95769 Trust, Voice, and Incentives Learning from Local Success Stories in Service Delivery in the Middle East and North Africa ‘‘ We are often forced to pay bribes and give out gifts to get our paperwork done at various governmental departments, ‘‘ especially the traffic and real Teachers are often unable estate departments, where to teach because of crowded ’’ corruption is rampant. classrooms, low salaries, and unresponsive school ’’ management. ‘‘ There is simply no point in expressing your opinion. No one listens. ’’ ‘‘The problem is that public servants are not employed based on their qualifications, but because they ’’ belong to specific parties. ‘‘ Public services are not good. For instance, we have an obvious problem in sectors. If you go to the hospital, you may just find one doctor, lots of sick people, ‘‘ Citizens have a role to play in changing the current reality of institutions by fighting all forms and not enough medications. of corruption and bribery. This Same problems occur in the can only happen with unity and ’’ education sector. popular consciousness. Such problems will not be resolved when citizens keep encouraging and giving bribes to public ’’ officials. ‘‘ Engagement with the community is different. It is regular now. Parents are involved in decision making; they oversee and support the school. They provide support to the teachers and ‘‘In the clinic, we are all one team. We work together and help each other to care for our patients. ’’ school by their active participation. But if a staff member does not ’’ care, he or she is out. Sources: Comments received on the World Bank Middle East and North Africa Facebook page (https://www.facebook.com/ WorldBankMiddleEastNorthAfrica) during online consultations for this report (quotes 1–6); comments received during local success case studies (quotes 7–8). Trust, Voice, and Incentives Learning from Local Success Stories in Service Delivery in the Middle East and North Africa Hana Brixi, Ellen Lust, and Michael Woolcock © 2015 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 18 17 16 15 This work is a product of the staff of The World Bank with external contributions. The fi ndings, inter- pretations, 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 IGO license (CC BY 3.0 IGO) http://creativecommons.org/licenses/by/3.0/igo. 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: Brixi, Hana, Ellen Lust, and Michael Woolcock. 2015. Trust, Voice, and Incentives: Learning from Local Success Stories in Service Delivery in the Middle East and North Africa. Washington, DC: World Bank. doi:10.1596/978-1-4648-0456-4. License: Creative Commons Attribution CC BY 3.0 IGO 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. Adaptations —If you create an adaptation of this work, please add the following disclaimer along with the attribution: This is an adaptation of an original work by The World Bank. Views and opinions expressed in the adaptation are the sole responsibility of the author or authors of the adaptation and are not endorsed by The World Bank. Third-party content—The World Bank does not necessarily own each component of the content con- tained within the work. The World Bank therefore does not warrant that the use of any third-party- owned individual component or part contained in the work will not infringe on the rights of those third parties. The risk of claims resulting from such infringement rests solely with you. If you wish to re-use a component of the work, it is your responsibility to determine whether permission is needed for that re-use and to obtain permission from the copyright owner. Examples of components can include, but are not limited to, tables, figures, or images. All queries on rights and licenses should be addressed to the Publishing and Knowledge Division, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@ worldbank.org. ISBN (paper): 978-1-4648-0456-4 ISBN (electronic): 978-1-4648-0457-1 DOI: 10.1596/978-1-4648-0456-4 Cover design: Critical Stages Library of Congress Cataloging-in-Publication Data has been requested. Contents Foreword . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xv Acknowledgments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xvii About the Authors and the Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xix Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xxiii Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Citizens’ expectations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Local successes: Autonomy, accountability, and participation . . . . . . . . . . . . . . . . . . . . . . . 2 The cycle of poor performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Bases for improvement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 Reference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 Data sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Organization of this report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 Note . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 PART I Expectations, Reality, and Inspiration in Education and Health Services Delivery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 1. A Demand for Better Services but Not Formal Accountability. . . . . . . . . . . . . . . . . . . . . . 21 The right to education and health services in the MENA region . . . . . . . . . . . . . . . . . . . . 21 Citizens’ satisfaction and demands on government . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 Citizens’ trust and engagement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 Data sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 v vi CONTENTS 2. The State of Education and Health Services Delivery: The Quality Challenge amid Impressive Advances . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Outcomes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Access . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 The quality challenge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 Citizens’ satisfaction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 Reflections in the wider historical and institutional contexts . . . . . . . . . . . . . . . . . . . . . . . 61 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 Data sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 3. Local Successes: Satisfaction, Accountability, and Quality at the Local Level . . . . . . . . . . 67 Case study 1: Education services in Jenin, West Bank . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 Case study 2: Education services in Yarqa, Jordan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 Case study 3: Health care services in Jordan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 Case study 4: Health care services in Morocco . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 Case studies to share ideas and inspire change . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 Data source . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 PART II The Historical and Institutional Drivers of Performance . . . . . . . . . . . . . . . . 103 4. Historical and Cultural Roots of Citizens’ Attitudes and State Performance . . . . . . . . . . 107 Education and health as elite privilege . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108 From privilege to state building . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110 Independence and the new social compact: Services, a right for all . . . . . . . . . . . . . . . . . 111 Institution building in newly independent states . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112 Economic crises, neoliberal policies, and the failure to reconstruct the social compact . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113 High expectations, weak institutions, and low performance . . . . . . . . . . . . . . . . . . . . . . 115 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 Data sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 5. Institutions Influencing the Cycle of Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123 Political institutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125 Administrative institutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135 Social institutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144 Potential for reform? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158 Data sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163 PART III Service Delivery Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165 6. Service Delivery: Gaps in Providers’ Efforts and Abilities and in Inputs . . . . . . . . . . . . . 167 Providers’ efforts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168 Providers’ abilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172 Availability of key inputs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175 CONTENTS vii Interpreting service delivery performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179 Data sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179 7. Subnational Variation in Service Delivery Performance . . . . . . . . . . . . . . . . . . . . . . . . . . 181 Subnational variation in health services delivery in the Republic of Yemen . . . . . . . . . . . 182 Subnational variation in health services delivery in Morocco . . . . . . . . . . . . . . . . . . . . . . 187 Education and health services in Egypt. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195 Education services in the Republic of Yemen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197 Education services in Morocco . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199 Education services in Saudi Arabia, Oman, Qatar, the United Arab Emirates, Bahrain, and Tunisia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203 Concluding observations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207 Annex A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209 Annex B . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210 Reference . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211 Data sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211 PART IV Citizens’ Trust and Engagement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213 8. Citizens’ Response to Poor Performance and Unresponsive Institutions? Lower Trust . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215 Blaming the state for poor services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216 Lacking trust in institutions and actors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224 Annex Predicted probabilities tables on trusting the national government. . . . . . . . . . . 225 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 230 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231 Data sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232 9. Effects of Lower Citizen Trust on Citizen Engagement: Circumventing the State, Relying on Wasta, and Pursuing Confl ict . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233 Directly engaging the state and public services delivery institutions . . . . . . . . . . . . . . . . . 234 Circumventing the state . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239 Clashing with the state . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241 The use of social media . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243 Annex Predicted probabilities tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 248 Data sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249 PART V New Openings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251 10. Transitions, Conflicts, and Refugees: Both an Opportunity for and a Strain on the Cycle of Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253 Transitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 254 Confl ict . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 262 viii CONTENTS Refugees . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 270 Data sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274 11. Donor Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275 An overview of the World Bank’s education and health activities in the MENA region . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 276 Factors shaping variation in operational effectiveness . . . . . . . . . . . . . . . . . . . . . . . . . . . 279 The challenge of project implementation in the MENA region . . . . . . . . . . . . . . . . . . . . . 283 Lessons and opportunities going forward . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 288 Conclusion: Are there even more “lessons” to be learned? . . . . . . . . . . . . . . . . . . . . . . . . 289 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292 Data sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293 12. Incremental Changes in the Social Contract and Local Empowerment . . . . . . . . . . . . . . 295 Lessons from experience . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 296 Strengthening accountability and incentives through incremental changes. . . . . . . . . . . . 297 Empowering communities and local leaders to fi nd best-fit solutions . . . . . . . . . . . . . . . . 304 Delivering quick wins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 310 Moving forward . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 312 Boxes 5.1 Examples of underperforming health centers in the MENA region . . . . . . . . . . . . . . 146 5.2 Improving quality of care against all odds: Mobilizing social ties in Jordan . . . . . . . 151 6.1 The demand-side story in the quality of service provision . . . . . . . . . . . . . . . . . . . . . 178 7.1 Notable performers in health services delivery in Morocco . . . . . . . . . . . . . . . . . . . . 193 11.1 World Bank’s adaptable program loans and learning and innovation loans. . . . . . . . 284 11.2 The independent evaluation group’s process for conducting implementation completion reports: How it works . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285 11.3 Implementing education projects in the Republic of Yemen: The Secondary Education Development and Girls Access Project (SEDGAP) . . . . . . . . . . . . . . . . . . 289 12.1 How assessing schools and clearly informing the public promote learning in Dubai . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 305 12.2 Tunisia’s experience in building citizen feedback loops on service delivery . . . . . . . . 308 Figures O.1 Voiced opinion to public officials in the last month: MENA and other regions, 2013 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 O.2 The cycle of performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 O.3 Importance of wasta in obtaining a government job, by municipality: Jordan, 2014 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 O.4 Percentage of respondents reporting payment of informal fees, education and health care sectors: MENA region and globally, 2013 . . . . . . . . . . . . . . . . . . . . . . 7 CONTENTS ix O.5 Percentage of students whose principals report that teacher absenteeism is a serious problem in their school: MENA region and globally, 2011 . . . . . . . . . . . . . . . 8 O.6 Adherence to care protocols for diabetes and coronary heart disease in health facilities: Arab Republic of Egypt, 2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 O.7 Availability of aspirin, by province: Morocco, 2011 . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 O.8 SABER analysis of World Bank School Autonomy and Accountability Interventions, by region, FY03–13 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 I.1 The cycle of performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 1.1 Prioritizing values: MENA and other regions, 2010–14. . . . . . . . . . . . . . . . . . . . . . . . 23 1.2 Citizens’ priorities among services: Various regions, 2014 . . . . . . . . . . . . . . . . . . . . . . 23 1.3 Top two issues in need of the greatest investment: Arab Republic of Egypt, 2011 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 1.4 Satisfaction with education services: MENA and other regions, 2013 . . . . . . . . . . . . . 24 1.5 Satisfaction with the availability of quality health care: MENA and other regions, 2013 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 1.6 Satisfaction with services across sectors: MENA and other regions, 2013 . . . . . . . . . . 25 1.7 Education and health systems—corrupt or extremely corrupt: Selected MENA economies, 2013 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 1.8 Public sector corruption score: Selected MENA economies, 2014 . . . . . . . . . . . . . . . . 26 1.9 Perceptions of government’s efforts to fight corruption: MENA and other regions, 2013 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 1.10 Satisfaction with efforts to deal with the poor: MENA and other regions, 2013 . . . . . 28 1.11 On a continuum of 1 to 10, citizens’ expectations of their government: MENA and other regions, 2010–14 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 1.12 Perceptions of performance of current government in improving basic health services: Selected MENA economies, 2010–11 . . . . . . . . . . . . . . . . . . . . . . . . . 29 1.13 Citizens’ perceptions of service delivery responsibility: Tunisia, 2014 . . . . . . . . . . . . . 29 1.14 Satisfaction with education services and trust in government: MENA and other regions, 2013 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 1.15 Satisfaction with health services and trust in government: MENA and other regions, 2013 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 1.16 Voiced an opinion to a public official in the last month: MENA and other regions, 2013 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 2.1 Performance of MENA countries in education and health: Females who completed primary school and child mortality based on income per capita, 1975 and 2011 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 2.2 Percentage of children aged 12–23 months immunized for measles: MENA and OECD countries, selected years, 1975–2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 2.3 Net enrollment rates (adjusted) for primary education: MENA and OECD countries, selected years, 1971–2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 2.4 Ranking of top 25 causes of death in the Arab world, 1990 and 2010 . . . . . . . . . . . . 38 2.5 Gross enrollment ratios by education level: MENA and OECD countries, 1999–2012 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 2.6 Regional variation in child malnutrition: Morocco, 2010–11 . . . . . . . . . . . . . . . . . . . 42 2.7 Child mortality (under 5 years): Arab Republic of Egypt, 2008 . . . . . . . . . . . . . . . . . 42 2.8 Geographic inequities in access to health services: Selected MENA countries, 2010. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 2.9 Income inequities in access to health services: Selected MENA countries, 2010 . . . . . 43 2.10 Differentials in the receipt of antenatal care by wealth quintile: Arab Republic of Egypt (2008) and Morocco (2003) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 x CONTENTS 2.11 Limitations in access to antenatal care in rural areas: MENA and other countries, various years . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 2.12 Percentage of married women mentioning lack of money as a barrier to health care, by region: Morocco, 2010–11 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 2.13 Difficulty or ease in obtaining medical treatment in a nearby clinic or public hospital: MENA region, 2010–11 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 2.14 Benchmarking student math scores against levels of GDP per capita: MENA economies, 2011 and 2012 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 2.15 Student performance on TIMSS mathematics benchmarks (grade 8): MENA and OECD economies, 2011 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 2.16 Student performance on TIMSS mathematics (grade 8), by socioeconomic background of schools: MENA economies and internationally, 2011 . . . . . . . . . . . . . 49 2.17 Preliminary estimates of efficiency frontier, life expectancy at birth: Selected MENA countries, 2000–10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 2.18 Preliminary estimates of efficiency frontier, net primary enrollment: Selected MENA countries, 2000–10 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 2.19 Preliminary estimates of efficiency frontier, 2012 PISA math scores and 2011 TIMSS math scores for eighth-grade students: Selected MENA countries . . . . . . . . . . 53 2.20 Preliminary estimates of efficiency frontier, percentage of women receiving antenatal care: Selected MENA countries, available years from 1997 to 2013 . . . . . . 53 2.21 Repetition rates, by education level and geographic coverage/gender in public schools: Morocco, 2010–11 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 2.22 Patients’ dissatisfaction with aspects of health care: Republic of Yemen, 2010 . . . . . . 56 2.23 Variations in perceptions of service quality: Tunisia, 2014 . . . . . . . . . . . . . . . . . . . . . 57 2.24 Satisfaction with quality of garbage collection, by region: Tunisia, 2014. . . . . . . . . . . 57 2.25 Satisfaction with quality of education, by region: Tunisia, 2014 . . . . . . . . . . . . . . . . . 58 2.26 Percentage of respondents reporting payment of informal fees, education and health care sectors: MENA region and globally, 2013 . . . . . . . . . . . . . . . . . . . . . . . . . 59 2.27 Perceptions of role of corruption in services: Republic of Yemen, 2013 . . . . . . . . . . . . 60 2.28 Transparency in fees at health facilities: Morocco (2009), Arab Republic of Egypt (2010), and Republic of Yemen (2010) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 2.29 Fees for examination by doctor, by background characteristics: Arab Republic of Egypt, 2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 4.1 Per capita government expenditure on health: MENA and other countries, 2012 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115 4.2 Per capita government expenditure on primary education (percentage of GDP per capita): MENA and other countries, 2011 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116 4.3 Per capita government expenditure on secondary education (percentage of GDP per capita): MENA and other countries, 2011 . . . . . . . . . . . . . . . . . . . . . . . . . 117 4.4 Private primary school enrollment as a percentage of total primary enrollment: MENA region, selected years, 1985–2010 . . . . . . . . . . . . . . . . . . . . . . . 118 5.1 Political, administrative, and social institutions affecting service delivery . . . . . . . . . 124 5.2 Institutional constraints on the decision-making power of the executive: All regions, 2014 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126 5.3 Degree of legislative autonomy: All regions, 2014 . . . . . . . . . . . . . . . . . . . . . . . . . . . 126 5.4 Regional averages of judicial independence: All regions, 2014 . . . . . . . . . . . . . . . . . . 127 5.5 Performance of MENA region compared with other regions in effectiveness of laws governing civil service administration, 2009 . . . . . . . . . . . . . . . . . . . . . . . . . 128 5.6 Public access to information: All regions, 2009 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129 5.7 Open budget index: All regions, 2013 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130 CONTENTS xi 5.8 Readiness subindex: Selected MENA countries and world average, 2013 . . . . . . . . . 130 5.9 Right to information law index: All regions, 2013. . . . . . . . . . . . . . . . . . . . . . . . . . . 131 5.10 Legislative influence, autonomy, and power: MENA region, 2009 . . . . . . . . . . . . . . 134 5.11 Index of extent to which citizens can participate in government selection and freedoms of expression, association, and media: MENA and other regions and selected MENA economies, 2012 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136 5.12 Percentage of duplication in supervisory visits to primary health facilities (ESSBs): Morocco, 2011 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143 5.13 In-kind transfer flows to primary health facilities (ESSBs) in the Moroccan public health system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143 5.14 Percentage of leakage in medical supplies in shipments from provincial delegations to ESSBs: Morocco, 2011 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144 5.15 Global integrity scores, MENA region . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145 5.16 Influence of families and clans: Tunisia and Jordan, 2014 . . . . . . . . . . . . . . . . . . . . . 148 5.17 Percentage of fi rms making informal payments to public officials: Selected economies, 2005–13 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153 5.18 Importance of wasta and bribes in obtaining medical treatment and education: Tunisia, 2014 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155 6.1 Percentage of students whose principals report that teacher absenteeism is a serious problem in their schools: OECD, non-OECD, and MENA economies, 2011. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169 6.2 Percentage absent of total employed in health facility: Arab Republic of Egypt, Morocco, and Republic of Yemen, 2010 and 2011 . . . . . . . . . . . . . . . . . . . . . . . . . . 169 6.3 Percentage of chronic care observations conducted: Arab Republic of Egypt, 2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172 6.4 Percentage of child outpatient visit observations conducted: Arab Republic of Egypt, 2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173 6.5 Percentage of students (grade 8) who report reliance on memorizing lessons in math: OECD, non-OECD, and MENA economies, 2011 . . . . . . . . . . . . . . . . . . . 