negotiated solution. Social insurance reform to have access to all relevant economic and in Jordan is one rare example (see annex). sociodemographic data, at the anonymized 300 JOBS FOR SHARED PROSPERITY BOX 10.4 Data access in the Russian Federation Over several years, the Russian Federation moved • A policy decision to move to greater transparency from maintaining the complete secrecy of its main and performance-based budgeting was initiated household budget survey (HBS) toward posting sur- by the Ministry of Finance in the late 1990s and vey data on the Internet in a downloadable form. accelerated by the fi nancial crisis of 1998–99. The The process started in 2001, when increased pub- Ministry of Finance made “open access” a key lic attention to poverty reduction and monitoring dimension of performance indicators for Rostat. forced Russia’s statistical agency (Rostat) to engage • A parallel advocacy campaign targeted Rostat with researchers and international donors to review officials, making them aware of international its practices and procedures. By 2007, micro data- practices and professional standards. The World sets from several years of HBS were available free Bank supported Rostat’s exposure to interna- from Rostat’s website (http://www.gks.ru). Several tional standards through study tours of data internal factors combined to produce this change: access practices and technical assistance for pre- paring anonymized microdata ready for dissemi- • A decade-long effort by Russian researchers (sup- nation, which reduced the cost of public access to ported by donors) to collect public-access house- Rostat data. hold survey data challenged Rostat’s monopoly and helped build a critical mass of expertise out- Source: Programmatic poverty assessment for Russia, cited in Scott and Yemtsov 2010. side the main government agency. individual level.7 A number of approaches is critical to providing feedback and improv- can help, as discussed below. ing the quality of data. Statisticians inside Limits on researchers’ access to ano- the office and in academia need to work on nymized individual records can be reduced, the data to produce results and feedback for along with contradictions within the law future surveys. When data are not accessible or between the basic statistical law and any and their use is limited, statistical offices do derived decrees. Removing procedures that not benefit from the feedback from data users introduce a high degree of discretion in and, in turn, have little or no incentive to granting access to data, such as requiring improve measurement, precision, and survey personal authorization from the head of the design. statistics office or even ministerial approval, In this context, labor market observa- could eliminate bottlenecks. Tunisia is a tories (LMOs), especially when indepen- recent successful example of overcoming dent from the government, can provide an legal and other barriers to data access. A June external quality check and a trusted source 2011 decree on access to information clearly of analysis and information (Rutkowski indicates that anonymized individual data 2011). As issues of trust cannot be ruled must be made available to the public by the out where data are concerned, appointing national statistics office before June 2013. an independent institution to produce the To support evidence-based policy mak- relevant labor market statistics can be ben- ing, statistics offices need the capacity to eficial to all. LMOs can improve the qual- carry out large-scale household surveys to ity of labor market information and enable provide data that are comparable over time, labor market actors to make more informed are representative at a sufficiently disaggre- choices. The role of an LMO is usually to gated level, and meet other standards. The use existing data more efficiently, to fill data exchange with experts and outside data users gaps through new special surveys, and also THE POLITICAL ECONOMY OF INCLUSION 301 to help articulate evidence of employers’ unemployment. This arrangement also makes needs. Using the new and better data, policy it possible for LMOs to evaluate the perfor- makers can then tailor policies more closely mance of the employment office. Selecting to regional labor market conditions, and LMOs through competitive bidding can also employment offi ces can match job seekers ensure that the best capacity is mobilized. and vacancies more efficiently. Educational Box 10.5 illustrates the successful use of institutions can adjust curricula in response LMOs in Poland. to changes in occupational demand, and As the example of the LMO in box 10.5 students and job seekers can invest in skills shows, the quality and reach of dissemina- that are in demand. tion are the keys to useful data. Dissemina- Some specific tools can improve the likeli- tion should include existing surveys (survey hood that LMOs will succeed and mitigate summary reports) and administrative data, the risks of failure. Reporting arrange- such as monitoring data on employment pro- ments are important. If LMOs report to grams and vocational training. It should also regional authorities and not to the regional include international benchmark surveys, labor office, they naturally have a broader such as PISA, TIMMS, the Progress in Inter- focus on the labor market rather than just national Reading Literacy Study, and the BOX 10.5 Labor market observatories in Poland Poland established labor market observatories to TABLE B10.5.1 Main differences between the two improve the understanding of labor market per- types of labor market observatory in Poland formance and to produce much-needed informa- tion on regional and local labor market conditions Bottom-up initiative Top-down initiative and trends. This information was also intended to Regular activity of RLO Temporary project (e.g., for inform the allocation of budgets in local labor offices 3 years) and employment programs. The country established Financed initially (mid-2000s) European Structural Funds 16 regional LMOs, 14 in a top-down procedure by the by RLO, then ESF Ministry of Regional Development using European Objectives defined by RLO: Objectives defined by the Union funds, and 2 in a demand-driven, bottom-up Development of labor market project: Output indicators (e.g., initiative of regional labor offices (RLOs). All LMOs diagnostic capacity of RLO number of studies/workshops/ operate under the RLO and are not independent (see publications) table B10.5.1). Labor market analysis done Analysis and surveys contracted LMOs typically use a rich array of data to moni- in-house by LMO staff; out; LMO acts as a management surveys contracted out to unit tor the labor market: (1) administrative data from research institutions local labor offices on unemployment, vacancies, Research staff: 4–5 people Administrative staff: 5–10 and active labor market programs (expenditures Administrative staff: 5–10 and participants); (2) data provided by the national Budget: a 0.2 m per year Project budget of statistics office, broken down by region, such as a 0.7 m– a1.0 m for 3 years labor force surveys and employment and wage data; and (3) data from special-topic surveys (usually Source: Rutkowski 2011. Note: LMO = labor market observatory; RLO = regional labor offices; “sociological”). Currently, administrative data are ESF = European Structural Fund; m = million. underutilized compared