173 6.6 Percentage of students (grade 8) who attend schools with severe shortages of specialized math teachers: OECD, non-OECD, and MENA countries, 2011 . . . . . . 174 6.7 Percentage of students (grade 8) who attend schools with severe shortages of instructional materials: OECD, non-OECD, and MENA economies, 2011 . . . . . . . . 176 6.8 Medication availability: Morocco (2011) and the Arab Republic of Egypt (2010) . . . 177 7.1 Availability of electricity and water in health care facilities, by governorate: Republic of Yemen, 2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183 7.2 Average number of beds per health care facility, by district: Republic of Yemen, 2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184 7.3 Absenteeism in health care facilities, by district: Republic of Yemen, 2010 . . . . . . . . 185 7.4 Availability of medical equipment at health facilities, by region: Morocco, 2011 . . . 187 7.5 Availability of medical equipment at health facilities, by province/prefecture: Morocco, 2011 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188 7.6 Availability of aspirin, by region: Morocco, 2011 . . . . . . . . . . . . . . . . . . . . . . . . . . . 188 7.7 Availability of aspirin, by province/prefecture: Morocco, 2011 . . . . . . . . . . . . . . . . . 189 7.8 Availability of metformin, by region: Morocco, 2011 . . . . . . . . . . . . . . . . . . . . . . . . 190 7.9 Availability of metformin by province: Morocco, 2011 . . . . . . . . . . . . . . . . . . . . . . . 190 7.10 Absenteeism among health facility staff, by region: Morocco, 2011 . . . . . . . . . . . . . 191 7.11 Absenteeism among health facility staff, by province/prefecture: Morocco, 2011 . . . 191 7.12 Absenteeism among doctors, by region: Morocco, 2011 . . . . . . . . . . . . . . . . . . . . . . 192 7.13 Absenteeism among doctors, by province/prefecture: Morocco, 2011 . . . . . . . . . . . . 192 xii CONTENTS 7.14 Adherence to sick child care protocol, counting breaths: Arab Republic of Egypt, 2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195 7.15 Adherence to diabetes care protocol, examining feet and legs for pulsations: Arab Republic of Egypt, 2010 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195 7.16 Student performance on TIMSS mathematics (grades 4 and 6), by sex and location (rural or urban): Republic of Yemen, 2011 . . . . . . . . . . . . . . . . . . . . . . . . . . 198 7.17 Level of teacher education by ISCED level, grades 4 and 6 TIMSS math scores: Republic of Yemen, 2011 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199 7.18 Teacher absenteeism, by governorate: Republic of Yemen, 2006 . . . . . . . . . . . . . . . . 200 7.19 Student performance on TIMSS mathematics (grade 8), by region: Morocco, 2011 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 200 7.20 Availability of instructional materials, by region: Morocco, 2011 . . . . . . . . . . . . . . . 201 7.21 Teacher absenteeism, by region: Morocco, 2011 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202 7.22 School performance, by sex and location: Saudi Arabia, 2011. . . . . . . . . . . . . . . . . . 203 7.23 Education outcomes, by sex and location: Oman, 2011 . . . . . . . . . . . . . . . . . . . . . . . 204 7.24 Education outcomes, by sex and location: Qatar, 2011 . . . . . . . . . . . . . . . . . . . . . . . 204 7.25 Student achievement scores, by region: United Arab Emirates, 2011 . . . . . . . . . . . . . 205 7.26 Availability of instructional materials, grades 4 and 8, by region: United Arab Emirates, 2011 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205 7.27 Teacher absenteeism, grades 4 and 8, by region: United Arab Emirates, 2011 . . . . . . 206 7.28 Student achievement scores, public and private schools, by region: Bahrain, 2011 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206 7.29 Availability of instructional materials, public and private schools, by region: Bahrain, 2011 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207 7.30 Teacher absenteeism, public and private schools, by region: Bahrain, 2011 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207 8.1 Perceptions of the effectiveness of governments: MENA region, 2006–08, 2010–11 . . . . 219 8.2 Perceptions of the effectiveness of parliaments: MENA region, 2010–11 . . . . . . . . . . . . 219 8.3 Subnational variation in trust in state institutions, by region: Tunisia, 2014 . . . . . . . 220 8.4 Predicted probability of trusting the national government as a function of satisfaction with education and health services and believing that corruption is pervasive: Selected MENA economies, 2013 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221 8.5 Predicted probability of trusting the judiciary as a function of satisfaction with education and health services and believing that corruption is widespread: Selected MENA economies, 2013 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222 8.6 Predicted probability of trusting the government to a great extent compared with absolutely not trusting it as a function of perceptions of the importance of connections in obtaining employment: Selected MENA countries, 2010–11 . . . . . 223 9.1 Rate of respondents who had and acted on a complaint about health services: Republic of Yemen, 2009 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235 9.2 Percentage of respondents participating and voting in elections at the subnational level: Jordan and Tunisia, 2014. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236 9.3 Predicted probability of becoming a member of a political party as a function of trust in political parties: Selected MENA countries, 2010–11 . . . . . . . . . . . . . . . . 237 9.4 Predicted probability of becoming a member of a youth/civil society organization (CSO) as a function of trusting CSOs: Selected MENA countries, 2010–11 . . . . . . . 238 9.5 Responsiveness of members of parliament (MPs): Arab Republic of Egypt (2012), Libya (2013), and Tunisia (2012) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240 9.6 Primary uses of government social media: MENA region, 2014 . . . . . . . . . . . . . . . . 243 9.7 Sources of information about government services: MENA region, 2014 . . . . . . . . . 243 CONTENTS xiii 10.1 The cycle of performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 254 10.2 Survey of citizens of transitioning economies: Prospects for good governance, the economy, government corruption, security and safety, the overall economy, and quality of government leadership: MENA region, 2013 . . . . . . . . . . . . . . . . . . . 257 11.1 Education and health lending, World Bank: MENA region, 1994–2014 . . . . . . . . . . 276 11.2 The changing role of voice, accountability, and transparency in World Bank education and health projects: MENA region, 1994–2013 . . . . . . . . . . . . . . . . . . . . 278 11.3 Analytical and advisory activity commitments, education and health, World Bank: MENA region, FY01–13 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 278 11.4 SABER analysis of World Bank school autonomy and accountability interventions, by region, FY03-13 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283 11.5 World Bank education and health lending instruments: MENA region, 1994–2014 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284 Maps I.1 Human Development Index (HDI) values for MENA, 2013 . . . . . . . . . . . . . . . . . . . . 20 II.1 Government Effectiveness Index values for MENA, 2013 . . . . . . . . . . . . . . . . . . . . . 105 III.1 Student scores on TIMSS mathematics in MENA, 2007–11 . . . . . . . . . . . . . . . . . . . 166 IV.1 Percentage of citizens voicing opinion to a public official in the last month in MENA, 2013 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214 Photos 3.1 Abla Habayeb, principal of Kufor Quod Girls’ Secondary School . . . . . . . . . . . . . . . . 71 3.2 Jordanian youth in an after-school setting at an elementary school . . . . . . . . . . . . . . . 79 3.3 Young Jordanian practicing her arithmetic at an elementary school . . . . . . . . . . . . . . 86 3.4 Sign greeting patients at a health center in Jordan: “Welcome to our center. We promise to present you with the highest quality of health services.” . . . . . . . . . . . 89 3.5 Suggestion and complaint box at a health care center in Jordan . . . . . . . . . . . . . . . . . 90 3.6 A well-managed and well-stocked pharmacy in one of Morocco’s primary health care centers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94 Tables 2.1 Preliminary estimates of efficiency scores, by region, 2010–11 . . . . . . . . . . . . . . . . . . 52 2.2 Correlation between individuals’ evaluations of different services (Spearman’s Rho): Tunisia, 2014 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 2.3 Correlation between individuals’ satisfaction with different services: Globally, 2013 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 2.4 Correlation between individuals’ satisfaction with different services: MENA region, 2013 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 5.1 Division of responsibility among central ministries, provincial and regional administrations, and schools on policy setting, planning, fi nance, human resource management, and pedagogy: Selected MENA countries, 2014 . . . . . . . . . . 138 5.2 Division of responsibility among the central ministries, provincial and regional administrations, and health facilities on policy setting, planning, fi nance, and human resource management: Selected MENA countries, 2014 . . . . . . 139 xiv CONTENTS 6.1 Key service delivery indicators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168 7.1 Correlation of absenteeism in health care facilities and other measures at the governorate and district levels, Republic of Yemen . . . . . . . . . . . . . . . . . . . . . . . 186 7.2 Correlation between absenteeism and other measures at regional and provincial/prefectural levels, Morocco . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193 7.3 Subnational variation in textbook distribution, by governorate: Arab Republic of Egypt, 2007 and 2008 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197 7.4 Correlation between inputs to education: Tunisia, 2011 . . . . . . . . . . . . . . . . . . . . . . 208 7A.1 Number of health facilities, by region and province: Republic of Yemen, 2010 . . . . . 209 7B.1 Health facilities surveyed by PETS, by region and province/prefecture: Morocco, 2011 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210 8.1 Correlation between evaluation of various services and belief in the importance of wasta and bribes: Tunisia, 2014 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217 8.2 Correlation between percentage of respondents who trust their national government and WGI measures: MENA region, 2013 . . . . . . . . . . . . . . . . . . . . . . . . 218 8A.1 Predicted probabilities of satisfaction with education services on trusting the national government: Selected MENA economies, 2013 . . . . . . . . . . . . . . . . . . . 225 8A.2 Predicted probabilities of satisfaction with availability of quality health care on trusting the national government: Selected MENA economies, 2013 . . . . . . . . . . 225 8A.3 Predicted probabilities of belief in pervasiveness of corruption within government on trusting the national government: Selected MENA economies, 2013 . . . . . . . . . . 226 8A.4 Predicted probabilities of satisfaction with education services on trusting the judiciary: Selected MENA economies, 2013 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227 8A.5 Predicted probabilities of satisfaction with availability of quality health care on trusting the judiciary: Selected MENA economies, 2013 . . . . . . . . . . . . . . . . . . . 227 8A.6 Predicted probabilities of belief in pervasiveness of corruption within government on trusting the judiciary: Selected MENA economies, 2013 . . . . . . . . . 228 8A.7 Predicted probabilities of perceptions of importance of connections in obtaining employment on trusting the government to a great extent compared with absolutely not trusting it: Selected MENA countries, 2010–11 . . . . . . . . . . . . . . . . . 229 9A.1 Predicted probability of trusting a political party on becoming a member of one: Selected MENA countries, 2010–11. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244 9A.2 Predicted probability of trusting a youth/civil society organization on becoming a member of one: Selected MENA countries, 2010–11 . . . . . . . . . . . . . . . 245 10.1 Labor strikes: Arab Republic of Egypt, 2006–12 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261 Foreword T he Middle East and North Africa unprecedented improvements in basic human (MENA) is in a state of volatile development outcomes, as well as food and change. This period of change is fuel subsidies and employment in the public imposing hardships on the people of the sector. It gave rise to a social contract bind- region, and the outbreaks of conflict and ing governments and citizens, establishing violence represent a clear danger not only political legitimacy and strong nationhood regionally, but globally as well. and instilling citizens’ support for govern- Under the circumstances, it is essential to ment, as well as the expectation that the state focus on the welfare of the region’s youth and would assume responsibility for economic its economic and institutional health in order and social welfare. to build the foundation for lasting stability The post-independence social contract and shared prosperity. Public institutions and became unsustainable because it hampered services are central to this effort. In particu- the creation of the inclusive institutions and lar, it is crucial to harness the ability of pub- accountability mechanisms at the political, lic institutions to align their incentives with administrative, and social levels that would the needs of the people; the ability of educa- motivate providers, public servants, and tion systems to help the region’s children and policy makers to deliver quality services to youth develop the competencies and skills the poor and other non-privileged popula- that will serve them well for their future lives tions. As a consequence, the majority of Arab and livelihoods; and the ability of health sys- children are not learning well enough while tems to address the rapidly growing burden at school, and many mothers with young of chronic diseases, as well as the remaining children find health clinics closed, doctors pockets of high maternal and child mortality absent, or essential medicines out of stock and malnutrition. when they need medical help. In the aftermath of independence, lead- This report focuses specifically on the pub- ers of MENA countries promised their citi- lic institutions underpinning service delivery. zens industrialization and better standards In the right institutional environment, the of living through state intervention. That interactions between these various stakehold- intervention included the expansion of edu- ers create a virtuous cycle of performance. In cation and health services, which generated such a cycle, citizens have the means to inform xv xvi FOREWORD policies and comment on the quality of ser- become a shared responsibility, and the results vices, and both public servants and provid- are impressive. The students excel in national ers are attuned to those opinions by a system tests despite the poverty and instability that of incentives that reward their effectiveness surround them. There are equally power- in responding to them. As services improve, ful examples in schools and health clinics in citizens’ trust in government increases, which Jordan and Morocco of citizens finding better serves to consolidate social cohesion. ways to provide services to their communities. However, this report documents that in There is no blueprint for creating the right the Middle East and North Africa this vir- accountability mechanisms to monitor and tuous cycle has not been achieved. Many of motivate public servants and service provid- the public institutions that deliver basic social ers. Every environment has its own distinct services are not responsive to citizens’ needs. social and political characteristics, and solu- As a result, services have suffered, leaving tions have to be either drawn from or adapted citizens with little recourse but to abandon to them. In this respect, we hope that the the system and seek alternate means of meet- report will prove useful to policy makers, ser- ing their needs. This negative cycle will have vice providers, and citizens alike. While map- to be broken in order for investments in edu- ping out the key drivers of effective service cation and health to begin delivering mean- delivery, it aims to launch a regional discus- ingful outcomes for the Arab people. The sion on local strengths as the source of poten- evidence clearly indicates that the poor— tial solutions. lacking the needed personal connections, the The World Bank Group stands ready to means to pay informal fees, or the luxury of contribute and help the governments and opting out from the public service system— people of the region promote inclusive insti- are the ones suffering the most from ineffec- tutions and create efficient service delivery in tive service delivery. order to better respond to citizens’ needs and In its search for solutions, the Bank team boost shared prosperity. has identified local examples of effective ser- vice delivery and shared the lessons learned. For instance, at a girls’ school in rural West Bank, a school principal has managed to Hafez Ghanem build a culture of inclusion and commit- Vice President ment among community members, parents, Middle East and North Africa Region and teachers. The success of the school has The World Bank Acknowledgments T his report is the product of the col- Jordanian governorates of ‘Ajlun, ‘Amman, laborative effort by a core team led by Al Balqa’, and Jarash; and a number of health Hana Brixi and composed of Ellen Lust clinics in Morocco. and Michael Woolcock (principal authors), This report benefited greatly from the as well as Jumana Alaref, Samira Halabi, feedback received during extended consul- Luciana Hebert, Hannah Linnemann, and tations. In particular, we are grateful to the Manal Quota. Brixi and Edouard Al Dahdah representatives of governments, academia, co-led the initial study design and coordi- nongovernmental organizations, and civil nated the early data analysis feeding into the society organizations who shared valuable background papers for this report. Those comments on the framework and findings papers were written by Hebert, Rohini Pande, of this report at the World Development and Quota. Alaref, Linnemann, Quota, and Report 2004 10th anniversary conference Woolcock developed the case study on edu- on “Making Services Work for Poor People: cation in Jordan, and Alaref, Osama Mimi, The Science of Politics of Delivery,” held in and Woolcock developed the second educa- Washington, DC, on March 1, 2014, and at tion case study, which is set in the West Bank. the Arab Development Symposium held in Melani Cammett, Cari Clark, Linnemann, Kuwait on November 5, 2014. We would also Lust, and Tamer Rabie delivered the case like to acknowledge the excellent comments study on health services in Jordan, and and suggestions received from participants in Cammett and Nejoua Balkaab produced the the Euro-Mediterranean Cycle of Economic one on health services in Morocco. Perspectives discussion series facilitated by the We sincerely appreciate the immense Center for Mediterranean Integration (CMI) hospitality and candor of the government in a seminar held in Paris on September 19, officials, health professionals, educators, par- 2014. Moreover, this report gained insights ents, and many other wonderful individuals from comments received from citizens of in various functions and roles who shared the Middle East and North Africa (MENA) their time and insights for the case studies: region in response to blog posts, online con- Kufor Quod Girls’ Secondary School in Jenin sultations, and social media outreach. in the West Bank; Jordan’s Zeid Bin Haritha Colleagues across the World Bank Group Secondary School in Yarqa; six clinics in the and members of the international academic xvii xviii ACKNOWLEDGMENTS and research community contributed valu- We thank our peer reviewers, including able insights in a series of workshops and Mustapha Kamel al-Sayyid, Nick Manning, brainstorming meetings held from October Hart Schafer, and Allen Schick at the con- 2013 to November 2014, including the cept note stage, and Melani Cammett, Sami May 9, 2014 workshop hosted by the Middle Hourani, Stuti Khemani, Allen Schick, and East Initiative at Harvard Kennedy School; Joel Turkewitz in the decision review, for and the November 19, 2014, seminar spon- excellent comments and suggestions, which sored by the Global Partnership for Social have made a difference. Accountability in collaboration with the Our special thanks go to Shanta Devarajan Governance Global Practice, Education for strategic guidance; to Enis Baris, Mourad Global Practice, and Health, Nutrition and Ezzine, Caroline Freund, Guenter Heidenhof, Population Global Practice of the World Bank Steen Jorgensen, Harry Patrinos, and Hisham Group. They included Lindsay Benstead, Waly for insightful discussions and feedback Franck Bousquet, Marylou Bradley, Angela along the way; and to Mario Marcel Cullell Demas, Deon Filmer, Helene Grandvoinnet, and Joel Hellman for advice and qual- April Harding, Elena Ianchovichina, ity assurance at the fi nal review stage. The Emmanuel Jimenez, Steen Jorgensen, Stuti work was supported by the MENA Chief Khemani, Elisabeth King, Pierre Landry, Economist’s Office, as well as the Governance Keit h M c L e a n , B a l a k r i sh n a Menon Global Practice, Education Global Practice, Parameswaran, Juan Manuel Moreno, Amr and Health, Nutrition and Population Moubarak, Stephen Ndegwa, Aakanksha Global Practice of the World Bank Group. Pande, Tamer Rabie, Halsey Rogers, Pia It was complemented by active engagement Schneider, and Joel Turkewitz. in governance and service delivery in Jordan The following also provided very use- (Multi-Donor Trust Fund) and Morocco ful inputs: Dina Abu Ghaida, Ali Ahmed (Institutional Development Fund) and in Al-Mudhawahi, Edouard Al Dahdah, Ghazia the Local Governance Performance Index Aslam, Robert Beschel, Jurgen Blum, Kamel (MENA Governance and Anti-Corruption Braham, Dorothee Chen, Ernesto Cuadra, Fund), which are building on and seeking to Heba Elgazzar, Lire Ersado, Amira Kazem, operationalize the approach developed in this Hans Lueders, Alex Mckenzie, Marcos report jointly with clients and partners. Mendiburu, Paul Prettitore, Namrata Moving forward, the Governance Global Saraogi, Roby Senderowitsch, Samira Practice, Education Global Practice, and Nikaein Towfighian, Maria Vagliasindi, Health, Nutrition and Population Global Jeffrey Waite, Lianqin Wang, Clara Welteke, Practice of the World Bank Group have agreed Jakob Wichmann, and Noah Yarrow. to use the framework and conclusions pre- Heba Shamseldin, together with Ghanimah sented in this report as a prism in their effort Al-Otaibi, Ashraf Al-Saeed, Naga El-Swais, to enhance education and health services in Lara Saade, and William Stebbins, assisted the MENA region (see Maximizing the World in the online consultations and outreach con- Bank’s Impact in the Middle East and North ducted for the report. Sabra Bissette Ledent Africa: Global Practices—Middle East and edited the report. Etsehiwot Albert and North Africa Region Partnership Report, Emma Etori provided administrative support. published by the World Bank in 2014). About the Authors and the Contributors Hana Brixi leads the World Bank’s Public Ellen Lust is Professor in the Department Service Delivery Global Solutions Group, of Political Science and Founding Director which fosters innovative context- specific of the Program on Governance and Local ser vice deliver y solutions for clients. Development at Yale University, and Non- Concurrently, she serves as Program Leader Resident Senior Fellow at the Project on for the Gulf countries. With the World Middle East Democracy. She received her Bank since 1995, Brixi has led innova- M.A. in Modern Middle East and North tive engagement on public fi nance and fi s- African Studies and PhD in Political Science cal risk management, governance, and from the University of Michigan, and has human development. She has authored the conducted fieldwork and implemented Promoting Growth with Equity report on surveys in the Arab Republic of Egypt, China; led the Bank-wide Quality of Fiscal Jordan, Libya, Morocco, West Bank and Adjustment Thematic Group; and published Gaza, Syrian Arab Republic, and Tunisia. several books, including Government at She was a founding associate editor of Risk (Oxford University Press), and numer- Middle East Law and Governance and cur- ous articles. Brixi has also served as WHO rently chairs its board of directors. Her books Health Sector Development Coordinator include Structuring Conflict in the Arab in China and as UNICEF Social Policy World; Political Participation in the Middle Chief in China, contributing to China’s East , co-edited with Saloua Zerhouni; policy reforms in health, social protection, Governing Africa’s Changing Societies , and service delivery and financing. She co-edited with Stephen Ndegwa; the 12th holds a graduate degree in economics and and 13th editions of The Middle East; and public policy from the Woodrow Wilson Taking to the Streets: The Transformation School, Princeton University; a postgradu- of Arab Activism , co-edited with Lina ate certificate in finance from the Sloan Khatib. She has also published articles Business School, Massachusetts Institute in Perspectives on Politics, Comparative of Technology (MIT); and a PhD in physics Political Studies, International Journal of from Masaryk University, Czech Republic. Middle East Studies, Politics and Society, xix xx ABOUT THE AUTHORS AND THE CONTRIBUTORS Comparative Politics, and other journals. Manal Quota is an Education Specialist More can be found at http://gld .commons in the World Bank’s Education Global .yale.edu/, http://transitionalgovernance- Practice. She specializes in educational project .org, and http://politicalscience.yale issues in the Middle East and North Africa, .edu/people/ellen-lust. with a focus on fragile and conflict-affected states addressing issues related to institu- M i c h a e l Wo o l c o c k i s L e a d S o c i a l tional resilience, education to employment Development Specialist in the World gaps, and improvement of service delivery in Bank’s Development Research Group, adverse situations. She has worked on coun- a nd a L e c t u rer i n P ubl ic Pol ic y at tries and territories such as Libya, Jordan, Harvard University’s Kennedy School of Syrian Arab Republic, and West Bank and Government. He is a founder of the World Gaza. Prior to joining the World Bank, Bank’s Justice for the Poor program, the Quota worked in the private sector in the University of Manchester’s Brooks World MENA region, consulting on projects rang- Poverty Institute (where he was in residence ing from development of a school feeding in 2007– 09), and the Harvard Kennedy program in the Republic of Yemen to admin- School’s Building State Capability initia- istration of a cluster of private schools. tive. His published work spans a broad Quota also worked in various international spectrum of issues in development, ranging organizations such as the International from social institutions, economic history, Criminal Court (ICC) in The Hague, the and legal pluralism to program evaluation, United Nations Development Programme service delivery, and popular culture. He (UNDP) in Riyadh, and the Organisation for is a recipient of the 2012 “best book” and Economic Co-operation and Development 2014 “best article” awards by the American (OECD) in Paris. She holds a master’s degree Sociological Association’s section on the in international education policy (Harvard sociology of development. An Australian University), a master’s in law and diplomacy national, Woolcock has an MA and PhD in (The Fletcher School, Tufts University), and sociology from Brown University. an undergraduate degree in political science with a specialization in international rela- Jumana Alaref is a Junior Professional tions (UCLA). Associate in the World Bank’s Social Protection and Labor Global Practice. Her Luciana (Lucy) Hebert is a Research work focuses on labor markets and educa- Specialist with Ci3 and the Section of Family tion-to-work transition through business Planning and Contraceptive Research at the creation. Alaref also works on cross-cutting University of Chicago. Her work primarily governance issues around transparency, focuses on sexual and reproductive health public accountability, and service delivery among adolescents and young adults in urban systems, and on connecting these issues to settings. She has conducted research on fam- sector-specific challenges, such as health ily planning, gender, partner dynamics, STI/ and education. Alaref’s country experience HIV risk perceptions, and sexual and repro- includes the West Bank and Gaza, Saudi ductive attitudes and behaviors among youth Arabia, and Tunisia. She holds a master’s in both domestic and international urban degree in public policy from the University contexts. She holds a PhD in public health of Chicago’s Harris School of Public Policy from Johns Hopkins University’s Bloomberg Studies and a bachelor’s degree in foreign School of Public Health. service from Georgetown University. Prior to joining the Bank, Alaref worked at the Hannah Linnemann is a consultant with National Opinion Research Center (NORC) the Human Development Unit of the at the University of Chicago, supporting Middle East and North Africa Region at impact evaluation projects worldwide. the World Bank. Linnemann joined the ABOUT THE AUTHORS AND THE CONTRIBUTORS xxi World Bank in 2013 as a Fellow of the Practice, where she provides analytical, German Carlo-Schmid-Program, which operational, and capacity-building sup- is supported by the German Academic port to improve the quality of education in Exchange Service (DAAD) and the German a range of countries. Currently, she leads National Academic Foundation. She holds work on strengthening accountability for an undergraduate degree in political sciences improved education services in Jordan. from the Free University of Berlin and is cur- Her other country experience includes rently pursuing a master’s degree in inter- Senegal, Nigeria, Iraq, Kuwait, the United national public management at the Paris Arab Emirates, and Afghanistan. Halabi School of International Affairs at Sciences has also served as Managing Editor of Po. Her research has focused on governance the Comparative Education Review, and and service delivery the Middle East and has co-published research on educational North Africa Region as well as aid effective- opportunities for at-risk youth. She holds a ness. Linnemann is a member of the German PhD in educational theory and policy from National Academic Foundation and holds the Pennsylvania State University, a master’s a Kuwait Excellence Scholarship from the degree in international education policy and Kuwait Program at Sciences Po. management from Vanderbilt University, and undergraduate degrees in business and Samira Halabi is an Education Specialist educational leadership from the American in the World Bank’s Education Global University of Beirut. Abbreviations AFR Africa EAP East Asia and Pacific ECA Europe and Central Asia EHGS Egypt Health and Governance Study ESSB établissement de soins de santé de base GAC Governance and Anti-Corruption GCC Gulf Cooperation Council GDP gross domestic product HCAC Health Care Accreditation Council HIV/AIDS human immunodeficiency virus/acquired immune deficiency syndrome HNP Health, Nutrition and Population ICR Implementation Completion Report IEG Independent Evaluation Group ISCED International Standard Classification of Education ISIS Islamic State of Iraq and Syria LAC Latin America and the Caribbean MENA Middle East and North Africa MP member of parliament OECD Organisation for Economic Co-operation and Development PETS Public Expenditure Tracking Survey PFM public fi nancial management PHC primary health care PHCC primary health care clinic PISA Programme for International Student Assessment PPP purchasing power parity PTA parent-teacher association SABER System Approach for Better Education Results SAR South Asia SEDGAP Secondary Education Development and Girls Access Project xxiii xxiv ABBRE VIATIONS SIAAP Service d’Infrastructures d’Action Ambulatoires Provincial (Provincial Service of Infrastructure and Ambulatory Care) TIMSS Trends in International Mathematics and Science Study UN United Nations UNESCO United Nations Educational, Scientific and Cultural Organization UNHCR United Nations High Commissioner for Refugees UNRWA United Nations Relief and Works Agency Overview T he Middle East and North Africa out to the community, inspiring citizens’ trust (MENA) is a rising middle-income and engagement through transparent and region, and its citizens rightly expect inclusive decision making and the delivery of quality public services. Yet too often they excellent services. experience disappointment: students attend- Learning from such local successes is vital ing local schools are insufficiently prepared because there are no blueprints for solving for the 21st century economy, and those service quality problems. Countries around needing health care too often find that public the world are striving to improve education clinics have no doctors or medicines. Few in and health care quality. But simply modern- positions of authority are held accountable izing school and hospital facilities and train- for such shortcomings. This situation both ing staff are no longer sufficient. Delivering undermines the potential for improvement quality services requires motivated staff. And and heightens people’s unhappiness with the staff motivation arises in turn from values delivery system. and accountability, which are grounded in Although dissatisfaction with educa- the wider political, administrative, and social tion and health services is widespread in the rules, practices, and relationships. Providing MENA region, local successes do exist and high-quality services is hard; the World Bank offer inspiration. At the Kufor Quod Girls’ itself has struggled to ensure that its projects Secondary School in the rural West Bank, enhance incentives in country systems to for example, Ms. Abla Habayeb, the school’s achieve better learning and health outcomes. principal, provides her teachers with daily Finding solutions is especially difficult in encouragement and support, and she involves the institutional and sociocultural contexts community members, parents, and teach- in the MENA region. Its citizens not only ers in decisions about improving the school. demand better services but also expect their Teachers, students, and the community then government to provide them, reflecting the reciprocate that commitment. Thus, amid the promises made by Arab leaders at the out- surrounding poverty and instability, Kufor set of independence five decades ago that Quod girls excel in national tests. Similarly, the state would provide better living condi- in some poor villages in Jordan and Morocco, tions. Since then, MENA citizens have gained the leaders of schools and clinics are reaching nearly universal access to education and 1 2 TRUST, VOICE, AND INCENTIVES health care, which is a commendable achieve- adequate human and material resources, ment. Now they expect these services to be of and its governments are perpetually engaged high quality and create greater opportunities in reform efforts aimed at better service for all. provision. And yet dissatisfaction among the Although they have excelled at building public runs so deep that failure to receive schools, constructing hospitals, and train- adequate services underpinned the calls ing staff, the region’s societies have fallen for karama (dignity) that echoed through- short in fostering the accountability and val- out the Arab Uprisings in 2010–11. Today, ues needed to motivate public servants and as the region continues to reel from deep service providers to deliver quality services. dissatisfaction—struggling with ongoing Whether and how teachers teach, doctors transitions, conflicts, and fragility—it is ever treat the sick, and bureaucrats react to citi- more critical to assess the status of service zens’ demands do not seem to matter in the delivery, recognize the underlying causes of selection of leaders (and their promotions and problems, and look for effective solutions. salary increases) or in the social acceptance The dissatisfaction with services is wide- of public servants. Rather than performance, spread. In the 2013 Gallup World Poll, what seems to matter most for political and on average about half of respondents in professional careers are personal relation- the MENA region, compared with about ships and social ties. This limits citizens’ 30 percent in Asia and Latin America and trust in and formal engagement with public the Caribbean, expressed their dissatisfaction institutions. To fulfill their needs, citizens with education services and health care in rely on their own relationships or informal their country. The 2010–11 Arab Barometer fees. Thus a cycle of poor performance has found that about two-thirds of MENA emerged, perpetuating a culture of privilege respondents perceived the performance of and cronyism. their government in improving basic health Improving the quality of public services services as “bad” or “very bad.” More will require breaking this cycle of poor broadly, citizens of the MENA countries tell performance: making politicians, public pollsters that their government should do servants, and providers accountable to citizens better in ensuring service delivery and fight- and promoting citizens’ trust in and engage- ing corruption. And yet they also express ment with state institutions. At the local little trust in their government’s involvement level, leaders and communities can inspire the in the social sectors. Moreover, they are less needed changes by demonstrating possible likely than citizens of other regions to seek local solutions and identifying the remaining accountability and tell public officials what systemwide constraints. Communities, states, they think (figure O.1). and donors that succeed in improving service quality and accountability will go a long way toward earning and retaining citizens’ trust. Local successes: Autonomy, accountability, and participation The average levels of citizens’ dissatisfac- Citizens’ expectations tion, however, mask significant variation The quality of social services provided in within countries and possible excellence in the Arab world lags surprisingly behind some localities. As we document here, some its potential. Historically, its citizens have communities have managed, often despite cared deeply about education, health, and difficult circumstances, to attain extraordi- other services; indeed, universal access to nary outcomes using innovative local solu- education and health is a constitutional right tions to the prevailing problems. These local in most countries in the Middle East and successes can provide useful insights and North Africa. The region is predominantly inspiration for practitioners, policy makers, composed of middle-income countries, with and donors. OVERVIEW 3 FIGURE O.1 Voiced opinion to public officials in the last month: MENA and other regions, 2013 100 90 80 70 60 Percent 50 40 30 20 10 0 Al p. Jo p. Ku p. n, m a Ar co Ye eba n en n Un t Ba Rep isia iA a ain A ria d an lic ic s Ba ia Qa t Sa L r W Ara Tu q A P R CD R C Ira ab E Gaz ta i am te ud iby L a m no e EN Re Re wa EA AF Ira EC b SA LA ite nk ub t, oc rd ,R ge Isl ira hr ra es b n OE M ab yp Mor Ar d Eg n ria Sy No Yes Source: Gallup World Poll, 2013. Note: AFR = Africa; EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin America and the Caribbean; MENA = Middle East and North Africa; OECD= Organisation for Economic Co-operation and Development; SAR = South Asia. Examples of local successes in service supervisors for Kufor Quod’s teachers, and delivery highlight the importance of auton- some autonomy in implementing school omy, accountability relationships, and par- improvement plans. ticipation at the local level. Although central Similarly, deep in Jordan the Zeid Bin management systems (such as the School Haritha Secondary School in the village of and District Development Program and the Yarqa has been achieving excellent results Health Care Accreditation Council in Jordan amid poverty and low capacity. Jordan’s or the Concours Qualité in Morocco) create national School and District Development environments conducive to providing quality Program, launched in 2009, encourages services, the impacts of such systems are not schools and directorates to collaborate with uniform. Our case studies uncover key drivers parents and communities, and it provides of change at the local level. They highlight the small school grants allowing some auton- importance of accountability relationships— omy. Involving parents and citizens as part- and the role of local leadership in sparking ners, however, has not come naturally in and institutionalizing such relationships—to traditional communities such as in Yarqa, trust and effective citizen engagement. with its deeply embedded lines of author- As noted earlier, the Kufor Quod Girls’ ity. It meant changing the leadership style Secondary School in a small village near of the school principal and teachers, creat- the city of Jenin in the West Bank traces ing a sense of common purpose around the its success to parental and community school in the community, and establishing engagement and the ability of the school’s new relations through a parent-teacher asso- principal to build and maintain a motivating, ciation and Education Council and making encouraging work environment for teachers. these new structures effective. Significantly, The principal has partnered with the Jenin the Education Council in Yarqa has reached school district to secure support, such as out to parents and the wider community the pedagogical support of the district and gained their trust for its transparency 4 TRUST, VOICE, AND INCENTIVES and inclusive decision making. Furthermore, quality and innovative performance improve- friendly competition and rivalry among local ments among health workers. These clinics schools and communities have helped to exhibited a sense of a shared mission and a improve student outcomes. Student perfor- collaborative ethic, supported by good record mance in national tests has become a source keeping, transparency, participatory training of community prestige and pride. Supporting workshops, and attention to interpersonal student learning—and addressing obstacles relations. The clinic staff regularly communi- such as the school’s supply of electricity and cates with the provincial and regional officials some students’ vision impairments—has of the Ministry of Health to ensure adequate become a shared responsibility. supplies of medications and to solve related In the health sector, top clinics across implementation problems. Moreover, clinic Jordan benefit from partnering with local health workers actively engage with nearby social institutions and health committees and social organizations and individuals—such as from formalizing health management pro- youth groups, the murshidat (women serving cedures at the local level. These steps have as religious guides), and women working in been facilitated by Jordan’s accreditation the local hammams (public baths)—to build process, which supports improvements in awareness of health issues such as family facility administration by establishing clear planning, HIV/AIDS testing, breast cancer, rules and regulations and supporting moni- and chronic diseases. toring and transparency. Local leadership and engagement enhance the impact of such administrative reforms. In the locations we The cycle of poor performance visited, health committees have reached out Notwithstanding such examples of local to both citizens and health workers in gather- successes, a majority of MENA citizens rou- ing and addressing community needs, some- tinely experience a cycle of poor performance times uncovering—and resolving—hidden in their daily lives: political, administrative, challenges such as previously undetected high and social institutions fail to instill adequate levels of diabetes and hypertension in the Zay accountability and motivation in policy mak- community. In some communities, social ties ers, public servants, and service providers to have facilitated the process of establishing meet citizens’ needs. As citizens experience priorities, extending public health outreach, poor service quality, they increasingly regard and mobilizing resources to support health the government as corrupt and ineffective. clinic activities and development. Thus their trust in public institutions suffers, Finally, the best-performing rural health leaving them with few options other than clinics in Morocco draw effectively on their turning to informal social networks and strong partnerships with local communi- paying informal fees to tackle their individ- ties, on positive competition devised by the ual needs. As illustrated by our case studies, Ministry of Health, as well as on support the cycle of performance (figure O.2) may be from the provincial and regional offices of virtuous at the local level—with local formal the Ministry of Health available to reform- and informal accountability relationships oriented local leaders. Launched in 2007, filling the institutional gaps, motivating bet- Morocco’s Concours Qualité program of ter performance, and inspiring citizens’ trust competition among health facilities, involv- and formal participation—but it appears to ing self-assessment and audit by peers, recog- be stuck in a low equilibrium at the national nizes good work and motivates improvement. level in most MENA countries. The MENA It has been especially effective in the pres- region’s historical development explains ence of dynamic and visionary leadership at the initial factors underlying this cycle. As the local, provincial, and regional levels. The noted, that development led citizens to place excellent clinics we visited have translated a high value on education, health, and other such effective leadership into a culture of services and created expectations—stronger OVERVIEW 5 than in other regions—that the state would FIGURE O.2 The cycle of performance provide these services, but the region largely eschewed the establishment of institutions geared toward meeting these expectations. Citizens’ engagement: formal and informal Without dependable institutions and citi- zens’ trust in public institutions, there is little formal citizen engagement, institu- tions remain stagnant, and service delivery is poor. A central concern is thus how can this cycle of poor performance be overcome? Institutions: Below, we examine how the prevailing cycle Cycle of political is structured and reinforced. performance at administrative, Citizens’ trust in the national and and social public institutions local levels Political, administrative, and social institutions: Lack of formal accountability Political institutions in most MENA coun- tries lack accountability mechanisms, with citizens unable to obtain adequate informa- Performance: effort and ability tion, voice demands, or incentivize policy to meet citizens’ needs makers and public servants through formal channels. Authoritarian regimes—which have the least constrained executives, the weakest parliaments, and the lowest levels Citizens have few opportunities to provide of judicial independence in the world— feedback on performance quality and to seek dominate the MENA region. The region accountability. Accountability institutions lags behind other regions in its transparency, such as justice sector services, independent objectivity, and professionalism in civil ser- audit agencies, and ombudsmen are under- vice appointments and management. Directly developed in the region, making it difficult elected local governments are found in only a for citizens to submit complaints, hold public small minority of countries in the region, and servants and service providers accountable, where they do exist, councils have limited and obtain their rights. Executive authori- budgets and responsibilities. With the excep- ties tend to exercise influence over the judicial tion of the Arab Republic of Egypt, Morocco, branch and over agencies designed to address and the West Bank and Gaza, local council corruption. Independent audit agencies, budgets are less than 5 percent of total public inspectors general, and ombudsmen tend expenditures, far behind the world average of to lack resources, authority, and autonomy. 