to survey data. Data collec- tion and monitoring of regional labor market condi- which include regular monitoring of both regional tions can be done in house or contracted out. and local labor market conditions (such as trends in LMOs typically publish a variety of information: unemployment, the structure of unemployment, and monthly bulletins (in paper and electronic versions), trends in job vacancies) but no analysis; newsletters (continued next page) 302 JOBS FOR SHARED PROSPERITY BOX 10.5 Labor market observatories in Poland (continued) (monthly or quarterly), which report on activities conditions. Local authorities fi nd the publications and signal upcoming events (workshops, confer- useful but too complex to be effectively processed ences); and reports summarizing the results of spe- and used for policy purposes, and they consider cial surveys. LMOs can also offer workshops and them only loosely linked to the operation of the conferences for stakeholders on study design and local labor office. Educational institutions fi nd the results. Some LMOs issue biannual “labor market data useful but too complex and not always rele- mismatch” reports, which assess the supply of and vant. Investors and employers are currently not very demand for labor by occupation. One LMO has an aware of the LMOs’ work. Those who do use the “occupational barometer,” which is an assessment data report favorably. of labor shortages and surpluses by occupation done In Poland, the impact of LMOs on the labor by local labor offices. LMOs disseminate their infor- market has been very positive. LMOs have become mation on websites; in conferences for local experts, a platform for exchanging ideas, networking, and policy makers, and other stakeholders; and through improving cooperation and coordination among the local media. Reports are sent to heads of regional stakeholders. Owing to improvements in the pro- governments, local labor offices, local education duction and dissemination of information, employ- supervision bodies, and employer associations. They ment policies are more closely aligned with the needs are also sent to other LMOs to share experience. of local labor markets, public employment services The actual use and appreciation of LMO data are more responsive to the needs of employers, and vary by type of user. Regional authorities fi nd the investors and educational and training institutions data a valuable diagnostic tool that informs pol- are able to respond more effectively to local labor icy and raises awareness of regional labor market markets. International Adult Literacy Survey, which Generating credible are vital to signaling which skills are needed interventions with early in the population at different ages. For the measurable gains efficient use of data on individuals, research- ers need an archive of household survey data The evidence shows that politicians in with clear access criteria. young democracies appear less able to make Dissemination of data can extend well credible promises to broad groups of citizens: beyond national boundaries, just as interna- citizens are not sure if the new, inexperi- tional educational benchmarks have spread. enced government can actually deliver on its A regional data exchange can help unify promise; they are unable to hold the govern- standards and encourage peer review for ment accountable if it reneges on a reform better quality. At a more advanced stage, promise; they are not sure if the government countries can also agree on standards for will gain from reneging at all; and those that data sharing and dissemination and develop benefit from a reform (winners) might not be a monitoring mechanism for data access sure of the power of the losers and whether in the region, with periodic monitoring the losers’ influence might sway the govern- reports. The Program for the Improvement ment (Keefer 2007; Myerson 2009). of Surveys and the Measurement of Liv- Under these circumstances, early and mea- ing Conditions in Latin America and the surable gains could signal a young govern- Caribbean (MECOVI) initiative developed ment’s ability to deliver results and follow in Latin America in the late 1990s is one through with plans, despite a weak account- example (box 10.6). ability structure or vocal opposition. At the THE POLITICAL ECONOMY OF INCLUSION 303 BOX 10.6 Improving statistical capacity through regional peer learning: The MECOVI project in Latin America In 1996, the World Bank, the Inter-American Devel- learn about developing policy-relevant surveys and opment Bank, and the United Nations Economic basic analytical techniques. Finally, a database of all Commission for Latin America and the Caribbean surveys available in the region was created and made jointly developed a program called the Improve- accessible to researchers and policy makers. ment of the Surveys of Living Conditions (ISLC/ When the program started, a key goal was to MECOVI). The program was a response to the per- increase the access to data. Countries participating ceived poor quality typical of household surveys in the full program (4 originally, later expanded to produced in the region. The capacity to produce 10) had to agree to a policy of public access to data- policy-relevant data was limited, and datasets were sets. This position was very controversial in Latin rarely disseminated: statistical abstracts were usu- America at the time. Interestingly, if one looks for ally all that was produced. data on Latin America today, many statistical offices The project sought to engage countries at several offer the datasets on their websites and, instead of levels. A limited set of countries received extensive fighting users, are actively engaged with the research support in a multiyear program to develop, imple- and policy-making community to improve the data ment, analyze, and disseminate surveys relevant to and increase their use. While the message is still not the policy discussions of the time and, most impor- universally accepted, the combination of provid- tant, to welfare. A second and parallel effort sup- ing incentives, of the demonstration effect (Peru’s ported regional workshops for all statistical offices national statistical office made data publicly avail- on specific topics of interest (poverty lines, measure- able and benefited from doing so), of the interplay ment of consumption and income, and the advan- between peers and counterparts, and of the long- tages of each in welfare analysis, for example). term investment in increasing capacity in both sur- Third, two-week training courses were offered twice veys and analysis has paid off. a year to bring data producers together with users to Source: Scott and Yemtsov 2010. same time, governments must find a way • Expansions of access to fi nance to generate these early wins without com- • Changes that open up public services to promising their capacity to undertake more citizens’ participation significant structural reforms in the medium While the technical aspects of the broader term. Both within and outside MENA, gov- reforms are discussed in more detail in part 2 ernments have made a number of innovative of this report, the main lines of these inter- interventions that have relied on short-term, ventions are summarized below. measurable gains and addressed the strong demand for inclusion, democratization, and accountability without compromising fis- Improving opportunities and cal sustainability in the longer run. These employability through youth programs interventions are compatible with a fiscally New policy makers can choose from a rich constrained environment such as the current array of tried and tested employment policies one, and they target priority concerns for the that target youth. These policies do not typi- region. These include: cally entail a political cost, and several are • Reforms that facilitate women’s transition both sustainable and lay a good foundation into the labor force and that improve the for more difficult reforms later on. Examples employability of young people of such policies include subsidizing the wages • Interventions that fi ll infrastructure gaps or social security contributions for women with a view toward job creation and young people; providing well-designed 304 JOBS FOR SHARED PROSPERITY active labor market programs; combining be 9 million additional jobs in the course of labor-intensive public works with training 10 years. in job-specific skills for low-skilled workers; investing in early childhood development and Democratizing access to finance home-based work; and helping youth cre- ate their own jobs through entrepreneurship Given the severe restrictions on borrowing training and coaching. To be demonstrably that many fi rms face, improving access to effective, these programs should have clear credit can be an important step toward less- targets and be rigorously evaluated; and ening barriers to entry for fi rms and increas- information on their outcomes should be ing employment, productivity, and general accessible and broadly disseminated. business development. The 2011 financial sector flagship report on MENA (World Bank 2011) laid out a clear road map for Facilitating female participation achieving the ultimate goal of democratiz- in the labor force ing access to credit. Although some of these E a rly ch i ld hood prog ra m s t a rget i ng interventions will take time, others could 0 –3-year-olds are almost nonexistent in have an immediate effect on expanding MENA, even though subsidized child care access to finance for young entrepreneurs and preschool education represent a high- and small enterprises: (1) enacting laws to return public investment. A developed child simplify borrowing and collateralization care education industry could directly cre- methods for small amounts; (2) permit- ate a large number of jobs for women, while ting lenders to lend on the basis of personal also allowing many more parents (young evaluations and projected income streams; mothers) to be economically active. In addi- (3) permitting lending institutions to take tion, international evidence suggests that deposits, if they can meet requirements that high-quality early childhood education and are practical as well as prudential; (4) sup- care help prepare young children to succeed porting capacity building and product devel- in school and eventually in life. For exam- opment for microfi nance; and (5) ensuring ple, they are associated with lower rates of that financial services effectively reach rural grade repetition and lower dropout rates communities (WIEGO Network and World throughout a student’s lifetime (Calahan Bank 2008). and Tarr-Whelan 2006). Opening up public services to citizens’ Filling important infrastructure gaps participation Investment in infrastructure could be a Improving the quality of social services is a natural “early gains” measure to create jobs visible intervention that can rapidly estab- in MENA because of its high labor content lish credibility for governments. When and its strategic role in connecting people citizens can monitor the delivery of local to markets. The region could generate 2.0 services, the performance of service pro- million direct jobs and 2.5 million infra- viders improves, and citizens’ demand for structure-related jobs just by fi lling current greater inclusion is met in a highly visible domestic infrastructure gaps, according to way (Reinnikka and Svensson 2003). For a recent study (Estache et al. 2012). Apart instance, countries in other regions are from quick wins, the long-term employ- allowing schools greater autonomy and ment effect of infrastructure investment in are encouraging stronger local scrutiny MENA could be significant: the employ- from parents through the empowerment ment response induced by infrastructure of parents’ councils. In El Salvador, for investment resulting in one percentage example, giving parent committees a moni- point of additional growth is expected to toring role and a budget led to measurable THE POLITICAL ECONOMY OF INCLUSION 305 BOX 10.7 Improving learning through parents’ participation: Strengthening school councils in Morocco A principal objective of the Programme d’Urgence of would include regular discussion forums, moderated the Moroccan Ministry of National Education is to by a local nongovernmental organization, in a fi rst improve the quality of school life. Such improvement test group of colleges. The forums would formulate will require, among other things, reinforcement of action plans to improve learning quality, and the supervision and monitoring and evaluation, especially CdG would be in charge of implementing the plans. of teachers. Past reforms in this area did not achieve Discussion themes could include, for example, pupil the intended objectives. For example, for 2008, the and teacher absenteeism, curriculum adherence, Conseil Superieur d’Enseignement reported 80,200 or infrastructure. The pilot also included a second days of unexcused teacher absences in lower secondary test group, which would benefit from the forums in schools and 45,700 days in primary schools. A World addition to the legal accreditation of the CdG as an Bank study in 2005–06 arrived at similar conclusions. association and a fi nancial allocation to realize the A weak monitoring and incentive system for teachers action plan. A random sorting of colleges into one did not help the situation. Efforts by school inspectors of the two test groups or into a control group would had limited effectiveness owing to geographic barri- enable a rigorous impact evaluation. ers and scarce resources, such as cars. The Conseils In implementing the Plan d’Urgence, the ministry de Gestion (CdG)—that is, the official school manage- set forth several objectives for the pilot: ment committees that include teachers, parents, and • Discover mechanisms that offer effective incen- local stakeholders—were generally weak, and local tives for teachers. The pilot would test the extent stakeholders had little participation in school life. to which the interaction with the local community New and innovative measures were needed to real- could provide incentives through recognition and ize the Plan d’Urgence. The ministry therefore decided criticism of the teachers. to explore the possibility of strengthening the CdG to • Prepare for decentralization in school administra- serve as local guarantors of education quality. This tion. The pilot would examine the extent to which action was inspired by international experience from the CdG could constitute a decentralized civic El Salvador, which saw measurable improvements in body of quality assurance. learning achievement when school committees were • Improve the quality of school life. The quality empowered with a monitoring role and a budget. forums under the leadership