38 percent for federal systems and 22 percent Political capture has promoted a system that for unitary ones. lacks the information needed to monitor and Opacity further undermines accountability. evaluate performances. As a result, there are Freedom of information and public disclo- few consequences when violations occur and sure laws and practices that would allow very few performance-oriented incentives citizens and intermediaries to monitor govern- and norms for providers and administrators. ment activities are either lacking or are not Weak, politically captured regimes are implemented. On the Global Integrity Index, coupled with—and compound—ineffective the region ranks the lowest on the public administrative systems and accountability access to information indicator, the legal right mechanisms. Administrative institutions to access information, and whether the right suffer from highly centralized and opaque of access to information is effective. bureaucracies and weak management 6 TRUST, VOICE, AND INCENTIVES systems, and service providers and public regularly produce or disclose criteria against servants are rarely held accountable. Local which their performance could be indepen- administrators and service providers have dently monitored. This is in part because little influence on policy formulation and of capacity weaknesses—that is, a lack of implementation, and they lack autonomy facility-level information and weak monitor- to manage human resources, make finan- ing and internal controls—and little enforce- cial decisions, enforce rules, or bring about ment and performance management. It also change on their own without the blessing reflects political capture at the national and of the central authority. At the same time, subnational levels that subverts incentives they face little oversight of and pressure on toward establishing accountability. their own performance. Information on Because state institutions lack both inter- the performance of frontline service pro- nal and external accountability, social norms viders, such as schools and health facilities and regulations within society and com- and their staffs, is generally not collected, munities can play a vital role in motivating evaluated, or followed up on. For example, policy makers, public servants, and service surveys in Egypt and Morocco suggest that providers. Social institutions emphasize obli- teachers appear to be minimally assessed, gations to members of social networks over and school inspections generate little action. national welfare. The result is the widespread With some exceptions (such as the United practice of wasta (figure O.3), a form of Arab Emirates) government agencies do not clientelism, as well as a willingness to treat FIGURE O.3 Importance of wasta in obtaining a government job, by municipality: Jordan, 2014 100 90 80 70 60 Percent 50 40 30 20 10 0 n A Bayba in l A an a gh ha a ad ra rn n n M ab ib ad o ra ss uda Al ah bia M Sh kh W ssa S za fa dis ra ha l R a Ka ayt za Al Al nje s ha a Al ssak Afrah W Ma hn h a A ’O ib sh GAl A kun Ha aribKara na im Ka Ass a i A fra uq Al alith a Sh ia s Ni ia 1 Al asi ara l M asi 1 a2 nn je 3 W di M sou 2 ad A S a 1 a A an ar Al am ba 2 a w Deir adis a 1 a K A As ia 2 ’A A am 1 Al Al ma 2 ik yb 1 al 3 Haith 1 Ka 1 A ou a 1 Kh Al l Ma sa 2 B am 2 M Th 2 ou 5 M Al 2 W Kaf Al Jo 3 i u 4 As Al Tqab ni A a i Al Mousa 4 l Al l B lad Al Syim b’i ir ma f i M hn f ba fr i Ta ari yn l B al fra Ra hm e fri da da uk A wd ut iri a m Th ayd A Al lat Ja Ka ashm Am am Su ad k u il Al Al Z Raj ow Qa na a ub a n a Sh nje Ta ra Th sa Al ia id l ak m i M usa a ad o e un h br as ’E Ra az a a h l s’a Anjunu a r s Q m B w W i M nj Umshm u ro y H n r Ar S l A M ka Al lJ sh As aA yb un Al Ta sh As Al Essential Very useful Moderate use Some use No use Source: Governance and Local Development survey, 2014. Note: The numbers next to some localities (municipalities) indicate that they are sublocalities or areas within them. The numbers were added for survey purposes; they are not official administrative boundaries. OVERVIEW 7 informal payments as a necessary practice. the local level. Surveys suggest that provid- Wasta allows individuals to obtain public ers may not possess the qualifications and services, jobs, and other economic opportuni- professional autonomy, among other things, ties, and preferential treatment when dealing needed to deliver quality services, par- with administrative procedures. In this way, ticularly in rural localities. They also often wasta can undermine fairness and equality of lack key resources such as teaching materi- opportunity as well as erode administrative als and medicines. Meanwhile, their efforts systems and overall state performance (as, for appear to be lagging: 30 percent of students example, when the recruitment and advance- in MENA countries attend schools in which ment of administrators and service providers principals reported that teacher absenteeism are based on wasta rather than merit). The is a serious problem (figure O.5). Similarly, strength of social institutions at the local health professionals exhibit high levels of level varies and is likely to explain some of dissatisfaction, and absenteeism surpasses the variation in performance, including edu- 30 percent in countries for which data cation and health services delivery, as well as exist: Egypt, Morocco, and the Republic of some of the elements of success in the local Yemen. Where observations exist, the adher- case studies described in this report. ence to curricula in schools and to clinical care protocols in health facilities appears low (figure O.6). Public sector teachers and Performance in education and health: health workers tend to offer some services The quality challenge and subnational as private efforts for a fee, which can create variation a conflict of interest. In Egypt, for example, E ducation and health outcomes have 89 percent of private physicians also work in improved in recent decades, but they have public facilities where they may be absent or not kept up with demands. In most MENA extend little effort during official hours while countries, outcomes such as school enroll- giving their best performance at their private ments or child mortality have converged practice. to their expected levels based on economic development. The quality of services has not, however, kept pace with the broader socio- FIGURE O.4 Percentage of respondents reporting payment of economic transitions. MENA students score informal fees, education and health care sectors: MENA region low on international competency tests, and and globally, 2013 graduates struggle to find jobs while employ- ers report vacancies unfilled due to skill 60 gaps. Health inequities based on income, gender, degree of urbanization, and age per- 50 sist. The out-of-pocket health expenditure is 40 high by international standards, leading to Percent 30 impoverishment and to forgoing health care because of its cost. Citizens typically find 20 little information publicly available about the 10 performance of schools and health facilities or about fees at health facilities. The lack of 0 p. . ria an co za sia ep q A al transparency can give rise to informal user Re Ira EN Ga oc ob rd ge ,R ni b M Tu Jo or en Al Gl nd ra M fees, which about one-third of citizens of m ,A ka Ye t an yp MENA countries have reportedly paid in the tB Eg es education sector and especially in the health W Education sector Health sector sector (figure O.4). Service delivery is characterized by the Source: Global Corruption Barometer, 2013. weak efforts and capacity of providers at Note: MENA = Middle East and North Africa. 8 TRUST, VOICE, AND INCENTIVES FIGURE O.5 Percentage of students whose principals report that teacher absenteeism is a serious problem in their school: MENA region and globally, 2011 60 56 50 40 40 33 Percent 30 27 26 24 22 20 18 13 12 10 7 7 5 2 1 0 s ic co sia a an an za r ain n CD CD p. A ta te bi no bl EN Re Ga oc rd Om ni Qa ira OE OE hr ra pu ba M Tu Jo or ic iA Ba d Em n- Re am Le M n ud No ka ab b Isl Sa ra an Ar n, nA tB Ira d ite ria es W Un Sy Source: TIMSS, 2011. Note: MENA = Middle East and North Africa; OECD = Organisation for Economic Co-operation and Development. FIGURE O.6 Adherence to care protocols for diabetes and coronary heart disease in health facilities: Arab Republic of Egypt, 2010 Take blood pressure in the arm Examine the chest Examine the abdomen while the patient is in supine position Weigh the patient Examine feet or any wound or pressure points Examine the back of the thorax with a stethoscope Examine the patient for sensations and reflexes Measure the blood sugar with stick Measure the circumference of the waist Conduct an eye exam using an ophthalmoscope Measure the circumference of the hips 0 20 40 60 80 100 120 % of chronic care observations CHD Diabetes Source: EHGS, 2010. Note: CHD = coronary heart disease. OVERVIEW 9 The quality of service provision varies sig- citizens and demonstrated by the available nificantly within countries, in part because of evidence can be traced to the weaknesses in the weak national political and administra- the effort and capacity of providers. Such tive institutions and the resulting influence weaknesses in turn reflect the characteristics of social institutions and local governance of the prevailing political, administrative, practices. Indeed, the results in education and social institutions, especially the weak (such as student test scores) and the quality accountability mechanisms facing policy indicators in health (such as the adherence to makers, public servants, and service provid- care protocols) exhibit significant subnational ers. The lack of simple monitoring and inter- variation. nal controls undermines the distribution of Service delivery process indicators vary textbooks to schools and medicines to health as well within countries. Examples are staff facilities. The interplay of formal and infor- absenteeism and qualifications and the avail- mal pressures and norms at the national ability of instructional materials, essential and local levels influences the efforts of medicines, and other key inputs (figure O.7). providers. The patterns across service delivery indica- tors also vary, showing little correlation, for Citizens’ trust and engagement: example, between staff effort and the avail- Shaped by service delivery and state ability of key inputs. This may indicate that performance staff, materials, and other inputs in service delivery are managed in an insular manner, Citizens’ experiences when they visit a health along the vertical administrative lines in the facility or observe their child’s learning, generally heavily centralized service delivery when they seek a job or deal with admin- systems in the MENA countries. istrative procedures, affect not only their The poor quality of education and health view of performance but also their attitude care as perceived by the majority of MENA toward the state. The low satisfaction with FIGURE O.7 Availability of aspirin, by province: Morocco, 2011 25 120 100 20 80 15 Number Percent 60 10 40 5 20 0 0 ho an nm a sa Be sick ich fa ua ar ne M ch Ra s i b Set t Al elaa ra z en r Ou Ag ra m a To a oh Fée Iar ate er tat Tiz i El nit Na ia El dor Ta da am s Ou i me t c e a a ir ra J sset Ta jad za l em d l s ba Kh ge ne Er ou n rir -te ad ar Be eim h ar lla ta an an us jd ad Ch a-an id if Kh ma ke ao Al sult M Ifra ac i z jad B e ou n ek ha te er ch n no k bl rk i ra c s er -m Sid m hi da ay Sk Ca fi -h Al aa eb ns Ai % Facilities stocked out Number of facilities Number of weeks stocked out (mean) Source: PETS (health), Morocco, 2011. 10 TRUST, VOICE, AND INCENTIVES public services, perceived corruption and and holding them accountable), and voting nepotism, and, indirectly, unresponsive insti- (choosing political leaders committed to tutions appears to erode citizens’ trust in ensuring better services). In the face of weak public institutions in many MENA countries. institutions, poor performance, and low trust Across countries, fi ndings from the Gallup in public institutions, citizens tend to disen- World Poll, Arab Barometer, and Worldwide gage. In much of the region, people believe Governance Indicators reveal a high cor- they have little chance of succeeding by sim- relation between citizens’ satisfaction with ply following the rules. In surveys in Algeria service provision and indicators based on and Morocco, for example, only a quarter underlying measures of state institutional of respondents considered it effective to seek quality and performance such as government services or fi le complaints directly through effectiveness, rule of law, and control of cor- the relevant government agency; higher num- ruption. Furthermore, the data suggest that bers of respondents believed it was more trust in national government is highly associ- effective to go through family, friends, and ated with citizens’ satisfaction with education other social ties. In the Republic of Yemen, and health services and with their percep- only 10 percent of citizens who believed they tions of the pervasiveness of corruption. For had a valid reason to make a complaint actu- those MENA countries for which data are ally did. Furthermore, civil society organiza- available, our analysis suggests that the tions are subjected to state intervention and probability of trusting the national govern- crackdowns in most MENA countries, which ment increases by 13 percentage points when limit their effectiveness and may explain why respondents are satisfied with education and citizens rarely join such organizations. by 11 percentage points when they are sat- When citizens need services from the state, isfied with health care. On the other hand, try to resolve complaints and grievances, or citizens’ trust in public institutions declines have to deal with administrative procedures, by 35 percentage points when respondents they often do so through informal channels, believe that corruption is widespread within resorting to survival mechanisms such as their governments. These results also hold wasta or informal payments or, more rarely, across MENA countries when measuring demonstrations or rebellions. Unfortunately, trust in the judiciary. With respect to nepo- in doing so they exacerbate the existing tism, respondents are 10 percentage points problems, eroding formal accountability and likelier to trust their national government norms of public service, undermining pub- when they believe that qualifications are more lic welfare, and widening the inequality of important than connections. Interestingly, opportunities. In more extreme cases, such though, we found that the influence of tribal as the Houthis’ rebellion in the Republic of affinities has only a small impact on citizens’ Yemen, stemming in part from the Houthis’ trust in national government. unresolved grievances related to poor public Low trust in public institutions—and services, they may unravel postconflict insti- perceived powerlessness over the decision- tutions and trust building. There are, how- making processes—undermines citizen- ever, instances—as our examples of local driven, bottom-up approaches and social successes illustrate—in which citizens’ trust accountability tools (such as public hearings, and engagement can be inspired by local community scorecards, public opinion polls, leaders and can support improvements in and civil society oversight committees) that service quality. Finally, the rising popularity could be used to improve service delivery. of social media and their use by citizens to Institutions in most MENA countries offer share their experience and demands, partly their citizens few opportunities to encour- facilitated by successful e-governance initia- age better service delivery through choice tives in several MENA countries, are opening (selecting better providers), voice (giving potentially effective engagement avenues for feedback to providers and public servants the future. OVERVIEW 11 Bases for improvement gain citizens’ acquiescence to reform despite poor performance. Such opportunities may Because of the complex circumstances facing be short-lived, however, and the extent MENA countries, it is necessary to build on to which they can be exploited is likely to evidence of local successes and positive trends depend on a number of factors such as the that show where and how the cycle of gen- strength of state institutions, the degree of erally poor performance can be challenged. polarization within society, and levels of We seek to identify the bases for improve- regional or international intervention. But if ment and encouragement so that citizens, citizens see improvements, they may remain civil servants, policy makers, and donors engaged, thereby positively motivating ser- can act on them. Many policy makers across vice providers and public servants and sup- the MENA countries want to deliver visible porting statewide reforms and institution results and, in doing so, bolster their author- building. ity and public support. Conflicts, crises, and It is therefore critical to seize the oppor- political transitions in the region may give tunities offered in crises to buoy service pro- national and local leaders a unique opportu- vision, press for institutional reforms, and nity to reform institutions and accountability foster citizens’ trust and positive engagement. mechanisms and tackle service delivery chal- Even in the midst of enormous difficul- lenges, as well as boost citizens’ trust in pub- ties that citizens and states face, there is an lic institutions and constructive engagement. opportunity to escape the cycle of poor per- Donors, including the World Bank Group, as formance. Preparing to face these challenges well as governments and civil service organi- and seeking ways to open new opportuni- zations, need to learn from their own (often ties for breaking the cycle of poor perfor- failed) efforts to support quality in education mance require a clear understanding of how and health services. In moving forward, the the international community, local policy MENA countries can explore possible incre- makers, civil society, and citizens can work mental approaches to systemic reforms, the together to improve the quality of public options for empowering communities and services provision. local leaders to find local solutions, and pos- sible quick wins. Donor intervention Regardless of such extraordinary shocks, Extraordinary shocks donors have been only partly effective in sup- The extraordinary shocks spreading from porting education and health care quality in North Africa to the Arabian Peninsula—as the MENA countries. Over the last 10 years, unsettling, costly, and risky as they are— the World Bank Group has been increasingly can open possibilities for reform. Confl icts, highlighting the role of incentives and citi- refugee crises, and regime changes can make zen engagement in achieving equitable and national and local leaders more likely to take high-quality services. In the MENA region, risks, disrupt existing institutional arrange- this emphasis has figured prominently in the ments, and alter elite coalitions. Costly Bank’s analytical and reimbursable advi- events today may prompt changes in insti- sory activities but has been less pronounced tutions, trust, and engagement that result in operations. Only about 10 percent of the in better performance. For example, tran- Bank’s projects in the MENA countries sitions triggered by dramatic ruptures can (compared with about 30 percent on average potentially, at least in the short run, increase in other regions) have promoted autonomy trust, engagement, and the possibilities of and accountability in education (figure O.8) institutional reforms. Elites often experi- or transparency and accountability in health ence a brief honeymoon period with much services delivery. This low rate—along with of the population, during which they can a relatively modest success rate of Bank 12 TRUST, VOICE, AND INCENTIVES FIGURE O.8 SABER analysis of World Bank School Autonomy and Accountability Interventions, by region, FY03–13 80 70 60 % of all operations 50 40 30 20 10 0 AF R EAP ECA LAC MENA SAR Budget autonomy Autonomy in personnel management Role of school councils (participation) Assessment Accountability Source: World Bank forthcoming. Note: SABER = Systems Approach for Better Education Results; AFR = Africa; EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin America and the Caribbean; MENA = Middle East and North Africa; SAR = South Asia. operations supporting education and health of a problem and developing a hypothesis services delivery in MENA countries—partly while being agnostic about the solution; by reflects the difficulty in developing interven- using evidence to inform the implementa- tions to address the cycle of performance tion of solutions; by taking an adaptable, beyond capital investments and capacity creative, and context-driven approach; and building and to motivate better performance by being able to capture cumulative knowl- by policy makers, public servants, and service edge when finding and fitting local solutions. providers and inspire citizens’ trust and con- Furthermore, the World Bank’s initiative on structive engagement. social accountability in the MENA region If programs, systems, and civil service appears to be a promising approach, even organizations supported by donors are to if its implementation requires time, money, be trusted by citizens, donors may need to and expertise in order to appropriately reach out to citizens directly to build trust. engage with local contexts. Examples of Simple measures such as providing them positive deviance—such as the Secondary with detailed information on donor support, Education Development and Girls Access objectives, interventions, and cost, disag- Project (SEDGAP) in the Republic of Yemen, gregated to the village level, would be a step which is seeking to engage community lead- forward and a possible model for promoting ers, parents, girls, and female teachers— transparency domestically. offer valuable lessons on how donors can Experience suggests that the usual focus help. Investing in the high-quality collection on identifying policy reform needs must be of local data can pay handsome dividends matched by a corresponding focus on how in terms of enhancing the efficiency and any given policy will actually be imple- effectiveness of everyday decision making, mented, by whom, and why. One might hope providing an evidence base for promot- that the new “science of delivery” approach ing organizational learning, and helping to provides a space within which such issues identify where, when, and how such positive can be explored by looking at the nature deviance occurs. OVERVIEW 13 An incremental approach to systemic approach to institutional and policy reforms changes, local initiatives, and can combine considerations about feasibil- quick wins ity and political support with considerations about possible solutions. Such an approach Improvements in education and health ser- also can result in design reforms that align vices will not come simply through policy more closely with the existing reform space reforms, through modernization of schools and thereby gradually expand the space for and health facilities, or through training reform. of educators and health professionals. Our Drawing on the available evidence, we set analysis indicates that, to foster better perfor- forth two broad sets of recommendations. mance, policy reforms and investments need First, we argue for a stronger social contract the backing of institutions—especially incen- among public servants, citizens, and providers. tives and norms embedded in both formal and To that end, we propose an approach of informal accountability relationships—and strategic incrementalism toward improv- citizens’ trust and engagement. Experience ing institutional quality and accountability suggests that performance improves when mechanisms and motivating public servants political institutions are the primary driv- and providers to serve the poor and other ers of outcomes, or—as our case studies nonprivileged populations. In particular, we illustrate—when skillful leaders use them to explore options to tap into and exploit social institutions for better outcomes. Decentralization, incorpo- • Develop effective external accountability rated in a broad package of reforms aimed at institutions such as courts, independent putting more power into the hands of local auditors, and ombudsmen to monitor— officials, can help strengthen incentives for and subject to public scrutiny—the perfor- better performance if supported by adequate mance of service providers and public ser- accountability mechanisms and resources. vants and provide tools for the resolution The evidence on citizen engagement shows of citizens’ complaints and grievances. that information is necessary but not suf- • Strengthen monitoring, internal controls, ficient to motivate collective action, to make and performance management in the pub- local or central officials accountable, and to lic sector as well as within the education influence public sector performance. More and health systems, including mechanisms promising results emerge from multipronged to share and act on performance informa- strategies that encourage enabling environ- tion, exposing service providers and pub- ments for collective action and bolster a state’s lic servants to internal accountability for capacity to actually respond to the voices of performance. its citizens. An ability to respond to citizens’ • Modify mechanisms for selecting, encourag- feedback on the quality of service delivery is ing, and rewarding leaders, public servants, crucial to sustaining trust and participation. and service providers so as to internalize Meanwhile, to have an impact, institu- norms of personal responsibility, profes- tional and policy reforms must emerge from sional accountability, and public service. problem-led learning processes, facilitate • Learn from intracountry variations to the “fi nding and fitting” of context-specific design solutions that fit local contexts, solutions, and seek the participation of evaluate and strengthen policy imple- broad groups to ensure that new institu- mentation, inform citizens, energize local tions are shared, legitimate, and contextually leaders, and scale up local successes. appropriate. Reforms also need to recognize Second, we call for empowering com- the actual incentives that prevail for stake- munities and local leaders to find “best- holders associated with a specific problem fit” solutions in motivating educators and in a specific setting. Within the existing health professionals and in harnessing social constraints, an incremental, problem-driven accountability to inspire trust and empower 14 TRUST, VOICE, AND INCENTIVES citizens to act. Possible options for govern- institutions, performance, and citizens’ trust ments include measures to and engagement. We hope that this report and its recommendations will help citizens, • Build coalitions among champions of civil servants, policy makers, and donors reforms in government, civil society, and alike jointly identify and build on the present the private