of the CdG would A pilot experiment tested two ways of strength- address all “quality” issues facing local schools. ening the CdG to determine which would yield the best results in learning and in pupil and teacher The CdG pilot has been planned but has not yet attendance. The ministry began planning a ran- been implemented. domized pilot among lower-secondary schools that Source: Grun 2009. improvements in children’s learning (Vegas and strategic location near mature markets, 2005). A pilot project inspired by this expe- countries in the Middle East and North rience has started in Morocco (box 10.7). Africa have tremendous potential. How- Publication of data on performance and ever, high and persistent rates of jobless- resource flows and the use of citizen score- ness and low-productivity informal employ- cards for evaluating delivery are additional ment, along with a formal sector dominated methods for increasing accountability. by public employment, are clear indicators that the region is not making the most of its endowments. The great promise these coun- Time for action in the Middle tries hold has yet to be realized. East and North Africa A comparative analysis of the region’s With their natural resource wealth, young employment dynamics suggests that the and increasingly well-educated populations, development model adopted by many 306 JOBS FOR SHARED PROSPERITY countries in MENA has failed to deliver The Arab Spring disrupted the political enough good quality jobs—the kind of jobs equilibrium in the region, and many coun- that would have met the aspirations of the tries are undergoing important and complex region’s large youth cohorts. While the pub- transitions. This presents a unique opportu- lic sector played a key role in generating nity to break this system of privileges and employment and raising living standards move towards a new and more inclusive during the fi rst decades after independence, model of development. However, it also car- demographics and global economic trans- ries challenges. In particular, the new gov- formations now make a dynamic private ernments need to walk the fi ne line between sector the indispensable factor for sustained responding to political pressure of differ- job creation. This report has shown in ent political groups—with risk of engaging detail how the incentives for formal private once again in populist and fiscally costly sector job creation, productivity growth policies without changing the rules of the in fi rms, skills formation, and the efficient game—and building ownership for reform. allocation of resources and human talent All of this in a context of high institutional are stifled in MENA. Not only has this uncertainty. curbed the dynamism of the economy, but Many structural reforms will be needed it has also produced inequitable outcomes to unburden countries from the lega- that have been especially detrimental to cies of the past and promote job creation. women, young people, and the poor. The These include creating a level-playing field public sector and the state do have a vital among firms, establishing a fair de facto enabling role to play, but this is a very dif- investment climate, extending access to ferent role from that seen in the majority of credit, rebalancing employment conditions the region’s countries at present. between the private and the public sec- Part of the underperformance in MENA tor, improving governance in educational is grounded in the complex political econ- and training systems and removing energy omy of the region, which also explains the subsidies while protecting the poor and inability of countries to embrace the reforms vulnerable. that could really change their development The time for action is now. Global expe- trajectory. In the years following inde- rience has shown that it is possible to pro- pendence, ruling elites consolidated their duce tangible improvements in the lives of dominant position in society with a mix of citizens in the short run without resorting limiting civil liberties on the one hand, and to populist policies that might be harmful implementing populist policies on the other. to the economy in the longer run. Improved The latter involved widespread fuel and food social service delivery, short-term employ- subsidies, and guaranteed public employ- ment and skills development programs with ment. During this time, elites were able to well-designed temporary interventions for adopt economic strategies, such as enforcing youth and the poor, and well-functioning policies only when it suited their aims, limit- social safety nets are but a few of the policies ing openness to trade, and restricting access that can deliver early, visible, and measur- to credit and competition, which helped able results. They can help new governments them secure large shares of the countries’ gain credibility, while a more open dia- economic rents. More importantly, this rent- logue with the population, strengthened by seeking strategy held back the broad-based, open access to public data and information, inclusive growth process that is necessary to builds consensus for the needed, longer-term promote innovation and move the economy structural reforms. towards higher value added production. And if not now, when? THE POLITICAL ECONOMY OF INCLUSION 307 Annex Using data and dialogue The Social Security Corporation (SSC) to tackle a politically difficult opted for raising public awareness about reform: Social insurance in the sustainability of the status quo through Jordan a public awareness campaign, the National Dialogue on Reform. In early 2007, the SSC An unsustainable social insurance shared the results of its actuarial valuations system with its main stakeholders, including the gov- The Jordanian pension scheme used to offer a ernment at the highest policy-making levels very generous package, with early retirement and the tripartite structure of government, at the age of 45, given a working period of employers, and employees. 15 years for women and 18 years for men. While the “need to do something” went Over time, more people opted for early retire- down well among some (mostly in govern- ment, with almost 80 percent of retirees ment), it did not go down well with those choosing that option by 2008. In addition, whose constituencies stood to lose—includ- there was no cap on pensionable salaries, ing contributors hoping to at least maintain allowing the rich and those who had some their current level of benefits, if not increase control over their stated salaries to earn them. Organized trade unions in particular extremely high returns on their contribu- actively resisted any attempt to tamper with tions with large increases in the last few years the law. They questioned the credibility of before retirement. As a result, lower-income actuarial reports, claiming that there was no contributors and future generations were sub- problem to begin with or, if there was one, sidizing the better-off. it was the fault of government and govern- Actuarial studies dating back to the ment should pitch in with the funds to fi x it. 1990s identified these features as risks to It was clear that regardless of whether they the sustainability of social security. By understood the problem, many union orga- 2026, Jordan’s social security would have nizers saw themselves as acting in the best had to start liquidating the reserve funds interest of their constituencies, which meant to pay benefits, and by 2036, it would have resisting any change to the status quo. The exhausted all its reserves and would be in common man or woman was more likely to deficit. be swayed by a mistrust of government and question the motives behind reform. The SSC was on the defensive. A politically challenging reform agenda The National Agenda of 2004 envisioned A well-crafted public awareness reforming social security by curtailing these campaign unsustainable benefits; yet there was little appetite for reform, especially for early At that point, the SSC decided to invest in a retirement, and little consensus on how this well-articulated public awareness campaign. change could be accomplished. More impor- The aim was to change the terms of the debate tant, the public had little or no understand- by attempting to get the messages across in ing of why reform was needed. Although not simpler ways, to more people, and to cover unique to Jordan, lack of trust in the govern- more issues than just early retirement. The ment made reform difficult. Reform of social campaign was directed at universities, politi- security could easily trigger suspicions that cal parties, parliamentary blocs, municipal the government wanted to cover up mis- town halls, women’s groups, youth and young management problems or, worse, to signal a entrepreneurs’ associations, pensioners, desire to use social security funds. nongovernmental organizations, and, not 308 JOBS FOR SHARED PROSPERITY least, the press. Methods to convey the mes- The awareness campaign evolved over the sage ranged from articles in the newspapers first year into a National Dialogue Campaign to TV interviews, public debates, and short over social security reform. The SSC had to documentaries. Many of the results were put commit itself up-front to clear policy goals as in graphs and figures for policy makers and well as a clear process for articulating them. academics. Only as the goals were spelled out and the To address the issue of “credibility,” the SSC made itself accountable did the SSC SSC offered Parliament the opportunity improve its credibility. to identify any licensed actuary, firm, or individual to review the actuarial report, Result: A new social security law at the SSC’s expense. While Parliament never took advantage of that offer, the offer Through this strategy, the SSC was able to itself helped bolster the credibility of the secure consensus for the reform. The main report. features of the new social security law In addition, to address arguments about included: “assumptions” about future demographic • Adjusting early retirement factors such as and behavioral factors, the SSC showed, in minimum age, minimum working period, graphic terms, stylized examples in which and penalties for early retirement so that individual workers contributed for the the incentives were to work longer minimum vesting period required and then • Capping wages eligible for social security decided to retire and how much they would and capping new contributors at five times receive in benefits if, say, they lived an aver- the average wage age lifespan. The results invariably showed • Introducing new protection schemes, all that people would make more than they put of which were designed to be self-sus- in and potentially much more if they gamed taining, namely, a maternity insurance the system. scheme (highly supported by women’s To explain seemingly complex issues of organizations); an unemployment sav- financial sustainability in an intergenera- ing scheme (introduced to allay fears of tional context, the SSC produced a simplified those who used to see early retirement as animation to show how, in the initial stages, an insurance against job loss); provisions contributions made by an expanding pool for a health insurance scheme (actively of employees far exceeded benefits paid to demanded by both contributors and the few retirees. Then the animation showed pensioners, not yet activated); and lim- that, as the system matured over time, the ited indexation of pensions to inflation ratio of pensioners to workers grew, and (demanded by pensioners, not financed illustrated the effects of increasing ben- independently, but contained through efit payments and decreasing contributions. limits) Next it showed the available pension funds • Reforming the management of both liabil- gradually diminishing and the underlying ities and assets assets disappearing one by one, ending with a big question mark around social security. Finally, in an attempt to convey the nature Reform winners and losers of a defi ned benefit pay-as-you-go system, the metaphor of a Jordanian olive tree was Winners—youth, women, and workers in used. The olive tree in Jordan can last hun- small enterprises. The social security reform dreds of years when each generation acts as package would yield clear winners by design. a true custodian. Alternatively, olive trees These were youth, who were guaranteed a can be cut down after just a few decades by sustainable pension system; women, who a generation eager not just to enjoy the fruit saw many of their grievances with the old of the tree but its wood as well. law addressed; workers in small enterprises, THE POLITICAL ECONOMY OF INCLUSION 309 who wanted their coverage to become man- Details on its construction and use can be datory; and pensioners, who also had their found in Cingranelli and Richards (1999). pensions partially linked to inflation. 5. For example, statistical agencies in Egypt, Potential losers—early retirees, the rich. Lebanon, and Qatar maintain their sur- vey and census microdata in Oracle/SQL The potential losers were in two distinct databases. Because they do not use SPSS, groups: those who would lose the abil- Stata, or equivalent user-friendly software for ity to retire early and those whose salaries data analysis, their datasets cannot be dis- were above twice the national average (and seminated readily. mainly exceeded JD 5,000). Potential losers 6. In February 2012 alone, 53 strikes were included some of the loudest voices in the counted (African Manager 2012). country: organized labor and highly paid 7. Note that concerns about anonymization are executives. often not warranted. Datasets may contain Compensation. While it was not possible two types of identifiers: (1) direct identifiers to appease those groups, it was possible to (such as name, telephone number, address, address some of their demands. First, all of email address, social security number, and GIS coordinates); and (2) indirect identifiers (such those who had accumulated the minimum as date of birth or age, sex, profession, income, number of years to qualify for early retire- and village or city of residence). A light ano- ment by the end of 2010 were grandfathered nymization procedure can quickly remove into the old law. Second, a special provision the direct identifiers. Indirect identifiers are a was put in place to allow early retirement for much more complex issue. In developed coun- those working in hazardous jobs through a tries, a dataset that contains the date of birth, higher employer contribution (this provision sex, and postcode of respondents makes it pos- was very important for unions). Third, a sible to identify a very high proportion of them defined benefit scheme was made possible for simply by matching this file with other pub- those making more than the allowed pension- licly available data files (such as a list of vot- able wages (this provision met the needs of ers in the United States). In MENA today, no publicly accessible files can be used to match executives who wanted firms to maintain the identities, so a light anonymization procedure contribution on the full amount of their sala- would suffice. ries). Together, these measures represented various forms of compensation that reduced the opposition of the losers, even though they References all would have preferred the status quo to the African Manager. 