sector to improve service deliv- foundation to improve the delivery of social ery, giving local actors space to engage in services, shifting the cycle of performance into piloting possible solutions. a virtuous gear. An improved cycle of per- • Systematically collect feedback on public formance is what those living in the MENA services from users, benchmark service countries deserve and what would enable them delivery and local governance perfor- to fulfill their hopes and dreams for the future. mance, and disseminate information on performance to provide a rigorous basis for citizen action. Reference • Close the feedback loop among citizens, World Bank. Forthcoming. World Bank Support service providers, civil society organi- for School Autonomy and Accountability zations, and the private sector in order FY03-13. Washington, DC: World Bank. to strengthen the coherence of policy development; provide a foundation for Data sources prioritizing problems and possible solu- Arab Barometer, http://www.arabbarometer.org tions (with an emphasis on the most dis- /instruments-and-data-files. advantaged, poor, and vulnerable); more EHGS (Egypt Health and Governance Study), equitably allocate fi nite public resources World Bank, http://documents.worldbank across the local and national levels; and .org /curated /en /2010/06/16332545/egypt improve implementation effectiveness by - m a n a g e m e nt- s e r v i c e - qu a l it y- p r i m a r y enhancing mutual accountability and the -health-care-facilities-alexandria-menoufia dissemination of useful information on -governorates. performance standards. Gallup World Poll, http://www.gallup.com /services/170945/world-poll.aspx. In addition, quick wins are needed— Global Corruption Barometer, Transparency especially in countries in transition or emerg- International, http://www.transparency.org ing from a conflict and fragility—to gain and /research/gcb/overview. retain trust and make the cycle of performance Global Integrity Index (no longer published), virtuous. Quick wins could come in any form Global Integrity, https://www.globalintegrity .org/. of improvements observable by citizens. The Governance and Local Development survey, state, for example, could reach out to iden- Prog ra m on G over na nce a nd L o c a l tify and popularize local successes through Development, Yale University, http://gld the media; hold public awareness campaigns .commons.yale.edu/research/. on citizens’ rights, service delivery standards, PETS (Public Expenditure Tracking Survey), and anticorruption; conduct solution-focused Wo r l d B a n k , h t t p : / / w e b .w o r l d b a n k public meetings to address problems such as .org / W BSI T E / EXT ER NA L / TOPICS absenteeism and material shortages in service / EXTSOCIALDEVELOPMENT/EXTPCENG delivery; expand opportunities for citizens’ /0,,contentMDK:20507700~pagePK:148956 engagement; and demand a response to citi- ~piPK:216618~theSitePK:410306,00.html. zens’ feedback. Citizens’ experience with the SABER (Systems Approach for Better Education Results), World Bank, http://saber.worldbank state and with service delivery can improve .org/index.cfm. more rapidly with the appropriate use of infor- TIMSS (Trends in International Mathematics mation communication technology. and Science Study), Boston College, http:// We argue that because of the complex timssandpirls.bc.edu/. circumstances found in MENA countries, Worldwide Governance Indicators, World Bank, it is necessary to build on evidence of local ht t p: //data.worldba n k.org /data- catalog successes and positive trends at the level of /worldwide-governance-indicators. Introduction T he Middle East and North Africa Development Report 2004 approach to (MENA) region is at a critical juncture, ”making services work” by exploring how with complicated transitions and tragic political, administrative, and social institu- conflicts on one side and a tremendous poten- tions affect provider responsiveness. It also tial based on rich human and natural resources explores how a state’s performance shapes on the other. The majority of the population in citizens’ trust and the nature of their engage- the region is lacking economic opportunities, ment at the national and local levels, which in facing inequalities, demanding social injustice, turn further influence institutional develop- and expressing frustration and mistrust. In ment and performance (World Bank 2004). parts of the region, children, women, and other Focusing on education and health, this vulnerable population groups are suffering report illustrates how the weak external and from extreme insecurity or even open conflict, internal accountability relationships prevalent struggling for survival and in need of essential in the MENA political and administrative services. Meanwhile, political uncertainties spheres undermine incentives toward policy and poor state performance are exacerbating implementation and performance, and how tensions in the society, deepening mistrust, such a cycle of poor performance can be coun- and discouraging citizens from engaging with teracted. According to the evidence, weak the state. With visionary leadership, the right accountability as well as low satisfaction with policies, and inclusive institutions, this vicious public services negatively affects citizens’ trust cycle can be broken and a renewed social con- in the state. Low trust in public institutions tract can be established allowing citizens to explains in part why citizens seldom engage receive better services. A renewed social con- with the state and service providers through tract and better service delivery would also formal channels, relying instead on wasta empower young women and men to realize (personal relationships), informal payments, their aspirations and potential and build a and other survival mechanisms. Case stud- brighter future for the next generation. ies of local successes reveal the importance This report examines the role of trust, of both formal and informal accountability incentives, and enagement as critical deter- relationships and the role of local leadership minants of service delivery performance in in inspiring and institutionalizing incentives the MENA countries. It expands the World toward better service delivery. 15 16 TRUST, VOICE, AND INCENTIVES Even more broadly, enhancing the services such services, but states fail to meet citizens’ received by MENA citizens requires forg- expectations and overcome the service qual- ing a stronger social contract among public ity challenge, particularly in the public sector. servants, citizens, and service providers, as That said, examples of local successes can well as empowering communities and local be found, demonstrating that better realities leaders to fi nd “best-fit” solutions. Lessons are possible. Why is this? Why does service learned from the variations within countries, delivery fall short of potential in the MENA especially the outstanding local successes, region? can serve as a solid basis for new ideas and We argue that the answer to this question inspiration for improving service delivery. lies in the cycle of poor performance that has Such lessons may help the World Bank Group emerged in much of the region (figure I.1). and other donors, as well as national and Institutions are a useful starting point for local leaders and civil society, develop ways understanding this cycle. In the MENA to enhance the trust, voice, and incentives for region, state institutions lack both internal service delivery to meet citizens’ needs and and external accountability, in part because expectations. of the shortage of information on perfor- mance that is needed to guide centralized decisions and in part because of the lack of Organization of this report incentives toward establishing accountability This report explores the perceptions, reali- mechanisms for performance in public sector ties, governance determinants, and possible services delivery. When institutions are weak, solutions for education and health services service delivery policies are not successfully delivery in the MENA region.1 The report is implemented. As Acemoglu and Robinson divided into five parts. (2012, 78) argue, “The low education level In part I, we describe citizens’ perceptions of poor countries is caused by economic insti- and expectations as well as the realities and tutions that fail to create incentives for par- local successes in social services delivery in ents to educate their children and by political the MENA region. First, we summarize institutions that fail to induce the government insights from surveys about citizens’ atti- to build, finance, and support schools and the tudes toward education and health services, wishes of parents and children.” their expectations about the government’s Based on their experiences with poor- role in providing those services, and how quality service delivery, as outlined in citizens try to fulfill their needs. Second, chapters 1 and 2, citizens perceive govern- drawing on the literature and databases, we ments as corrupt and ineffective. Not only briefly review the impressive achievements do public services seem captured by public of the MENA countries in expanding access servants and local elites with limited account- to basic education and health services and ability to citizens, but a large share of private improving core human development out- services also appear captured by the same comes over the last five decades. We then public servants as part of their employment in highlight the remaining challenge of poor both the public and private sectors. This prob- service quality and citizens’ dissatisfaction. lem is not evident in many of the Gulf states Finally, we examine cases of local success in because an abundance of resources allows which schools and health facilities are per- these states to provide high-quality services. forming far beyond expectations under very But even there, service provision falls short of difficult circumstances. the promise these resources should provide. Overall, we find that the MENA coun- Low trust in institutions undermines tries are not meeting their potential in pro- bottom-up pressures for improving service viding citizens with education and health delivery. Citizens can provide incentives for services. Citizens demand quality services and public service delivery through choice (using believe the state is responsible for delivering public services), voice (giving feedback to INTRODUCTION 17 providers), and voting (choosing political FIGURE I.1 The cycle of performance leaders who support service delivery systems). However, in the face of weak institutions, poor performance, and low trust people Citizens’ engagement: formal and informal often disengage. They turn instead to local nonstate actors and institutions for services. When they do demand services from the state, citizens tend to do so through informal channels and seek piecemeal, selective solu- tions to their individual problems. Institutions: Circumventing the state perpetuates insti- Cycle of political tutional weakness. When citizens walk away performance at administrative, Citizens’ trust in and social from public services or fail to give feedback, the national and public institutions local levels state actors lack the information they need to improve institutions. When political competi- tion is weak, there is little pressure to develop better state solutions. Indeed, the importance of political engagement to developing insti- tutions was clear in postcommunist eastern Europe. There, formal administrative institu- Performance: effort and ability tions and rules were established and enforced to meet citizens’ needs in countries that had the vibrant political competition needed to pressure governments for reform (Grzymala-Busse 2010). Cycles vary, of course, driven by differ- and extent of subnational variation in service ences in the available resources, societal com- delivery performance (chapter 7). The sub- position, and the agency/leadership of key national variation analysis underscores the actors. And they are alterable. As shown in message of chapter 3 that local successes can chapter 3, a virtuous cycle can develop at the be found and that much about service deliv- local level (even in context of a poor perfor- ery challenges and possible solutions can be mance at the national level) when local stake- learned in local contexts. holders are driven by individual will or social Completing the cycle of performance, in obligations to take initiatives. part IV we discuss how institutions and per- In part II, chapter 4 explains how histori- formance affect citizens’ perceptions of the cal experience has led citizens to value health state and the nature of citizen action vis-à-vis and education, fostered their dependence on the state. In particular, we seek to reveal how the state, and has limited state responsive- performance influences citizens’ trust in the ness. Chapter 5 provides a detailed picture of state (chapter 8), and how this trust in turn the political, administrative, and social insti- shapes the nature of citizens’ engagement at tutions that affect service delivery. both the local and national levels (chapter 9). In part III, we turn our attention to per- Because of the complex circumstances formance at the point of service delivery: we facing MENA countries, it is necessary to explore the efforts and abilities of teachers build on evidence of local successes and on and health professionals and the availability positive trends that buck the cycle of gen- of key inputs such as instructional materials erally poor performance. In this respect, in schools and medicines in health facilities. in part V, chapters 10, 11, and 12 identify We also discuss how these efforts and avail- the bases for improvement and encourage- ability are affected by institutions. Drawing ment so that citizens, civil servants, policy on surveys, we focus first on the national makers, and donors can act on them. We level (chapter 6) and then explore the nature acknowledge that many policy makers 18 TRUST, VOICE, AND INCENTIVES across the MENA countries want to deliver as enhance the political, administrative, and visible results and, in doing so, bolster their social institutions that affect state perfor- authority and public support. As we discuss mance. To devise effective solutions, these in chapter 10, confl icts, crises, and politi- efforts can build on the local success stories cal transitions give rise to a new dynamic, found in MENA countries. which may present national and local lead- ers with a unique opportunity to reform Note institutions and accountability mechanisms and to tackle service delivery challenges as 1. Because of differences in the transliteration of Arabic, place-names often differ slightly well as boost citizens’ trust and constructive from one source to another. The versions engagement. The cycle of poor performance of place-names appearing in this report are can also be altered by reforms in the absence largely those used in the various surveys cited of such major disruptions. That depends or those used by the World Bank’s country on appropriate learning by and incentives offices. from international donors (discussed in chapter 11), or incremental institutional References reforms and local reform coalitions in soci- ety and government (chapter 12), or a com- Acemoglu, D., and J. A. Robinson. 2012. Why bination of these factors. Thus by arguing Nations Fail: The Origins of Power, Prosperity that much of the MENA region is stuck in and Poverty. New York: Crown Publishers. Grzymala-Busse, A. 2010. “The Best Laid Plans: a low-equilibrium cycle of performance, we The Impact of Informal Rules on Formal are not suggesting that citizens are doomed Institutions in Transitional Regimes.” Studies to weak institutions, poor service delivery, in Comparative International Development dissatisfaction, and suboptimal engagement. 45 (3): 311–33. Rather, we are suggesting that efforts to World Bank. 2004. World Development Report reform service provision in the region should 2004: Making Services Work for Poor People. foster citizens’ trust and engagement as well Washington, DC: World Bank. Expectations, Reality, and Inspiration in Education and Health Services Delivery I n part I, we explore citizens’ perceptions and expectations of social services delivery in the Middle East and North Africa (MENA) as well as the realities and local successes. In chapter 1, we summarize insights from the available polls and surveys about citizens’ attitudes toward education and health services, their expecta- tions about the government’s role in providing those services, and their attempts to pursue their needs. Then, drawing on the available literature and databases, in chapter 2 we briefly review the impressive achievements of MENA countries in expanding access to basic education and health services and improving core human development outcomes over the last five decades. We also highlight the remaining challenge of poor service quality and citizen dissatisfaction. Finally, in chapter 3 we examine cases of “positive deviance” and offer four case studies of schools and health facilities performing far above expectations in very difficult circumstances. 19 MAP I.1 Human Development Index (HDI) values for MENA, 2013 IBRD 41533 | MARCH 2015 B l a c k S e a Ca s pi an Sea Mediterranean Sea ATLANTIC Me SYRI AN TUNISIA MALTA dite OCEAN rranean Sea LEBANON A.R. IRAQ ISLAMIC REP. MO ROCC O WEST BANK AND GAZA OF IRAN JORDAN KUWAIT A L G E R I A L I B Y A ARAB REP. BAHRAIN OF EGYPT SAUDI QATAR UNITED ARABIA ARAB EMIRATES Human Development OM AN Re Index (HDI), 2013 d Se a More than 0.8 0.71 – 0.8 REP. OF YEMEN 0.51 – 0.7 en Arabian Ad MENA average 0.72 f of Sea Gul 0.5 and less DJIBOUTI Non-MENA/non-OECD average 0.64 No data available OECD average 0.87 Data source: Human Development Index, United Nations Development Programme (UNDP). Source: World Bank (IBRD 41533, March 2015). A Demand for Better Services but Not Formal Accountability 1 • Universal access to education and health is a constitutional right in most countries in the Middle East and North Africa (MENA), and their citizens place a high value on education and health care. • Citizens of the MENA region are not satisfied with the existing services and with their government’s efforts to improve service delivery. • Citizens expect their government to play a strong role in providing welfare, but they rely mainly on informal mechanisms in pursuing their individual needs. S ocial services typically constitute a role in ensuring the provision and quality of critical interface between citizens and these services. By contrast, citizens are gener- their state, shaping in turn citizens’ ally dissatisfied with the available education expectations of the state, the level of trust and health services and their government’s they place in it, and the manner in which they efforts to improve these services, and they engage with it. These relationships are tend to rely on informal mechanisms in explored in detail in this report. In this chap- demanding accountabilit y for service ter, we mainly approach education and health delivery. services through the eyes of the citizens of the Middle East and North Africa (MENA) region, recognizing that citizens’ perceptions The right to education and and expectations reflect not only their cur- health services in the MENA rent experiences and realities but also the region promises and hopes of the past. Many countries in the Middle East and This chapter especially highlights the great North Africa consider the delivery of social emphasis placed by MENA citizens on edu- services—education and health—to be the cation and health and on the government’s state’s responsibility and the right of all 21 22 TRUST, VOICE, AND INCENTIVES citizens to universal access to be enshrined in the expansion in education and health has their country’s constitution and law. Since continued and so have expectations of a the Arab Uprisings in 2010–11, the revised strong role for government in these sectors constitutions in countries such as the Arab and in providing for citizens’ welfare. Republic of Egypt, Morocco, and Tunisia Paradoxically, many countries in the region have further strengthened these rights and have implemented the postindependence the state’s obligation to ensure their protec- social contract by drawing on administrative tion and fulfillment. The constitutions of all institutions established under colonization, MENA countries set education as a right for which has limited the effectiveness of policy all citizens and, with the exception of implementation and of state action (Cammett Morocco, mandate that basic education is 2013; Issawi 2013). 2 The historical under- compulsory and free for youth. Similarly, the pinnings of the current service delivery right of all citizens to health care is to some challenges are described in more detail in extent enshrined in the constitutions of all chapter 4. MENA countries except for Djibouti, Jordan, The Arab people place a high value on and Lebanon. Furthermore, as part of the education and health services along with global movement to ensure that all people employment opportunities in the context of obtain the health services they need without economic prosperity. Much like the rest of suffering financial hardship when paying for the world, for them economic growth is a top them (so-called universal health coverage), priority. In the World Values Survey, Egypt, Morocco, Tunisia, and the Gulf 2010 –14, the majority of respondents in Cooperation Council (GCC) countries1 are Egypt, Iraq, Jordan, Kuwait, Libya, Morocco, striving to increase health coverage as a right Qatar, the Republic of Yemen, Tunisia, and for all citizens. the West Bank and Gaza identified a high The pronounced right of citizens to educa- level of economic growth as the current top tion and health services and the state’s priority of their economies. In Algeria and responsibility in delivering them are rooted Lebanon, citizens gave priority in part to in the political economy of Arab indepen- economic growth and in part to defense and dence following World War II. With indepen- citizen engagement (figure 1.1). dence, Arab leaders sought to break away Against this backdrop, citizens give prior- from the earlier elitist policies and promised ity to education and health care, together with their people industrialization and better job opportunities. As a 16-year-old Palestinian living standards through state intervention. boy from Hebron put it, “Of course if some- That intervention included massive expan- one wants to find a respectable job, he has to sion of education and health services as well get educated,” or as an adolescent girl from as food and fuel subsidies and employment in Rafah said, “Education is a girl’s best weapon the public sector. It gave rise to a social to face the world” (Krishnan et al. 2012). In contract binding government and citizens to the My World survey organized by the United establish political legitimacy and strong Nations, MENA citizens—much like citizens nationhood and instilling in citizens support elsewhere—reported that education and for government as well as the expectation health care were among their top priorities that the state would assume responsibility for (figure 1.2).3 A recent opinion poll in Egypt economic and social welfare. In the region’s found similar results (figure 1.3). These results oil-rich countries, oil revenues further aug- were consistent with those of earlier surveys, mented citizens’ expectations that the state such as the poll conducted by Zogby would play a big role in the provision of International in 2005 (Attitudes of Arabs, services and welfare. Although the arrival of 2005). liberalization and marketization in the region Moreover, MENA citizens expect educa- du ring the 1980s par tly undermined tion and health services to be delivered well. the state’s provision of these services, They view the ability of governments to A DEMAND FOR BE T TER SER VICES BUT NOT FORMAL ACCOUNTABILIT Y 23 “deliver goods” as a key aspect of gover- FIGURE 1.1 Prioritizing values: MENA and other regions, 2010–14 nance. In this respect, they are no different from other people (Bratton 2010; Leavy and 90 Howa rd 2013; B erg h, Menocal, and 80 Takeuchi 2014). This is also illustrated by the 70 political economy of the recent uprisings in 60 Percent the region. These uprisings pushed to the 50 forefront the embedded economic and social 40 grievances, driven largely by deep dissatisfac- 30 tion with the provision of public services and 20 the lack of employment opportunities, as well 10 0 as with the signs of political capture and . ep Al on Jo co M ria an Q n A nd ar yp Tu za ia en . A R P M R C m ep da EN EC EA crony capitalism (Walker and Tucker 2011; AF SA t A nis LA k a at oc ,R Ga ge n Ye ab R r ba or Le Cammett and Diwan 2013).4 r tB Eg es W Citizens’ satisfaction and A high level of economic growth Making sure there are demands on government strong defense forces Seeing that people have more Trying to make our cities and On average, a large share of MENA citizens say about how things are rural areas more beautiful indicate in surveys that they are not satisfied done at their jobs and in their communities with education and health services. For example, the 2013 Gallup World Poll, which Source: World Values Survey, 2010–14 (Wave 6). covers citizens and Arab-speaking residents Note: Algeria, 2014; Arab Republic of Egypt, 2012; Iraq, 2013; Jordan, 2014; Kuwait, 2013; Lebanon, 2013; Libya, 2013; Morocco, 2011; Qatar, 2010; Republic of Yemen, 2013; Tunisia, 2013; across countries, found on average nearly West Bank and Gaza, 2013. AFR = Africa; EAP = East Asia and Pacific; ECA = Europe and Central Asia; 40 percent of respondents dissatisfied with LAC = Latin America and the Caribbean; MENA = Middle East and North Africa; SAR = South Asia. the education services in their country (figure 1.4) and 45 percent dissatisfied with the avail- ability of quality health care (figure 1.5). On FIGURE 1.2 Citizens’ priorities among services: Various regions, 2014 the positive end, respondents in Bahrain, Oman, Qatar, and the United Arab Emirates 80 reported relatively high levels of satisfaction 70 with both education and health services. By contrast, the majority of citizens in Egypt, 60 Iraq, Morocco, and the Republic of Yemen % of votes received were dissatisfied. Overall in the MENA 50 region, levels of satisfaction with services 40 across sectors appear lower than those of other regions, except for Africa (figure 1.6). 