2012. “Tunisie: Février reform. le mois de toute s le s g rève s.” ht t p: // Source: This annex summarizes findings from Razzaz 2011. w w w. a f r i c a n m a n a g e r. c o m / 1 3 8 61 9. html?pmv_nid=1. Angel-Urdinola, D., A. Hilger, and I. B. Ivins. 2011. “Enhancing Access to Micro-Data in the Notes Middle East and North Africa.” Arab World 1. See, for example, Diwan and Squire (1993), Brief 4, World Bank, Washington, DC. Hoekman (1995), Shafik (1995), and López- Beblawi, H. 1990. “The Rentier State in the Arab Claros and Schwab (2005). World.” In The Arab State, ed. G. Luciani, 2. See chapter 4 and figure 4.1 (World Bank 85–98. Berkely, CA: University of California 2006, 97). Press. 3. See, for instance, National Lawyers Guild Benassy, J.-P. 1995. “Nominal Rigidities in Wage (2011); and, for Egypt, Solidarity Center (2010). Setting by Rational Trade Unions.” Economic 4. The Physical Integrity Rights Index is an addi- Journal 105 (430): 635–43. tive index constructed from the torture, extra- Bentolila, S., J. Dolado, and J. Jimeno. 2011. judicial killing, political imprisonment, and “Reforming an I nsider- Outsider Labor disappearance indicators. It ranges from 0 (no Market: The Spanish Experience.” CEPR government respect for these four rights) to 8 Discussion Paper 8691, Centre for Economic (full government respect for these four rights). Policy Research, London. 310 JOBS FOR SHARED PROSPERITY Calahan, L., and L. Tarr-Whelan. 2006. “Early Fisman, R., and R. Gatti. 2006. “Bargaining for Childhood: E ducation for A ll, a Wise Bribes: The Role of Institutions.” In Economic Investment.” MIT (Massachusetts Institute of Handbook of Corruption , ed. S. Rose- Technology) Workplace Center, Cambridge, Ackerman. Northampton, United Kingdom: MA. Edward Elgar. Cassarino, J. P. 2011. “Confidence-Building in Forteza, A., and M. Rama. 2006. “Labor Market Tunisia after the Popular Uprising: Strategies ‘Rigidity’ and the Success of Economic and Dilemmas of the Interim Government.” Reforms across More than 100 Countries.” Istituto Affari Internazionali Working Papers Journal of Policy Reform 9: 75–105. 11(04), Istituto Affari Internazionali, Rome. Gaertner, M. 1981. “A Politicoeconomic Model Cheikhrouhou, A. 2012. “Le NON a tout, of Wage Inflation.” De Economist 129 (2): et tout de suite” [NO to everything, and 183–205. immediately)]. February. Gobe, E. 2008. “Les syndicalismes Arabes au Cherkaoui, M., and D. B. Ali. 2007. “The prisme de l’autoritarisme et du corporatisme” Political Economy of Growth in Morocco.” [Arab trade union movements under the lens Quarterly Review of Economics and Finance of authoritarianism and corporatism]. In 46: 741–61. Autoritarismes démocratiques et démocraties Ci ng ra nelli, D., a nd D. R icha rds. 1999. autoritaires au XXIe siècle [Democratic “Measuring the Level, Pattern, and Sequence authoritarianism: Authoritarian democracies of Government Respect for Physical Integrity in the 21st century], ed. O. Dabène, V. Geisser, Rights.” International Studies Quarterly and G. Massardier, 267– 84. Paris: La 43 (2): 407–18. Découverte. CIRI (Cingranelli-Richards) Human Rights Grun, R. 2009. “Renforcer les responsabilités Dataset. http://ciri.binghamton.edu. les Conseils de Gestion des établissements Dillman, B. 2001. “Facing the Market in et améliorer la qualité des apprentissages.” North Africa.” Middle East Journal 55 (2): [Strengthening the responsibility of school 198–215. management councils and improving learning Diwan, I., and L. Squire. 1993. “External Finance outcomes]. Note de Cadrage d’une opération in the Middle East: Trends and Prospects.” pilote [Concept Note of a Pilot]. Issue Note Discussion Paper 10. Middle East and North JSDF 093996, World Bank, Washington, DC. Africa, World Bank, Washington, DC. Guven, A. B. 2012. “Structural Reforms Under Dolado, J. J., M. Jansen, and J. F. Jimeno. 2002. Broad-based Policies: Turkey’s Justice and “A Matching Model of Crowding-Out and Development Party in the 2000s.” Working On-the-Job Search (with an Application to paper, World Bank, Washington, DC. Spain).” Working Papers 2002–16, Foundation Heydemann, S. 2007. Upgrading Authoritarianism for Applied Economics Studies, Madrid. in the Arab World. Saban Center for Middle Duval, R., and J. Elmeskov. 2005. “The Effects East Policy. Washington, DC: Brookings of EMU on Structural Reforms in Labour Institution. and Product Markets.” OECD Economics Hoekman, B. 1995. “The WTO, the EU and Department Working Papers 438, OECD, the Arab World: Trade Policy Priorities and Paris. Pitfalls.” CEPR Discussion Papers 1226, E con L it (d at abase). A mer ic a n E conom ic C ent re for E conom ic Polic y Resea rch, Association. http://www.aeaweb.org/econlit/ London. Esfahani, H. S. 1994. “Lessons from the Political Høj, J., V. Galasso, G. Nicoletti, and T.-T. Dang. Economy of Privatization and Public Sector 2006. “The Political Economy of Structural Reform in the Middle East and Southeast Reform: Empirical Evidence from OECD Asia.” Quarterly Review of Economics and Countries.” OECD Economics Department Finance 34 (Supplement 1): 301–31. Working Paper 501, OECD, Paris. Estache, A., E . Ianchovichina, R. Bacon, Kaufmann, D. 2011. Governance and the Arab and I. Salomon. 2012. Infrastructure and World Transition: Ref lections, Empirics Employment Creation in the Middle East and Implic ations for the Intern ation al and North Africa. Directions in Development Community. Washington, DC: Brookings Series. Washington, DC: World Bank. Institution. THE POLITICAL ECONOMY OF INCLUSION 311 Keefer, P. 2007. The Political Economy of Private Posusney, M . P., a nd L . J. C ook. 20 02 . Sector Development in the Middle East and Privatization and Labor: Responses and North Africa: Credible Commitment, Rent- C o n s e q u e n c e s i n G l o ba l Pe rs pec t i ve . Seeking and Institutionalization. Washington, Northampton, United Kingdom: Edward DC: World Bank. Elgar. Kienle, E. 2001. A Grand Delusion: Democracy Premand, P., S. Brodmann, R. Grun, and and Economic Reform in Egypt. New York: R. Almeida. Forthcoming. “Entrepreneurship IB Tauris. Trai n i ng a nd S el f-E mploy ment a mong López-Claros, Augusto, and Klaus Schwab, eds. Universit y Graduates: Evidence from a 2005. Arab World Competitiveness Report Randomized Trial in Tunisia.” Working paper, 2005. New York: Palgrave MacMillan . World Bank, Washington, DC. Luciani, G., ed. 1990. “Allocation vs. Production Razzaz, O. 2011. “The Political Economy of States: A Theoretical Framework.” In The Social Security Reform: Lessons from the Arab State, ed. G. Luciani, 65–84. Berkeley: Jordan Experience.” Working paper, World University of California Press. Bank, Washington, DC. Myerson, R. 2009. “Local Foundations for Strong Reinikka, R., and J. Svensson. 2003. “The Power Democracy in Pakistan.” Mimeo. http://home of Information: Evidence from an Information .uchicago.edu/~rmyerson/research/paklocal.pdf. Campaign to Reduce Capture.” Development National Lawyers Guild. 2011. Promises and Research Group, World Bank, Washington, Challenges: The Tunisian Revolution of DC. 2010 –2011. New York: National Lawyers Reporters without Borders. 2012. “Press Freedom Guild. Index 2011–12.” http://en.rsf.org /press- North, D. C., J. J. Wallis, S. B. Webb, and freedom-index-2011-2012,1043.html. B. R. Weingast. 2009. “Limited Access Orders: Rutkowski, J. 2011. “Labor Market Observatories: Rethinking the Problems of Development T he Pol ish E x p er ienc e” Unpubl ished. and Violence.” http://web.cenet.org.cn / Washington DC, World Bank. upfile/121138.pdf. S a i n t- Pau l , G . 2 0 0 2 . “M a c r o e c o n o m i c North, D. C., J. J. Wallis, and B. R. Weingast. Fluctuations and the Timing of Labour 20 09. Viol e n c e a n d S oci al O rd e rs: A Market Reforms.” CEPR Working Paper Conceptual Framework for Interpreting 3646, Centre for Economic Policy Research, Recorded Hum an Histor y . Cambridge, London. United Kingdom: Cambridge University Press. Scott, K., and R. Yemtsov. 2010. “MENA Olsen, M . 20 0 0. Powe r an d Pros pe rit y: Economic and Social Statistics Monitoring.” Outgrowing Communist and Capitalist Concept Note for the Regional Activity Dictatorships. New York: Basic Books. (P119449).” World Bank, Washington, DC. Onis, Z., and S. B. Webb. 1992. “Political Shafi k, N. 1995. Claiming the Future: Choosing Economy of Policy Reform in Turkey in the Prosperity in the Middle East and North 1980s.” Policy Research Working Paper 1059, Africa. Washington, DC: World Bank. World Bank, Washington, DC. Solidarity Center. 2010. The Struggle for Workers Paczynska, A. 2006. “Globalization, Structural Rights in Egypt. Washington, DC: Solidarity Adjustment, and Pressure to Conform: Center. Contesting Labor Law Reform in Egypt.” New Tversky, A., and D. Kahneman. 1991. “Loss Political Science 28: 45–64. Aversion in Riskless Choice: A Reference- ———. 2007. “Confronting Change: Labor, State, Dependent Model.” Quarterly Journal of and Privatization.” Review of International Economics 106 (4): 1039–61. Political Economy 14: 333–56. Vegas, E., ed. 2005. Incentives to Improve Posusney, M. P. 1997. Labor and the State in Teaching: Lessons from Latin America . Egypt: Workers, Unions, and Economic Washington, DC: World Bank. R e s t r u c t u r i n g . N e w York : C olu mbi a W IEGO Network and World Bank. 2008. University Press. Raising Productivity and Reducing Risks ——— . 20 03. “Globa l i z at ion a nd L abor of Hou sehold Ente r pri ses: Di ag nostic Protection in Oil-Poor Arab Countries.” Methodology Framework. Washington, DC: Global Social Policy 3: 267–97. World Bank. 312 JOBS FOR SHARED PROSPERITY World Bank. 2006. Access to Finance and ——— . 2 0 1 2 a . G e n d e r E q u a l i t y a n d Economic Growth in Egypt. Washington, DC: Development: World Development Report World Bank. 2012. Washington, DC: World Bank. ———. 2009. From Privilege to Competition: ———. 2012b. MENA Economic Development Unlocking Private-Led Growth in the Middle Prospects: Investing for Growth and Jobs. East and North Africa . Washington, DC: Washington, DC: World Bank. World Bank. Zakaria, A. 2006. “The Process of Issuing ———. 2011. Financial Access and Stability: the Work Code in Morocco.” Unpublished A Roadmap for the Middle East and North manuscript. Center for Development Research Africa. Washington, DC: World Bank. (ZEF), University of Bonn, Bonn, Germany. Appendix: Main Data Sources Economy Survey Year Description Egypt, Arab Labor Force 2010 Egypt’s LFS 2010 was conducted by the Central Agency for Public Mobilization and Rep. Survey (LFS) Statistics (CAPMAS), with the goal of measuring the composition of the Egyptian civilian labor force, employment and unemployment levels, the characteristics of those who are employed, and the geographic distribution of the different individuals composing the labor force. The survey was conducted quarterly, using the population census as a sample frame, with a total sample size of 85,408 households per year. Egypt, Arab Survey of Young 2009 Egypt’s SYPE was conducted by the Population Council and CAPMAS, with support from Rep. People in Egypt the Information and Decision Support Center and the National Center for Exams and (SYPE) Educational Evaluation. The sample was designed to be representative at the national level and for four administrative regions (urban governorates, Lower Egypt governorates, Upper Egypt governorates, and the frontier governorates). The sample size of approximately 17,000 young people, ages 10–29, was selected to provide estimates of key indicators related to adolescents and youth for the country as a whole and for the administrative regions and their urban and rural segments. The goal of the survey was to update the state of knowledge on youth and adolescents in Egypt on issues of health, education, employment, family formation, and civic participation. Egypt, Arab Egypt Labor 1998, 2006 Egypt’s LMPS was conducted by the Population Council and CAPMAS with support from Rep. Market Panel the U.S. Agency for International Development Egypt and the Ford Foundation. The LMPS Survey (LMPS) 2006 is designed as a panel survey and follows the same households and individuals that were interviewed in the Egypt Labor Market Panel Survey of 1998 and re-interviews them. Individuals who split from the original 1998 households in the intervening period are also tracked and interviewed, together with their entire household. Of the 8,371 households interviewed in 2006, 3,701 were households that were interviewed in LMPS 1998; 2,167 were splits from the original households; and 2,498 were part of an entirely new refresher sample. The sample size is 17,364 individuals and 8,371 households. The surveys contain rich information about individuals’ education, employment status, occupation, economic activity, firm size, wage, pension contribution, and the like. (continued next page) 313 314 APPENDIX: MAIN DATA SOURCES Economy Survey Year Description Iraq Iraq Household 2006–07 The Ministry of Planning and Development Corporation and the Central Organization for Socioeconomic Statistics and Information Technology conducted HSES with the support of the World Survey (HSES) Bank in an effort to reduce poverty and promote social development. Providing essential data for understanding the nature and causes of poverty among Iraqi households, HSES is the largest household social and economic survey ever conducted in Iraq; it reached a total of 18,144 households. HSES provides information on education, labor, health, income, and expenditure in Iraq. It has a sample size of 17,822 households and 127,189 individuals. Jordan Jordan Labor 2010 Jordan’s LMPS 2010 was carried out by the Economic Research Forum in cooperation with Market Panel the National Center for Human Resource Development and the Jordanian Department Survey (LMPS) of Statistics. For the first time in Jordan, detailed information about Jordanian labor market experiences and behaviors is available in the LMPS. It has a sample size of 25,969 individuals and contains rich information about individuals’ education, employment status, occupation, economic activity, firm size, wage, pension contribution, and the decision making related to labor force participation. It allows for a much richer linking of individual characteristics