30 In some MENA countries, citizens are 20 concerned about corruption in education and health care. According to the Global 10 Corruption Barometer, the majority of citi- zens in Algeria, Egypt, Lebanon, Morocco, 0 MENA ECA EAP LAC AFR SAR OECD and the Republic of Yemen perceive the edu- A good education Better health care Better job opportunities cation and health systems in their countries Honest and responsive Protection against to b e cor r upt or ex t remely cor r upt government violence and crime (figure 1.7). In some countries, respondents complained about being “constantly exposed Source: My World Survey, 2014. Note: AFR = Africa; EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin America to corruption, favoritism, poor customer and the Caribbean; MENA = Middle East and North Africa; OECD = Organisation for Economic service, and deficient information” when Co-operation and Development; SAR = South Asia. 24 TRUST, VOICE, AND INCENTIVES FIGURE 1.3 Top two issues in need of the greatest investment: Arab Republic of Egypt, 2011 Education programs Health care Private sector/business development Support for local governorates Infrastructure projects Democracy programs 0 10 20 30 40 50 60 70 Percent Source: Egypt Public Opinion Survey, April 14–27, 2011. FIGURE 1.4 Satisfaction with education services: MENA and other regions, 2013 100 90 80 70 60 Percent 50 40 30 20 10 0 yp or . ra co p. q a an ep p. nd lic Tu a Ku ia Jo it Sa Alg n i A ia Le bia Om n an Ba tes ain A A Em tar CD R R P C Eg M Rep n A lam by z a no Ira wa EC EN EA AF SA s ud er LA Re tB R e Ga k a ub t, A oc rd ni ira ab Qa hr es rab ic R ra OE ria Is Li ba M , b en m Ye Ar Sy n, Ira d ite W Un Dissatisfied Satisfied Source: Gallup World Poll, 2013. Note: AFR = Africa; EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin America and the Caribbean; MENA = Middle East and North Africa; OECD = Organisation for Economic Co-operation and Development; SAR = South Asia. dealing with public services and the bureau- Perceptions Index, 2014)—see figure 1.8. In cracy (World Bank 2007). The perceived addition, figure 1.9 reveals that MENA GCC prevalence of corruption varies across MENA respondents were on average more satisfied countries. Scores on public sector corruption with their government’s effort to fight cor- indicate low prevalence in the GCC countries ruption than those in other regions and non- when compared with Iraq, Libya, the GCC MENA countries (Gallup World Poll, Republic of Yemen, and Syria (Corruption 2013). A majority of respondents in Lebanon A DEMAND FOR BE T TER SER VICES BUT NOT FORMAL ACCOUNTABILIT Y 25 FIGURE 1.5 Satisfaction with the availability of quality health care: MENA and other regions, 2013 100 90 80 70 60 Percent 50 40 30 20 10 0 M ep. co an ep a nd lic Al za n, Tu ria am sia Le ep. iA n Ku ia Jo it Om n Ar Ba an Em ain s r R A A R C P CD ta te tB R y es rab Lib . ud no a wa ra aq EA AF SA LA b EC EN p Ga k a ub oc rd ge Isl ni Qa ira ab hr ,R R ra Re OE Sa ba t, A Ir M or ic en b m Ye nA yp Ira d Eg ite ria W Un Sy Dissatisfied Satisfied Source: Gallup World Poll, 2013. Note: AFR = Africa; EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin America and the Caribbean; MENA = Middle East and North Africa; OECD = Organisation for Economic Co-operation and Development; SAR = South Asia. FIGURE 1.6 Satisfaction with services across sectors: MENA and other regions, 2013 AFR 90 80 70 Roads/highways OECD 60 EAP 50 Quality of water 40 30 20 Quality of public transportation 10 0 Availability of quality health care SAR ECA Education services Availability of good, affordable housing MENA LAC Source: Gallup World Poll, 2013. Note: The figure shows percentages. AFR = Africa; EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin America and the Caribbean; MENA = Middle East and North Africa; OECD = Organisation for Economic Co-operation and Development; SAR = South Asia. 26 TRUST, VOICE, AND INCENTIVES FIGURE 1.7 Education and health systems—corrupt or extremely corrupt: Selected MENA economies, 2013 90 80 70 60 Percent 50 40 30 20 10 0 Morocco Egypt, Lebanon Yemen, Algeria Global Tunisia Iraq West Bank Arab Rep. Rep. and Gaza Education system Health system Source: Global Corruption Barometer, 2013. Note: MENA = Middle East and North Africa. FIGURE 1.8 Public sector corruption score: Selected MENA economies, 2014 100 90 80 70 60 Percent 50 40 30 20 10 0 q ya . am lic p. n ti p. co sia it an ud an a ain ab atar s t, A ria ep te bi no Ira ou wa Re Re b Lib oc Om rd ni ge ira hr ,R ra pu Q ba ib Ku Tu Jo or ic b iA Ba en Al Em Dj Re ra Le M m ab Isl Ye Sa yp Ar Ar n, Eg Ira d n ite ria Un Sy Source: Corruption Perceptions Index, 2014. Note: Score: 0 (highly corrupt) to 100 (very clean). MENA = Middle East and North Africa. (80 percent), Iraq (74 percent), Syria (62 per- and education in households. In most MENA cent), Egypt (55 percent), and the West Bank countries, higher-income respondents and and Gaza (52 percent), however, responded females tend to be more satisfied with educa- that government was not doing enough to tion and health services than lower-income fight corruption. respondents. More educated respondents Within countries, satisfaction with ser- tend to be more satisfied with health care but vices appears associated with levels of income less satisfied with education (Gallup World A DEMAND FOR BE T TER SER VICES BUT NOT FORMAL ACCOUNTABILIT Y 27 FIGURE 1.9 Perceptions of government’s efforts to fight corruption: MENA and other regions, 2013 100 90 80 70 60 50 Percent 40 30 20 10 0 n Eg R aq t B ra lic n p. m aza M ep. Al o ria Tu a Jo a an i A it Ba ia ain Em tar s A C CD P R R A te y si EA c AF SA no LA ud a EC EN b k a Re es , A b Lib oc rd r ge Sa Kuw ni ab Qa OE ira hr ,R ra I Ye d G W ypt epu ba M or an b en Le b ra Ar nA d ite ria Un Sy No, not doing enough Yes, doing enough Source: Gallup World Poll, 2013. Note: AFR = Africa; EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin America and the Caribbean; MENA = Middle East and North Africa; OECD = Organisation for Economic Co-operation and Development; SAR = South Asia. Poll, 2013).5 These differences in part reflect Reflecting the postindependence social differences in access (such as in the ability to contract, the Arab people expect their gov- pay formal or informal fees) and information ernments to take more responsibility for (including knowledge of objective perfor- their overall welfare and engage more effec- mance measures and evidence). “You need to tively in improving services. Nearly half of pay the staff to get good service,” said some respondents in the selected MENA countries citizens in recent surveys (World Bank 2013), believed their governments should take more which may imply that, as one Moroccan responsibility to ensure that everyone is pro- health professional stated, those able to pay vided for, whereas less than 10 percent benefit from services more (Allin, Davaki, believed that people should instead provide and Mossialos 2006). for themselves (as seen in figure 1.11 by con- In the broader socioeconomic context, trasting those responding 1 or 2 with those MENA citizens tend to express concern responding 9 or 10 in the given range). about exclusion and clientelism. Especially As figure 1.11 shows, the Arab people expect in the developing MENA countries, respon- significantly more government action on dents emphasized the need for pro-poor behalf of citizens’ welfare than do citizens policies and expressed dissatisfaction with of other regions. Furthermore, an over- efforts to deal with the poor (figure 1.10). whelming majorit y of respondents in By contrast, respondents in the GCC Lebanon (90 percent), Iraq (79 percent), countries, especially in Kuwait, Qatar, and Algeria (77 percent), the Republic of Yemen the United Arab Emirates, were largely sat- (76 percent), Egypt (67 percent), Tunisia isfied with their government’s effort to (65 percent), Jordan (60 percent), and the address poverty. West Bank and Gaza (57 percent) evaluated 28 TRUST, VOICE, AND INCENTIVES FIGURE 1.10 Satisfaction with efforts to deal with the poor: MENA and other regions, 2013 100 90 80 70 60 Percent 50 40 30 20 10 0 Le raq k a on en a Al p. Eg M ria ria , A cco Re p. Jo lic Tu n sia ud ya Ba bia ain Em it Qa s r R A A R C P D ta te m az EA AF SA a LA a EC EN C e e b Sa Lib rd ge Ar Kuw n ni OE ira hr ,R ra b R ra I Ye d G Sy ypt oro pu ba M iA n A ra n ab b an tB d ite es W Un Dissatisfied Satisfied Source: Gallup World Poll, 2013. Note: AFR = Africa; EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin America and the Caribbean; MENA = Middle East and North Africa; OECD = Organisation for Economic Co-operation and Development; SAR = South Asia. FIGURE 1.11 On a continuum of 1 to 10, citizens’ expectations of their government: MENA and other regions, 2010–14 40 35 30 25 Percent 20 15 10 5 0 The 2 3 4 5 6 7 8 9 People government should take should take for more responsibility responsibility to provide for to ensure themselves that everyone is provided for MENA ECA LAC AFR SAR EAP Source: World Values Survey, 2010-14 (Wave 6). Note: AFR = Africa; EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin America and the Caribbean; MENA = Middle East and North Africa; SAR = South Asia. A DEMAND FOR BE T TER SER VICES BUT NOT FORMAL ACCOUNTABILIT Y 29 FIGURE 1.12 Perceptions of performance of current government in improving basic health services: Selected MENA economies, 2010–11 100 90 80 70 60 Percent 50 40 30 20 10 0 a za sia p. an q ria . n ep bi no Ira Re Ga rd ge ni ,R ra ba Tu Jo b iA Al nd en ra Le ud m ka t, A Ye Sa an yp tB Eg es W Very bad Bad Good Very good Source: Arab Barometer, 2010–11 (Wave II). Note: Survey includes only citizens. MENA - Middle East and North Africa. their government’s performance in improv- FIGURE 1.13 Citizens’ perceptions of service delivery ing basic health services as either bad or very responsibility: Tunisia, 2014 bad (figure 1.12). For education and health as well as secu- 100 90 rity, citizens’ perceptions of government 80 70 responsibility mainly pertain to the central Percent 60 government. These perceptions largely reflect 50 40 the high extent of the centralization of 30 20 authority in the education and health systems 10 (as discussed in chapter 5). However, the 0 n n es s y ity ed nd io ad cit division of responsibilities is often not trans- tio iti ur at Ro ry y a tri lec cil c uc Se c fa da ar parent. In a survey in Tunisia, for example, Ele ol on im h ec alt Pr ag many citizens believed the local government He rb Ga c to be responsible for services that were under se the purview of the central government, and Local councils National government almost 25 percent of respondents in the sur- Source: Transitional Governance Project, 2014. vey were unable or unwilling to answer (figure 1.13). The next chapter discusses in more detail the realities of and the elements of Citizens’ trust and engagement citizens’ satisfaction with education and In view of the high value that MENA citizens health services. Chapters 8 and 9 explore place on education and health services and how citizens’ perceptions of service delivery their perceptions of state responsibility in and state performance affect their trust in ensuring the provision and quality of these public institutions and its different levels of services, it is not surprising that citizens government and the nature of their engage- exhibit higher trust in public institutions in ment with the state. countries—and, at the subnational level, in 30 TRUST, VOICE, AND INCENTIVES localities—where they are more satisfied with local governance program) illustrated the education or health care. Among those sur- importance of the perceived informal and veyed, a large share of respondents in Jordan, formal accountability relationships in gov- Kuwait, and Qatar expressed trust in their ernment and service delivery and the signifi- national government (with rates above cant role of the local customary institutions 80 percent), whereas less than half of respon- in constructing accountability relationships dents in Iraq, Lebanon, and the West Bank between citizens and the state, acting in and Gaza expressed trust in their govern- essence as intermediaries between them. ments (Gallup World Poll, 2013). Finally, in contrast to their high expecta- Globally, there seems to be a positive tions of state action, MENA citizens tend to correlation between trust in government and forgo formal engagement with the state, satisfaction with services (see figures 1.14 and including ad ministrators and ser vice 1.15). This relationship seems stronger across providers. Compared with the citizens of MENA countries, which may reflect the high other regions, those of the MENA region are demand that MENA citizens place on their the least likely to voice their opinions to public governments to ensure service provision officials (figure 1.16). As chapter 9 explains, (Gallup World Poll, 2013). the citizens of MENA countries are not likely Qualitative analyses indicate that apart to seek accountability for service delivery from the citizens’ experience with service through formal channels. Instead, anecdotal delivery, it is often the observed account- evidence suggests that it is not uncommon for ability relationships that influence their citizens to use their social networks to reach views of the state and their trust in govern- out to ministry officials or members of parlia- ment. Fieldwork on local governance and ment to voice a complaint or request state service delivery in the Republic of Yemen intervention such as capital investments in a (conducted in 2014 as part of the Yemeni given school or health facility. FIGURE 1.14 Satisfaction with education services and trust in government: MENA and other regions, 2013 100 Qatar % of respondents who trust their Kuwait 80 Iran, Islamic Rep. Jordan national government Egypt, Arab Rep. Tunisia Djibouti 60 Yemen, Rep. 40 Algeria Iraq West Bank and Gaza 20 Lebanon 0 20 40 60 80 100 % of respondents satisfied with education services 95% CI Fitted values AFR OECD ECA LAC MENA SAR EAP Source: Gallup World Poll, 2013. Note: CI = confidence interval; AFR = Africa; EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin America and the Caribbean; MENA = Middle East and North Africa; OECD = Organisation for Economic Co-operation and Development; SAR = South Asia. A DEMAND FOR BE T TER SER VICES BUT NOT FORMAL ACCOUNTABILIT Y 31 FIGURE 1.15 Satisfaction with health services and trust in government: MENA and other regions, 2013 100 Iran, Islamic Rep. Egypt, Arab Rep. Qatar Kuwait % of respondents who trust their 80 national government Djibouti Jordan 60 Tunisia Yemen, Rep. 40 20 Algeria Iraq Lebanon West Bank and Gaza 0 20 40 60 80 100 % of respondents satisfied with health services 95% CI Fitted values AFR OECD ECA LAC MENA SAR EAP Source: Gallup World Poll, 2013. Note: CI = confidence interval; AFR = Africa; EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin America and the Caribbean; MENA = Middle East and North Africa; OECD = Organisation for Economic Co-operation and Development; SAR = South Asia. FIGURE 1.16 Voiced an opinion to a public official in the last month: MENA and other regions, 2013 100 90 80 70 60 Percent 50 40 30 20 10 0 ra cco Jo . Ye eba n en n Al p. ria T q Un Ba ep sia d an lic n, m a ic s Ku p. it Sa L r iA a Ba ia ain A R A P C R CD p ta am te Ira ab E Gaz ud iby EA SA AF L a m no Ira LA wa EN EC b Re e Re ite nk ub rd ge es rab uni Qa OE Isl ira hr ,R ra t, A ro M b Ar d yp o Eg M R nA t ria W Sy No Yes Source: Gallup World Poll, 2013. Note: AFR = Africa; EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin America and the Caribbean; MENA = Middle East and North Africa; OECD= Organisation for Economic Co-operation and Development; SAR = South Asia. 32 TRUST, VOICE, AND INCENTIVES This chapter has looked briefly at citizens’ developments that brought about the perspectives on education and health ser- protests. In particular, the perceived inequal- vices in the MENA region. This report seeks ity of opportunity, underscored by the crony to explain why MENA citizens approach capitalism and political connections shaping economic opportunities in the region, was a education and health services, and the state, central concern. in the way they do and what the service 5. The reported results are based on two logis- delivery realities and challenges are. In the tic regressions performed using the 2013 next chapter, we present evidence on the Gallup World Poll data. The first equation state of education and health services deliv- lists satisfaction with education as a depen- ery in the MENA region. dent variable (binary variable coded as 1 = satisfied, 0 = dissatisfied) with the follow- ing explanatory variables: age, education, Notes employment status, gender, and income quintile. The second equation lists satisfac- 1. The GCC countries are Bahrain, Kuwait, tion with the availability of quality health Oman, Qatar, Saudi Arabia, and the United care as a dependent variable (binary variable Arab Emirates. coded as 1 = satisfied, 0 = dissatisfied) with 2. Arab states emerged from the era of coloni- the following explanatory variables: age, zation with varied administrative capacities education, employment status, gender, and and competence. Egypt and Tunisia, for income quintile. example, entered independence with strong functioning national administrations, while other countries such as Libya and Lebanon had weak administrative penetration and References bureaucracies with little experience, making Allin, S., K. Davaki, and E. Mossialos. 2006. full territorial control difficult. Other states “Paying for ‘Free’ Healthcare: The Conundrum were still struggling with the formation of of Informal Payments in Post-Communist social structures and economies within the Europe.” In Global Corruption Report, arbitrary territorial lines drawn by the 62–71. Berlin: Transparency International. Europeans. This was an acute problem for Bergh, G., A. R. Menocal, and L. R. Takeuchi. the nomadic tribal families in Jordan, Iraq, 2014. “W hat’s Behind the Demand for and the Gulf. Governance?” Overseas Development Institute 3. My World, an anonymous online survey, asks (ODI), London. respondents around the world to identify what Bratton, M. 2010. “The Uses of the Afrobarometer is most important to their families. Among the in Promoting Democratic Governance.” MENA economies, the sample includes Briefing Paper 93, Afrobarometer. http://www Algeria, Bahrain, Egypt, Jordan, Lebanon, .afrobarometer.org/. Morocco, Oman, Qatar, the Republic of C a m m e t t , M . 2 013. “ D e velopm e nt a nd Yemen, Saudi Arabia, Sudan, the Syrian Arab Underdevelopment in the Middle East and Republic, Tunisia, the United Arab Emirates, North Africa.” http://www.melanicammett and the West Bank and Gaza. Online voting .net / w p - content /uploads / 2012 / 03/OU P was open until 2015. Figure 1.2 reflects the _MENA_v1.3_MC1013.pdf. votes received until July 3, 2014. Cammett, M. C., and I. Diwan. 2013. The 4. In “After the Arab Spring: The Uphill Struggle Political Economy of the Arab Uprisings. New for Democracy,” Walker and Tucker (2011) York: Perseus Books Group. offer a comprehensive review of the grim and Issawi, C. 2013. An Economic History of the deteriorating conditions in the run-up to the Middle East and North Africa. Abingdon, Arab Spring. In their book The Political Oxford, U.K.: Routledge. Economy of the Arab Uprisings, Cammett Krishnan, N., T. Vishwanath, A. Thumala, and and Diwan (2013) provide an overview of the P. Petesch. 2012. Aspirations on Hold? Young important developments across the Arab Lives in the West Bank and Gaza. Washington, world prior to the Arab Uprisings. They DC: World Bank. argue that it was the interaction of political Leavy, J., and J. Howard. 2013. What Matters factors and real and perceived economic Most? Evidence from 84 Participatory Studies A DEMAND FOR BE T TER SER VICES BUT NOT FORMAL ACCOUNTABILIT Y 33 with Those Living with Extreme Poverty and Attitudes of Arabs: An In-Depth Look at Social Marginalisation. http://www.ids.ac.uk/files and Political Concerns of Arabs, 2005, Zogby /d m f i le / Pa r t i c ip at e Sy nt he si s -W h at International for Arab American Institute, MattersMost.pdf. http://www.aaiusa.org Walker, C., and V. Tucker. 2011. “After the Corruption Perceptions Index, Transparency A rab Spr i ng: T he Uph i l l St r ug g le for International, http://www.transparency.org Democracy.” Freedom House, Washington, /research/cpi/overview DC. https://freedomhouse .org/report/algeria Egypt Public Opinion Survey, International /overview-essay#.VI37pb5N3Hg. Republican Institute, http://www.iri.org World Bank. 2007. Decentralization and Local /MENA/PublicOpinionPolls Governance in MENA: A Survey of Policies, Gallup World Poll, http://www.gallup.com Institutions, and Practices. A Review of /services/170945/world-poll.aspx Decentralization Experiences in Eight Middle Global Corruption Barometer, Transparency East and North Africa Countries. Washington, International, http://www.transparency.org DC: World Bank. /research/gcb/overview ———. 2013. Fairness and Accountability: My World Survey, United Nations, http://vote Engaging in Health Systems in the Middle .myworld2015.org/ East and North Africa. Washington, DC: Tra nsit iona l G over na nce Projec t , ht t p: // World Bank. transitionalgovernanceproject.org World Development Indicators (database), World Bank, http://data.worldbank.org/data-catalog /world-development-indicators Data sources World Values Survey, 2010–14 (Wave 6), http:// Arab Barometer, 2010–11 (Wave II), http://www w w w.w o r l d v a l u e s s u r v e y. o r g / W V S .arabbarometer.org/instruments-and-data-files DocumentationWV6.isp The State of Education and Health Services Delivery: The Quality Challenge amid Impressive Advances • Education and health outcomes have improved over the decades, gradually approaching the levels expected based on economic development. • Although citizens have gained access to services, their needs have shifted with the economic, demographic, and epidemiological transitions. Conflicts and refugee crises have added extra pressure. • The private provision of services has grown, expanding the choices for some citizens. • Service quality has become the main challenge. A s discussed in chapter 1, the citizens in the MENA region, the advancements of the Middle East and North Africa achieved over the past decades have been (MENA) region value education and applied almost uniformly, except for the health services and expect the government to unfortunate local reversals due to conflicts. ensure their delivery and quality. This chapter briefly summarizes the region’s impressive Historical advancement advances in education and health outcomes and in promoting access to essential services. Over the last five decades, the citizens of The analysis highlights service quality as the MENA countries have seen their education key challenge facing MENA countries, in and health improve rapidly. As governments part reflecting the shifting needs driven by across the MENA region have invested in the region’s economic, demographic, and building a wide network of schools, hospitals, epidemiological transitions. and clinics and in training teachers and health professionals, citizens have rapidly gained access to essential services. Indeed, on several Outcomes fronts they have seen their education and Although education and health outcomes dif- health rapidly improving in absolute terms fer significantly across and within countries and relative to other countries at a comparable 35 36 TRUST, VOICE, AND INCENTIVES stage of economic development. Although sig- between 1990 and 2011, compared with an nificant differences exist within as well as increase from 65 to 70 years during the same across countries in the region, education and time period worldwide. Premature death and health indicators in MENA countries have disability from most communicable, new- largely converged to the levels expected based born, nutritional, and maternal causes with on income per capita (figure 2.1). the exception of HIV/AIDS have decreased Education and health outcomes look significantly and faster than expected based encouraging on several fronts. Life expec- on the growth in income per capita (Mokdad tancy at birth increased from 64 to 72 years et al. 2014). These improvements have been driven in part by successes in selected areas of public health such as immunization FIGURE 2.1 Performance of MENA countries in education and (figure 2.2). In education, primary, second- health: Females who completed primary school and child mortality ary, and tertiary enrollment and completion based on income per capita, 1975 and 2011 rates have risen rapidly and achieved parity for girls and boys. Net enrollment rates for a. Females who completed primary school primary education have approached levels 120 seen in member countries of the Organisation for Economic Co-operation and Development % of girls who completed primary school 100 (OECD)—see figure 2.3. And the average returns to primary education in the MENA 80 region have been estimated as the highest in the world, especially for females (Montenegro 60 and Patrinos 2014). 