with labor market outcomes. Lebanon Lebanon 2011 The Lebanon Employer-Employee Survey was conducted by the World Bank. It is a Employer- nationally representative household-based survey covering a sample of 1,841 households. Employee Survey The survey collects basic information such as age, education, and employment for the (EES) entire household; detailed information on employment (current and history), skills and training, wages, and work benefits for each individual in the household over 15 years of age who is either unemployed, self-employed, or a salaried employee; and the level of cognitive and noncognitive skills of workers. Morocco Morocco 2010 Morocco’s HYS 2009–10, administered from December 2009 through March 2010, Household and collected information from a nationally representative sample of 2,000 households across Youth Survey the country (1,216 households were urban and 784 were rural) on their demographic (HYS) and educational characteristics, economic activities, migration, and social program participation. Data on household asset ownership were used to construct a household wealth index and classify households into welfare deciles. In addition to the household module, which collected information on all members, a separate youth module focused on young people, ages 15–29, in the 2,000 surveyed households. Consequently, information related to youth economic inclusion, community participation, and use of key public services was collected from 2,883 young individuals. The survey thus gathered information on understudied issues related to youth, such as labor force participation and intermediation, career choices and perceived employment opportunities, use of free time, and use of youth-oriented recreational and educational services that complement formal education. Morocco Morocco Labor 2009 Morocco’s LFS 2009 was conducted by the Haut Commissariat au Plan, Direction de la Force Survey Statistique, on a nationally and regionally representative sample (the sample size was (LFS) 1 percent of the total population). The survey contains information on the size and structure of Morocco’s labor force, including their demographic and educational characteristics, employment status, occupation, and economic activities. Syria Syrian Arab Republic 2009 The Syria Employer-Employee Survey is a nationally representative household-based Employer- survey, conducted by the World Bank. The survey collects basic information such as Employee Survey age, education, and employment for the entire household; detailed information on employment (current and history); skills and training, wages, and work benefits for each individual in the household over 15 years of age and is either unemployed, self-employed, or a salaried employee; and the level of cognitive and noncognitive skills of workers. Tunisia Tunisia Labor 2005, 2009, Tunisia’s LFS 2005, 2009, and 2010 were conducted by the National Statistics Institute. This Force Survey 2010 survey covers the entire country using the population census as a sampling frame. The (LFS) sample size in 2010 was 549,015 individuals. The survey consists of a household module, a module for the employed labor force, and a module for the unemployed. The survey contains rich information on the size and structure of the Tunisian labor force, including individuals’ education, employment status, occupation, economic activity, place of work, hours and days worked, reasons for and duration of unemployment, as well as prior employment. (continued next page) APPENDIX: MAIN DATA SOURCES 315 Economy Survey Year Description Turkey Turkey Household 2010 Turkey’s HLFS was conducted by the Turkish Statistical Institute (TURKSTAT), using as a Labor Force sampling frame the Address Based Population Register System (updated every month) Survey (HLFS) The survey covers the entire country population, sampling around 14,100 households each month, 12 times per year, and it contains rich information on economic activity, occupation, status in employment, and hours worked for the employed persons; it also informs on the duration of unemployment and occupation sought for the unemployed. United Arab Labor Force 2009 The United Arab Emirates’ LFS 2009 was conducted by the National Bureau of Statistics. Emirates Survey (LFS) This survey collected information from a nationally representative sample of 11,024 households across the country and was conducted using the population census as a sampling frame. The survey contains information on the size and structure of the labor force of the United Arab Emirates, including their demographic and educational characteristics, employment status, occupation, economic activities, migration trends, and social program participation. West Bank Labor Force 2008 The West Bank and Gaza’s LFS 2008 was conducted by the Palestinian Central Bureau of and Gaza Survey (LFS) Statistics, as a part of the Palestinian Labor Force Survey Program. This survey is conducted quarterly, and, in 2008, the total sample size was 30,180 households and 107,991 individuals ages 10 and older. The survey contains rich information on the size and structure of the West Bank and Gaza labor force, including the individuals’ education, employment status, occupation, economic activity, place of work, hours and days worked, wages, pension contribution, and the like. Yemen, Household 2005–06 The Household Budget Survey 2005–06 was conducted by the Central Statistical Rep. of Budget Survey Organization of Yemen. The HBS data contain information on household roster, economic (HBS) activities, dwelling conditions, health, education, anthropometrics, income, durable goods, and consumption. One of the main objectives of the HBS 2005–06 is producing aggregates of the statistical indicators at the level of the urban and rural communities of each governorate to support economic and social development planning at the central and local levels. HBS 2005 contains observations on 13,136 households and 98,941 individuals. World Gallup World Poll 2009, 2010, The Gallup World Poll Survey data are collected in more than 150 countries. The target Survey 2011 population is the entire civilian, noninstitutionalized population, ages 15 and older, and samples are probability based and nationally representative (with some exceptions). A standard set of core questions is used around the world to measure global attitudes and behaviors, job creation, city prosperity, global migration, and well-being. In addition, supplemental questions might be asked in some regions. The typical World Poll survey includes at least 1,000 surveys of individuals per year; however, in some countries, oversamples are collected in major cities or areas of special interest. ECO-AUDIT Environmental Benefits Statement The World Bank is committed to preserving Saved: endangered forests and natural resources. • 19 trees The Office of the Publisher has chosen to • 9 million BTUs of print Jobs for Shared Prosperity: Time total energy for Action in the Middle East and North • 1,591 pounds of net Africa on recycled paper with 30 percent greenhouse gases postconsumer fiber in accordance with the • 8,632 gallons of recommended standards for paper usage set waste water by the Green Press Initiative, a nonprofit • 578 pounds of program supporting publishers in using solid waste fiber that is not sourced from endangered forests. For more information, visit www .greenpressinitiative.org.