2011 40 Recent shifts in needs 20 1975 The economic, demographic, and epidemio- logical transitions experienced during recent 0 decades have shifted the demands on service Log of GDP per capita (PPP-adjusted $) delivery. The region’s economies and labor b. Child mortality based on income per capita markets are increasingly dependent on the 400 creation of private sector jobs amid rapid 350 population growth. This situation has altered the skills needed and the demands on the Number of children 0–5 years 300 1975 region’s education and training systems. As dying per 1,000 born 250 for health, the MENA region’s middle- and high-income countries are experiencing rapid 200 2011 changes in lifestyles, bringing about major 150 shifts in the disease burden. Education and training systems in the 100 MENA region are finding it difficult to meet 50 the shifting labor market needs. These sys- 0 tems emerged as part of the postindepen- Log of GDP per capita (PPP-adjusted $) dence social contract, which included an World, 2011 MENA countries, 2011 emphasis on public sector employment and World,1975 MENA countries,1975 on the role of education and training systems Exponential (world, 2011) Exponential (world, 1975) in preparing youth for public sector jobs. In recent decades, however, public sector jobs Source: World Development Indicators (database). Note: MENA = Middle East and North Africa; GDP = gross domestic product; PPP = purchasing have become costly to generate for an power parity. expanding pool of applicants, driven by high T H E S TAT E O F E D U C AT I O N A N D H E A LT H S E R V I C E S D E L I V E RY 37 population growth, the increasing participa- FIGURE 2.2 Percentage of children aged 12–23 months immunized tion of females in the labor force, and rising for measles: MENA and OECD countries, selected years, 1975–2010 labor force qualifications. Moreover, the for- mal private sector has been curtailed in some 100 MENA countries by crony capitalism and a 90 poor business climate (Diwan, Keeefer, and 80 Schiffbauer 2014; Rijkers, Freund, and 70 60 Nucifora 2014). The result is that there are Percent 50 few “desirable” jobs—too few, in fact, to 40 motivate the education and training systems 30 to adopt reforms that would produce skills 20 applicable to the private sector. 10 These trends have contributed to an ero- 0 sion in the value of the skills currently pro- ra tes A CD iA . a Re q Om c an Ar Jo r yp m n Ba ep. Ku in it ya ud ep ta i bi Eg ab E rda b Ira wa EN bl a Lib Qa t, A ira OE R Sa n, R hr ra pu M duced by the region’s education and training b e m systems. In contrast to primary education, Ye ra nA d the average returns to secondary and tertiary ite ria Un Sy education, especially for males in the MENA 1975 1990 2000 2010 region, are currently estimated to be the lowest among all regions (Montenegro and Source: World Development Indicators (database). Note: MENA = Middle East and North Africa; OECD = Organisation for Economic Co-operation and Patrinos 2014). Moreover, the MENA region Development. has the world’s highest unemployment rate among youth, and particularly among edu- FIGURE 2.3 Net enrollment rates (adjusted) for primary education: cated youth and young females. For young MENA and OECD countries, selected years, 1971–2010 men and especially women, the transition from education to work is longer in the 100 MENA region than in any other region (ILO 2004). Although private sector job creation is a 80 key problem, MENA’s employment challenge also stems in part from the skill gaps and 60 mismatches generated through the education Percent and training system. On average, one-third of unemployed youth are university graduates, 40 even though enterprises report being unable to fill vacancies. About 40 percent of employ- ers in the MENA region identify skill mis- 20 matches as a major constraint to doing business and growth (Gatti et al. 2013). 0 Although further economic development and 1971 1975 1985 1990 1995 2000 2005 2010 employment in MENA countries increasingly MENA total OECD total rely on job creation in the private sector, the Source: World Development Indicators (database). education systems have a legacy of preparing Note: MENA = Middle East and North Africa; OECD = Organisation for Economic Co-operation and youth for public sector jobs. Development. In health, the region is experiencing a rap- idly growing burden of noncommunicable increased, especially ischemic heart disease, diseases, posing new demands on the health stroke, mental disorders, musculoskeletal system. According to Mokdad et al. (2014), disorders, and chronic kidney disease between 1990 and 2010 the disease burden (figure 2.4). The epidemiological profi le of from many noncom mu nicable causes many high-income and some middle-income 38 TRUST, VOICE, AND INCENTIVES FIGURE 2.4 Ranking of top 25 causes of death in the Arab world, 1990 and 2010 1990 2010 Mean rank Disorder Disorder Mean rank 164-1 1 Lower respiratory infections 1 Ischaemic heart disease 233.3 139-6 2 Ischaemic heart disease 2 Stroke 179.7 130-6 3Diarrhoeal diseases 3 Lower respiratory infections 130.9 119-7 4 Stroke 4 Diarrhoeal diseases 58.7 61-0 5 Preterm birth complications 5 Diabetes 58.5 54-1 6 Congenital anomalies 6 Road injury 57.7 43-9 7 Protein-energy malnutrition 7 Cirrhosis 56.9 37-1 8 Cirrhosis 8 Preterm birth complications 51.6 36-7 9 Road injury 9 Hypertensive heart disease 43.5 33-1 10 Other cardiovascular and circulatory 10 Congenital anomalies 41.4 disorders 11 Other cardiovascular and circlulatory disorders 39.0 32-6 11 Diabetes 30-9 12 Malaria 12 Chronica kidney disease 37.4 28-2 13 COPD 13 Malaria 36.4 26-2 14 Hypertensive heart disease 14 Cardiomyopathy 31.3 23-5 15 Tuberculosis 15 COPD 30.0 22-6 16 Meningitis 16 Protein-energy malnutrition 28.4 22-5 17 Neonatal encephalopathy 17 HIV/AIDS 27.4 21-7 18 Cardiomyopathy 18 Neonatal sepsis 24.7 20-8 19 Chronic kidney disease 19 Tuberculosis 22.9 20-7 20 Neonatal sepsis 20 Neonata encephalopathy 21.2 18-8 21 Maternal disorders 21 Meningitis 19.6 16-9 22 Rheumatic heart disease 22 Lung cancer 18.2 12-2 23 Measles 23 Rheumatic heart disease 17.9 11-8 24 Mechanical forces 24 Liver cancer 17.4 11-6 25 Asthma 25 Breast cancer 14.0 27 Lung cancer 26 Maternal disorders 37 Liver cancer 27 Asthma 43 Breast cancer 28 Mechanical force 52 HIV/AIDS 85 Measles Communicable, maternal, neonatal and nutritional disorders Ascending order in rank Non-Communicable diseases Descending order in rank Injuries Source: Mokdad et al. 2014. Arab countries resembles that of countries in Syrian Arab Republic, Tunisia, and the West Europe, the United States, and Canada, with Bank and Gaza, ischemic heart disease and health losses from most noncommunicable stroke became the leading causes of death in diseases increasing over the last 20 years. 2010. In Bahrain, Kuwait, Oman, Qatar, The exceptions are Djibouti and the Republic Saudi Arabia, and the United Arab Emirates, of Yemen, where lower respiratory infections the leading causes of death included road and diarrheal diseases continued to be the injuries as well as noncom municable main causes of death during 1990 and 2010. diseases. In Algeria, the Arab Republic of Egypt, Iraq, The rapid increase in noncommunicable Jordan, Lebanon, Libya, Morocco, the diseases in the Arab world suggests that the T H E S TAT E O F E D U C AT I O N A N D H E A LT H S E R V I C E S D E L I V E RY 39 change in the disease burden is a result of a nea rly t h re e - qu a r ters of t he ne eded change in behaviors, including lower levels health care is related to acute illnesses of physical activity, a less healthy diet, and (UNHCR 2013). Meanwhile, many locali- rising health risks such as smoking. Its obesity ties in Iraq, Syria, and the West Bank and rates are among the highest in the world, Gaza have seen their service delivery capac- especially among preschool children and ity partly or largely destroyed, and the defi- women in the Gulf countries (de Onis and cits in service delivery are deepening. Blössner 2000; Kelishadi 2007; Mirmiran Conflict and the limitation of available et al. 2010), and tobacco consumption is very resources, particularly when public infra- high among men (Rahim et al. 2014). structure is targeted, force citizens and gov- Health systems in the region are ill- ernments to severely stretch what resources prepared to tackle the shifting disease burden they have available. In cases of displacement, and the underlying behavioral drivers. many refugees experience trauma as a result Overall, the health system stewardship of the of their journey or the forced substandard national health ministries is underdeveloped. living conditions. Most refugees also arrive in Health information systems are weak, which the host communities with health conditions undermines the design and limits the effec- that require immediate attention. For exam- tiveness of health policies and prevention ple, the common health care needs of Syrian programs. Meanwhile, public health func- refugees entering Lebanon include reproduc- tions such as the promotion of healthy life- tive health care and family planning, child styles and preventive care are relatively health care such as vaccinations, and treat- underdeveloped; health systems in the region ment for acute illnesses such as respiratory have focused instead on curative care infections and gastrointestinal diseases, (Veillard et al. 2011; Mokdad et al. 2014; chronic diseases such as hypertension and Rahim et al. 2014). diabetes, and mental health problems. Yet refugees such as those in Lebanon often can- not access the limited resources available The effects of conflicts because they do not have the ability to pay Confl icts and refugee crises have generated the public fees when applicable (UNHCR education and health challenges across 2014). Some innovative resources such as MENA countries. As we discuss in chapter mobile medical units are available, but these 10, refugee populations have elevated the units are scarce and providing coverage is demand for public services in their host still a challenge. communities. In 2014 in Jordan, for exam- In terms of education, most refugee fami- ple, over 120,000 refugee children enrolled lies value education and would like to register in schools throughout the country (some their children in schools, but they cannot in 100,000 in host communities and 20,000 in part because of no space in nearby schools, camps), and over 35,000 refugee children the distance of schools that do have space received remedial, nonformal, or informal from the place of residence, and the costs of education as well as basic life skills interven- registration, books, and other fees. In certain tions (UNHCR 2013). The refugee influx circumstances, lack of adequate water, sani- has severely strained the health systems of tation, and hygiene facilities in schools is a host countries, most recently in Jordan and barrier to attendance for Syrian children, Lebanon, because of (1) the sharp rise in especially for girls. communicable diseases; (2) the emergence of Conflicts and refugee crises in the MENA new diseases not present earlier in host com- region have reversed decades of educational munities; and (3) the increasing risk of achievement and may create a “lost genera- epidemics such as water-borne diseases, tion.” Because of the collapse in Syria, for measles, and tuberculosis. Among noncamp example, more than 3 million children have Syrian refugees in Lebanon and Jordan, lef t school, and thousands of young 40 TRUST, VOICE, AND INCENTIVES school-age children have never been enrolled and the OECD average of 86 percent (Chahine et al. 2014). Among Syrian refugee (figure 2.5). In fact, on average, MENA children living in urban centers in Jordan, countries pay relatively little attention to school enrollment rates range from 31 to early childhood learning, despite strong evi- 61 percent, depending on the governorate, dence of its immense benefits (El Kogali and and refugee students have a 5 percent drop- Krafft 2015). out rate (UNHCR 2013). In Lebanon, only 6 percent of the surveyed Syrian refugees Equity in access aged 15–24 years were enrolled in formal education even though more than a third of Access disparities remain of concern, espe- the youth sample had been enrolled in Syria cially in health in the MENA region. MENA before displacement (Chahine et al. 2014). countries have broadly achieved equity in The gravity of the situation is even starker access to basic education and selected public when comparing the estimated 72 percent of health services such as immunization. Access Syrian refugees aged 15–18 years enrolled in to other essential services, however, is formal education in Syria in 2010 (reflecting unequal in some countries, driving a high a high 95 percent transition from primary to variation in health outcomes such as child secondary education in Syria in 2010) with malnutrition in Morocco (figure 2.6) and the mere 8 percent of 15–18-year-old Syrian child mortality in Egypt (figure 2.7). refugees enrolled in schools in Lebanon Poor water quality and poor sanitation (Chahine et al. 2014). facilities are largely found in rural areas, and the related deaths and illnesses are concen- trated among the children there. In Djibouti, Access nearly all those who reside in urban areas Access to essential services remains a prob- have clean drinking water compared with lem in the MENA region’s low-income coun- just over half of those in rural areas. Similar tries, Djibouti and the Republic of Yemen. differences in access to clean water are found Their residents face widespread constraints in between urban and rural communities in obtaining basic services, including clean Morocco (37 percentage point gap) and Iraq water—a challenge exacerbated by the effects (35 percentage point gap). In the Republic of of climate change and population growth in Yemen, just three-quarters of those in urban the region1—and sanitation, as well as essen- areas enjoy clean water compared with less tial health care. Partly as a consequence, their than half in rural areas (47 percent). life expectancy is 10 years less than the Consequently, child mortality, especially due regional average. to diarrhea, tends to be higher in less devel- Other countries in the region face nar- oped rural areas. rower access challenges related to education Similarly, disparities in access to critical and health. For example, 52 percent of goods and services, such as prenatal care, households in the MENA countries (the skilled delivery care, iodized salt, child devel- highest rate among all regions), compared opment activities, and early childhood educa- with about 29 percent globally, are without tion, remain large, giving rise to unequal adequately iodized salt, which is critical for opportunities for the healthy development of healthy brain development in children children and their prosperity later in life. The (UNICEF MICS). most advantaged child has, for example, a In education, the key remaining access 16– 69 percent chance of receiving early challenge relates to preprimary education. childhood care and education, depending on Gross enrollment in preprimary education the country, but the least advantaged child stands at 27 percent in the region, compared has a 0 –13 percent chance. Among the with the international average of 48 percent MENA economies for which data exist, the T H E S TAT E O F E D U C AT I O N A N D H E A LT H S E R V I C E S D E L I V E RY 41 gap is the narrowest, fourfold, in the West FIGURE 2.5 Gross enrollment ratios by education level: MENA and Bank and Gaza and sixfold in Djibouti OECD countries, 1999–2012 and Egypt. It is the widest—17-fold—in Iraq and Libya (El-Kogali and Krafft 2015). 120 Such disparities have been found closely related to a mother’s education as well as 100 household income and location. Many women in rural areas and in low- Gross enrollment ratio 80 income populations lack access to essential maternal health services. Rural women in 60 Djibouti, Morocco, and the Republic of Yemen and low-income women in Egypt, 40 Morocco, and the Republic of Yemen are sig- nificantly less likely to have their child deliv- 20 ery attended (figures 2.8 and 2.9). Similarly, women in rural areas and in low-income 0 99 00 01 02 03 04 05 06 07 08 09 10 11 12 populations—such as in Egypt, Morocco, 19 20 20 20 20 20 20 20 20 20 20 20 20 20 and the Republic of Yemen—are less likely to Pre-primary (MENA) Primary (MENA) receive antenatal care during pregnancy Secondary (MENA) Tertiary (ISCED 5 and 6) (MENA) Pre-primary (OECD) Primary (OECD) (figures 2.10 and 2.11). Secondary (OECD) Tertiary (ISCED 5 and 6) (OECD) Access barriers Source: EdStats. Note: MENA = Middle East and North Africa; OECD = Organisation for Economic Co-operation and In some MENA countries, citizens report Development; ISCED = International Standard Classification of Education. physical barriers to access to services. In the Republic of Yemen, the majority of schools do not have a paved road within their catch- Morocco’s rural areas and 8 percent in urban ment area (Public Expenditure Tracking areas reported finding their facility closed Survey, PETS, Republic of Yemen, 2006). In during opening hours at least once during the Egypt, citizens reported in 2010 needing six months prior to the survey (QSDS, almost an hour to reach a health facility Morocco, 2009). This seems consistent with (Egypt Health and Governance Study, the PETS survey’s finding that because of EHGS). Morocco’s citizens said they traveled staff shortages about 20 percent of health an average 14 kilometers to a health facility facilities had to close on the day survey staff (Quantitative Service Delivery Survey conducted mobile visits. [QSDS], Morocco, 2009). By contrast, in Financial barriers are more significant in Jordan and Lebanon, health facilities are health care than education. Out-of-pocket available within a 30-minute walking dis- payments are relatively high in several devel- tance to more than 90 percent of refugees as oping MENA countries, accounting for more well as citizens. than half of total health expenditures in Finding a public health facility closed dur- Egypt, Morocco, the Republic of Yemen, and ing opening hours appears to be a common Syria (World Health Statistics, 2010). In barrier. Patients in the Republic of Yemen Egypt, 37 percent of survey respondents reported in 2010 traveling on average a half- reported that the cost of medicines and fees hour to a health facility, but one in 20 patients were the main deterrent in seeking care who came to a health facility over the last six (EHGS, 2010). In Morocco, out-of-pocket months found it closed during opening payments seem to accrue disproportionately hours (QSDS, Republic of Yemen 2010). to lower-income population groups, who are Fifteen percent of survey respondents in less likely to access the free services offered 42 TRUST, VOICE, AND INCENTIVES FIGURE 2.6 Regional variation in child malnutrition: Morocco, 2010–11 Souss-Massa-Drâa Gharb-Chrarda-Beni Hssen Marrakech-Tensift-Al Haouz Taza-Al Hoceima-Taounate Meknès-Tafilalet Chaouia-Ouardigha Tanger-Tétouan Fès-Boulemane Tadla-Azilal Doukkala-Abda Region du Sud Oriental Rabat-Salé-Zemmour-Zaër Grand Casablanca 0 5 10 15 20 25 Percent Stunting among children <2 years Insufficient weight among children <2 years Source: Demographic and Health Survey, Morocco, 2010–11. FIGURE 2.7 Child mortality (under 5 years): Arab Republic of Egypt, 2008 50 45 40 Number of deaths per 1,000 births 35 30 25 20 15 10 5 0 Upper Egypt-Rural Upper Egypt-Urban Frontier Governorates Urban Governorates Lower Egypt-Rural Lower Egypt-Urban Source: Demographic and Health Survey, Arab Republic of Egypt, 2008. T H E S TAT E O F E D U C AT I O N A N D H E A LT H S E R V I C E S D E L I V E RY 43 FIGURE 2.8 Geographic inequities in access to health services: Selected MENA countries, 2010 100 80 % of births attended 60 40 20 0 6) 6) ) 8) ) ) ) 11 11 09 12 00 00 00 0 0 0 20 . (2 (2 (2 (2 (2 (2 n( ti co p. q ic ep a Ira ou bl Re oc rd ,R pu ib Jo or b en Dj ra Re M m ,A ab Ye pt Ar y Eg r ian Sy Rural Urban Source: World Health Statistics, 2010. FIGURE 2.9 Income inequities in access to health services: Selected MENA countries, 2010 100 80 % of births attended 60 40 20 0 Yemen, Morocco Egypt, Arab Iraq (2011) Syrian Arab Jordan Rep. (2006) (2003–04) Rep. (2008) Republic (2007) (2006) Lowest quintile Highest quintile 44 TRUST, VOICE, AND INCENTIVES FIGURE 2.10 Differentials in the receipt of antenatal care by wealth quintile: Arab Republic of Egypt (2008) and Morocco (2003) 70 % of women receiving no antenatal care 60 50 40 30 20 10 0 Egypt, Arab Rep. (2008) Morocco (2003) Lowest Second Middle Fourth Highest Sources: Demographic and Health Survey, Arab Republic of Egypt, 2008; Morocco, 2003. Note: Percentages shown are for births over last three years. FIGURE 2.11 Limitations in access to antenatal care in rural areas: MENA and other countries, various years 80 Yemen, Rep. 1997 % of rural women receiving no antenatal care 70 60 Morocco 2004 50 40 Egypt, Arab Rep. 2008 30 20 10 Jordan 2007 0 500 1,000 2,500 5,000 10,000 15,000 20,000 GDP per capita (PPP-adjusted), current international $ 95% CI Fitted values Low-income MENA countries Middle-income MENA countries Other countries Sources: Demographic and Health Surveys, ICF International; World Bank. Note: GDP = gross domestic product; PPP = purchasing power parity; CI = confidence interval; MENA = Middle East and North Africa. T H E S TAT E O F E D U C AT I O N A N D H E A LT H S E R V I C E S D E L I V E RY 45 FIGURE 2.12 Percentage of married women mentioning lack of money as a barrier to health care, by region: Morocco, 2010–11 Taza-Al Hoceima-Taounate Marrakech-Tensift-Al Haouz Gharb-Chrarda-Beni Hssen Doukkala-Abda Meknès-Tafilalet Tadla-Azilal Chaouia-Ouardigha Oriental Tanger-Tétouan Rabat-Salé-Zemmour-Zaër Fès-Boulemane Region du Sud Grand Casablanca Souss -Massa-Drâa 0 10 20 30 40 50 60 70 80 Percent Source: Demographic and Health Survey, Morocco, 2010–11. by public hospitals. This reflects in part the FIGURE 2.13 Difficulty or ease in obtaining medical treatment in a fact that among the 58 percent of respon- nearby clinic or public hospital: MENA region, 2010–11 dents unaware of their entitlement to certain health services and medicines for free, a large 100 proportion are poor (World Bank 2013a; 90 QSDS, Morocco, 2009). 80 70 Financial constraints are more pronounced 60 Percent in rural and less developed regions because 50 accessing care often requires relatively high 40 transport costs. Reportedly, lack of money is 30 a factor limiting the majority of married 20 women in Morocco from accessing health 10 care, especially in Morocco’s less developed 0 n . ria sia p. q za an a ep bi o Ira regions (figure 2.12). Re Ga rd ge an ni ,R ra Tu Jo b Al iA b en nd ra Le Other barriers can make it difficult to ud m ka t, A Ye Sa an yp access services. In Algeria, Egypt, Iraq, tB Eg es Lebanon, Tunisia, and the Republic of W Yemen, a majority of citizens reported diffi- Very easy Easy Difficult Very difficult culty in seeking treatment in the nearby clinic or hospital (figure 2.13). As we discuss in Source: Arab Barometer, 2010–11 (Wave II). chapters 6 and 7, these difficulties may relate to staff absenteeism, qualifications, or avail- ability of medicine. education and health, particularly in rural In some countries, cultural norms may areas where they face the obstacles of geogra- aggravate access to services. In the MENA phy and the limited availability of female region, conservative cultural norms place teachers and female health professionals. In women and girls at a disadvantage in access- richer countries, such norms coupled with ing services. Norms restricting the movement outdated health beliefs contribute to the high of women have real consequences for their levels of obesity found among women 46 TRUST, VOICE, AND INCENTIVES (El Zein et al. 2014). Apart from women, citizens cite overcrowding in public facilities older people, internally displaced persons, as a concern. migrants, and refugees, as well as the poor, In education, private providers have played also find themselves disadvantaged in access- a strong role, especially in the expansion of ing services, in part because of cultural, technical and higher education and in prepri- social, and other barriers (Kronfol 2012). mary education, where the majority of Yet other barriers may emerge as a conse- enrolled children attend private facilities in quence of conflicts and refugee crises. Even as most MENA countries. Private primary and host communities in Jordan and Lebanon, secondary education are particularly com- for example, contract additional teachers and mon in Lebanon, Qatar, and the United Arab introduce multiple shifts in schools to accom- Emirates, where the majority of students are modate refugee children, many children estimated to be enrolled in private schools. remain out of school, reporting reasons such In other countries, private schools generally as violence and intimidation at school, chal- cater to students from more affluent back- lenges in adjusting to a new curric ulum, grounds, and in some Gulf countries to inability to catch up after missing months or immigrant children. even years of schooling, and working in order Although the growth of formal private to earn money for their families (United services has improved choices in education Nations 2014). and health care, it has also contributed to inequalities because the private sector is con- centrated in urban areas, compounding the Choice disparities between urban and rural and rich Access to both private education and health and poor (Phillimore et al. 2013). Among services is becoming increasingly common in citizens in the MENA countries, private MENA countries. The situation, however, schools tend to have significantly higher pro- differs widely not only across countries but portions of affluent children than public also within countries, especially between schools (Trends in International Mathematics rural and urban areas. and Science Study, TIMSS). In health, the demand for private care is Overall, few private schools serve rural particularly high in the more specialized areas. In Egypt, for example, the 2010 System services such as mental health and dental Approach for Better Education Results care, where often long wait times and qual- (SABER) survey identified in rural areas only ity limitations in public provision push users 285 private primary schools and 76 private into the private realm (Kronfol 2012). In secondary schools, compared with 16,097 Egypt, nearly half of respondents in the public primary schools and 1,254 public sec- 2010 Egypt Health and Governance Study ondary schools. By contrast, in urban areas used a private service in the last six months, nearly 20 percent of primary and secondary reporting the following main concerns schools were private. The SABER survey in about public facilities: lack of specialized Egypt also found that public primary and doctors, unavailability of doctors, and over- secondary schools had on average about 450 crowding. Choice seems to be limited. For students, whereas private primary and sec- example, in Morocco 93 percent of rural ondary schools tended to be smaller, with citizens and 72 percent of urban citizens about 350 and 150 students, respectively. reported having only one provider, reflect- Meanwhile, the Republic of Yemen had ing in part the fi nancial barriers to accessing 294 private and 944 public primary schools private services (QSDS, Morocco, 2009). In in urban areas, but only 9 private primary Jordan and Lebanon, the demand for pri- schools, compared with 10,569 public pri- vate provision in both education and health mary schools, in rural areas (SABER, 2010). has sharply increased in communities Similarly, in Jordan and the West Bank and affected by conflict and refugee crises as Gaza, private school teachers accounted for T H E S TAT E O F E D U C AT I O N A N D H E A LT H S E R V I C E S D E L I V E RY 47 nearly a third of all primary and secondary activities—and these are among the lowest school teachers in urban areas, but for less estimated rates in the world. Meanwhile, vio- than 5 percent of teachers in rural areas lent child discipline involving psychological (SABER, 2010). aggression or physical punishment is wide- Beyond formal private health facilities and spread in the MENA region, negatively affect- schools, private service provision in both ing children’s physical, psychological, and education and health in the MENA region social development (El-Kogali and Krafft includes a shadow market driven in part by 2015). In the classroom, the results from early citizen choice and in part by distortions in grade reading and math assessments reveal provider incentives in the absence of any via- that MENA children are lacking the required ble regulatory mechanisms. This shadow foundational reading and math skills. market includes public sector teachers who Likewise, students in MENA countries fall engage in tutoring and health professionals behind those in most other countries in basic who operate dual practices, thereby facing reading, math, and science as measured by potential confl icts of interest (as we discuss TIMSS and the 2011 and 2012 Programme in chapter 6). for International Student Assessment (PISA) test scores (figure 2.14). The majority of The quality challenge Service quality is a major challenge through- FIGURE 2.14 Benchmarking student math scores against levels of GDP per capita: MENA economies, 2011 and 2012 out the MENA region. Increasingly, this challenge is taking center stage in the public 650 debate across MENA countries as well as in country development strategies and govern- ment education and health policy documents. 600 The quality challenge pertains to the entire education and health systems and to public providers specifically. Service quality appears 550 somewhat higher in private education and TIMSS and PISA scores, math health care. In Bahrain, Lebanon, Oman, and preconflict Syria (but not in Tunisia), stu- 500 dents in private schools outperformed their peers in public schools on TIMSS tests. In the West Bank and Gaza health sector, private facilities outperformed 450 Lebanon Tunisia public facilities in some elements of quality United Arab Emirates such as waiting times and the availability of Qatar Jordan Bahrain medicines and advanced equipment. As dis- 400 Saudi Arabia cussed shortly, MENA citizens on average Tunisia Qatar perceive the quality of education and health Jordan Oman care to be higher from private providers, 350 Morocco Egypt, Arab Rep. especially in education. 300 Student performance 3.0 3.5 4.0 4.5 5.0 5.5 Log of GDP per capita (PPP adjusted $) The challenge of attaining quality is immense PISA, 2012 MENA-PISA, 2012 TIMSS, 2011 from preprimary education to higher MENA-TIMSS 2011 Linear (PISA, 2012) Linear (TIMSS, 2011) education. Among young children, only 33 percent in the Republic of Yemen, Sources: TIMSS, 2011; PISA, 2012. Note: PISA = Programme for International Student Assessment; TIMSS = Trends in International 36 percent in Djibouti, and 48 percent Mathematics and Science Study; MENA = Middle East and North Africa; PPP = purchasing power in Morocco engage in developmental parity; GDP = gross domestic product. 48 TRUST, VOICE, AND INCENTIVES FIGURE 2.15 Student performance on TIMSS mathematics benchmarks (grade 8): MENA and OECD economies, 2011 100 90 80 70 60 Percent 50 40 30 20 10 0 co an ic e a sia za ain an ar n s e te bi ag ag no bl t Ga oc Om rd ni Qa ira hr ra pu er er ba Tu Jo or iA Ba nd Em Av Av Re Le M ud ka ab D ab b Sa C ra an Ar Ar OE nA tB d ite ria es W Un Sy Advanced (at or above 625) High (550–625) Intermediate (475–550) Low (400–475) Below low (<400) Source: TIMSS, 2011. Note: TIMSS = Trends in International Mathematics and Science Study; MENA = Middle East and North Africa; OECD = Organisation for Economic Co-operation and Development. fourth- and eighth-grade students in MENA The variation in educational attainment countries participating in these tests scored suggests gaps in achieving equality of oppor- “below low” in math and science (figure 2.15). tunity. Students’ school performance relates In higher education, the rapid quantitative more closely to their socioeconomic back- expansion in both public and private provi- grounds in MENA countries than interna- sion has raised significant quality concerns (El tionally. In the 2011 math TIMSS scores, Hassan 2013; Baporikar 2014). MENA students in schools with a high share Interestingly, girls on average outperform of disadvantaged students fell behind their boys across MENA countries. Not only do peers in the more affluent schools by girls typically achieve better scores in math, 41 points (and by over 77 points in Morocco), reading, and science in international tests compared with the 49-point difference in such as the Progress in International Reading OECD countries (figure 2.16). Literacy Study (PIRLS), TIMSS, and PISA, In part as a reflection of the extensive but they also tend to have lower repetition inequality in the quality of primary and sec- rates across urban and rural areas, in private ondary education, the recent expansion in as well as public schools, as measured by the access to higher education in MENA devel- national education statistics in, for example, oping countries has applied less to children Morocco. The gap is especially high in the in rural areas and those whose parents have Gulf Cooperation Council (GCC) countries, low levels of education and are engaged in perhaps reflecting in part the social contract basic occupations than to children from offering most boys an implicit public sector urban and stronger socioeconomic back- job guarantee based on their citizenship. grounds. In Egypt, children from the top T H E S TAT E O F E D U C AT I O N A N D H E A LT H S E R V I C E S D E L I V E RY 49 FIGURE 2.16 Student performance on TIMSS mathematics (grade 8), by socioeconomic background of schools: MENA economies and internationally, 2011 600 500 400 Score 300 200 100 0 an co ic p. n a ain a on e e za r D ta bi isi ag ag a bl Re C Ga oc rd Om an Qa hr ra OE n pu er er Tu Jo or ic iA b Ba nd Av Av n- Re am Le M ud no ka A CD ab Isl EN Sa an A, OE Ar n, M EN tB Ira ian M es r n- W Sy No 0–10% disadvantaged 11–25% disadvantaged 26–50% disadvantaged More than 50% disadvantaged Source: TIMSS, 2011. Note: TIMSS = Trends in International Mathematics and Science Study; MENA = Middle East and North Africa; OECD = Organisation for Economic Co-operation and Development. wealth quintile are over twice as likely to circumstances. The UNRWA advantage in complete a secondary education and four learning outcomes—despite apparent com- times as likely to complete a college educa- monalities with public school—is 25 points, tion than those from the bottom quintile, a quarter of a standard deviation, or about a and children from urban governorates are year’s worth of learning. Moreover, more more likely to complete a secondary or higher UNRWA students achieve the international education than those from rural Upper Egypt benchmarks in math and science. UNRWA or Frontier governorates. Consequently, students in Jordan and the West Bank and between 1998 and 2006 the share of college Gaza achieve on average scores that are graduates among the least advantaged youth 23–80 points higher than their peers in pub- increased by only 1 percentage point, com- lic schools, even after controlling for student pared with a 17 percentage point increase characteristics and for urban or rural con- among the most advantaged youth (Ersado texts (Patrinos et al. 2013). and Gignoux 2014). Although conflicts and refugee crises gen- Health services quality and the erally have a negative impact on education problem of opacity quality, exceptions exist. Most notably, chil- dren in schools operated by the United Health care in the MENA region does not Nations Relief and Works Agency for measure up to international standards and Palestine Ref ugees in the Near East people’s expectations, as chapter 1 illus- (UNRWA)2 are achieving higher than aver- trated. The actual practice of health care age learning outcomes despite their adverse often differs from evidence-based practices 50 TRUST, VOICE, AND INCENTIVES because of poor oversight and lack of generally not measured in MENA countries.4 accountability. The literature suggests that Service quality standards tend to be poorly long waiting lines, absent providers, lack of defined and, where defined, not publicly privacy and confidentiality, medication available. errors, and informal payments erode the As discussed in chapter 5, although quality of health care across the MENA schools and health facilities are inspected region (Zaky, Khattab, and Galal 2007; regularly in most MENA countries, informa- Jabbour and Yamout 2012). The difficulty in tion on performance is typically not shared obtaining appointments, long waiting times, throughout the centralized education and overcrowding, and poor infrastructure and health systems, receives no follow-up, and is hygiene have been also reported in surveys in not made available publicly. According to the Morocco, the Republic of Yemen, and, fol- TIMSS, significant variation in the transpar- lowing the Syrian refugee crisis, in host com- ency of information on performance exists in munities in Jordan and Lebanon trying to the MENA region. Nearly half of students in extend services to refugees. Qatar and the United Arab Emirates attend Facility surveys in Egypt revealed deficien- schools in which achievement data are posted cies in practices. The 2010 Egypt Health and publicly, compared with 20 percent and less Governance Study, which surveyed facilities students in such schools in Jordan and across Alexandria and Menoufia, used direct Tunisia. observations of clinical practices to measure Opacity on performance exacerbates whether protocols were followed and to information asymmetries that bar citizens calculate a “quality index”3 for the care from making informed choices and demand- provided. According to this quality index, the ing accountability, and it intensifies market average provider performed 63 percent inefficiencies (Keefer and Khemani 2005). of standard procedures that constitute a Meanwhile, the lack of information on the routine antenatal examination, 59 percent of quality of services provided by individual the standard procedures for examination of a schools and health facilities, and by the edu- sick child, and only about 38 percent of the cation and health systems at large, is one of standard procedures for diabetes or coronary the institutional roots of the quality challenge heart disease (CHD)/hypertension patients. in MENA countries. Much like the results of other studies, the analysis suggests relatively good standards Implications for efficiency for perceived priority interventions such as immunization, maternal health, and epidemic The quality challenge in the MENA region control, but poor care for chronic disease colors analysis of the efficiency of public management, prescribing patterns, health expenditure on education and health services education, and referral patterns (World Bank in MENA countries. 5 The efficiency of the 2013b). We will discuss some of these results education and health systems of these in more detail at the national level in chapter countries varies widely, depending on the 6 and the subnational level in chapter 7. specific result under consideration. For exam- ple, in primary education enrollments and life expectancy at birth, many MENA coun- Information about service quality and tries appear to be achieving high efficiency in standards—rarely available public spending on education and health. In Overall, little information on service quality fact, the MENA region ranks second globally is reported and collected within the education in efficiency of spending on education as and health systems, and even less is available measured by net primary enrollment and sev- to citizens. Key aspects of performance such enth globally in health as measured by life as continuity, comprehensiveness, and appro- expectancy at birth (Mottaghi forthcoming). priateness in health care provision are Compared with those of other regions, T H E S TAT E O F E D U C AT I O N A N D H E A LT H S E R V I C E S D E L I V E RY 51 MENA’s efficiency scores are strong for both education system. The GCC countries developing MENA countries and for the Gulf achieve an efficiency score of 0.678 in their countries. Lebanon and preconflict Syria public spending on education, which is rela- appeared to be on the efficiency frontier in tively lower than OECD countries’ score of their public spending on health (figure 2.17), 0.879 (table 2.1). The Republic of Yemen and and several MENA countries were near the Morocco appear to have relatively inefficient efficiency frontier on education during education and health systems because they 2000–10 (figure 2.18). Considering education are both far off the efficiency frontier in their enrollment but using a different methodol- public spending when considering student ogy, Grigoli (2014) estimates that efficiency performance and rural women’s access to of public spending on education varies antenatal care as the metric (figures 2.19 significantly in the MENA region, with and 2.20). Bahrain, Kuwait, and Qatar achieving the Another perspective on the efficiency of highest efficiency scores and Djibouti and education systems is grade repetition Morocco the lowest. and dropout rates. According to the World If efficiency is measured in relation to ser- Bank’s World Development Indicators, the vice quality or equity, MENA countries repetition rates in MENA countries exceeded appear to exhibit lower levels of efficiency in 6.4 percent (7.6 percent among boys) at the public spending on education and health. In primary level in 2012, which was higher than student performance on international tests, the international average of 4.6 percent only Lebanon—among the MENA countries (4.4 percent among boys). The rate was par- for which TIMSS data exist—appears on the ticularly high in Lebanon and Morocco efficiency frontier in its public spending on (above 8 percent in total and close to education and seems to have an efficient 10 percent among boys). National education FIGURE 2.17 Preliminary estimates of efficiency frontier, life expectancy at birth: Selected MENA countries, 2000–10 90 80 Life expectancy at birth (years) 14 102 9 1 70 168 11 15 7 17 13 54 12 6 60 318 50 40 0 2,000 4,000 6,000 Per capita health spending (PPP-adjusted), current international $ Source: Mottaghi forthcoming. Note: 1 = United Arab Emirates, 2 = Bahrain, 3 = Djibouti, 4 = Algeria, 5 = Arab Republic of Egypt, 6 = Islamic Republic of Iran, 7 = Iraq, 8 = Jordan, 9 = Kuwait, 10 = Lebanon, 11 =Libya, 12 = Morocco, 13 = Oman, 14 = Qatar, 15 = Saudi Arabia, 16 = Syrian Arab Republic, 17 = Tunisia, 18 = Republic of Yemen; MENA = Middle East and North Africa; PPP = purchasing power parity. 52 TRUST, VOICE, AND INCENTIVES FIGURE 2.18 Preliminary estimates of efficiency frontier, net primary enrollment: Selected MENA countries, 2000–10 120 100 . 11 512 34 1 7 Net primary enrollment (%) 10 6 80 9 8 60 40 2 20 0 500 1,000 1,500 2,000 2,500 Per capita education spending (PPP-adjusted), current international $ Source: Mottaghi forthcoming. Note: 1 = United Arab Emirates, 2 = Djibouti, 3 = Arab Republic of Egypt, 4 = Islamic Republic of Iran, 5 = Jordan, 6 = Kuwait, 7 = Lebanon, 8 = Morocco, 9 = Oman, 10 = Qatar, 11 = Syrian Arab Republic, 12 = Tunisia; MENA = Middle East and North Africa; PPP = purchasing power parity. TABLE 2.1 Preliminary estimates of efficiency scores, by region, 2010–11 Education Health Eighth-grade math scores (PISA, 2012; Percentage of women receiving TIMSS, 2011) antenatal care in rural areas MENA 0.785 0.611 GCC 0.678 — MENA dev oil export 0.853 0.290 MENA dev oil import 0.858 0.717 EAP 0.906 0.909 ECA 0.947 0.901 LAC 0.814 0.929 SA — 0.779 SSA 0.774 0.834 HIC 0.860 — HIC: non-GCC 0.882 — HIC: OECD 0.879 — HIC: non-OECD and non-GCC 0.909 — Lower MIC 0.903 0.826 Upper MIC 0.853 0.916 Source: Mottaghi forthcoming. Note: — = not available; PISA = Programme for International Student Assessment; TIMSS = Trends in International Mathematics and Science Study; MENA = Middle East and North Africa; GCC = Gulf Cooperation Countries; EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin America and the Caribbean; SA = South America; SSA = Sub-Saharan Africa; HIC = high-income country; OECD = Organisation for Economic Co-operation and Development; MIC = middle-income country. T H E S TAT E O F E D U C AT I O N A N D H E A LT H S E R V I C E S D E L I V E RY 53 FIGURE 2.19 Preliminary estimates of efficiency frontier, 2012 PISA math scores and 2011 TIMSS math scores for eighth-grade students: Selected MENA countries 600 500 TIMSS Math score LBN IRN 400 SAU JOR TUN SYR MAR QAT OMN KWT 300 0 5,000 10,000 15,000 Per capita public expenditure on secondary education (PPP-adjusted), current international $ Source: Mottaghi forthcoming. Note: TIMSS = Trends in International Mathematics and Science Study; PISA = Programme for International Student Assessment; MENA = Middle East and North Africa; IRN = Islamic Republic of Iran; JOR = Jordan; KWT = Kuwait; LBN = Lebanon; MAR = Morocco; OMN = Oman; QAT = Qatar; SAU = Saudi Arabia; SYR = Syrian Arab Republic; TUN = Tunisia; PPP = purchasing power parity. FIGURE 2.20 Preliminary estimates of efficiency frontier, percentage of women receiving antenatal care: Selected MENA countries, available years from 1997 to 2013 100 JOR % of women receiving antenatal care 80 EGY 60 MAR 40 YEM 20 0 100 200 300 400 500 Per capita public expenditure on health (PPP-adjusted), current international $ Source: Mottaghi forthcoming. Note: EGY = Arab Republic of Egypt; JOR = Jordan; MAR = Morocco; MENA = Middle East and North Africa; YEM = Republic of Yemen. PPP = purchasing power parity. 54 TRUST, VOICE, AND INCENTIVES FIGURE 2.21 Repetition rates, by education level and geographic coverage/gender in public schools: Morocco, 2010–11 20 16 Percentage of students 12 8 4 0 Primary Middle Secondary Total girls Urban girls Rural girls Total boys Urban boys Rural boys Source: Education statistics, government of Morocco. Note: Definition of repetition rate from United Nations Educational, Scientific and Cultural Organization, http://www.uis.unesco.org/Library/Documents /eiguide09-en.pdf. statistics—for example, in Morocco—indi- countries with lower illiteracy rates and a cate that in repetition rates, the gender gap is higher per capita gross domestic product significantly wider than the rural-urban (GDP), education quality is perceived more divide (figure 2.21). The primary completion positively—regardless of actual quality— rate surpasses 90 percent in most MENA than in countries with higher poverty and countries (compared with the international illiteracy rates. average of 86 percent). Dropout rates are reported to be a greater problem in middle Public sector versus private sector and secondary schools across MENA coun- tries. In the Gulf countries, which have Satisfaction is higher with both education opened higher education opportunities to all and health care provided by the private sector citizens, the dropout rates have reached rela- than by the public sector. Generally, citizens tively high levels in universities. perceive public schools and health facilities to be of lower quality than private establish- ments and cite overcrowding, inadequate Citizens’ satisfaction infrastructure, and nonmotivated staff Evidence on citizens’ satisfaction with (World Bank 2007 and surveys cited in this services is mixed across and within MENA report). This perception is consistent with the countries. As shown in chapter 1, the satis- available data showing on average better ser- faction rates with education and health vice delivery results—in terms of quality and services are relatively low among citizens in service delivery indicators (see chapter 6)— the MENA region, particularly in its develop- for private schools and facilities compared ing countries, compared with the averages in with public ones. other regions. Citizens’ satisfaction with Generally, satisfaction with public ser- education services and with the availability of vices tends to be low; users cite having few quality health care is relatively high in alternatives, giving them little ability to use Bahrain, Kuwait, Oman, Qatar, and the “choice” in holding providers accountable. United Arab Emirates (Gallup World Poll, Proximity is a key determinant in choosing 2013). Surveys suggest that in societies and a provider. When asked about the most T H E S TAT E O F E D U C AT I O N A N D H E A LT H S E R V I C E S D E L I V E RY 55 important factor for them in choosing a satisfaction with the timeliness and cost of health provider, 42 percent of patients in services. The main areas of dissatisfaction Morocco referred to proximity, whereas were the availability and cost of medica- only about a quarter referred to the cost of tions, the exchange of information about consultation and 16 percent to the unavail- health and medications with health provid- ability of the specific type of care sought at ers, and the quality of education. Apart another facility (QSDS, Morocco, 2009).6 In from these similarities, the reported levels the Republic of Yemen, 34 percent of of satisfaction with education and health respondents in the 2010 QSDS reported differ within as well as across countries, in choosing a particular facility because of its part reflecting the subnational variation in proximity to their home. Unfortunately, service delivery performance, a topic many have limited choice in service provid- explored in chapter 7. ers, having to make do with those that exist. In Egypt, respondents to the 2009 GAC In Morocco, the vast majority of both rural survey placed education services among the (93 percent) and urban (72 percent) patients top 3 of the 13 government services, with reported in the 2009 QSDS that they did not three-quarters of respondents indicating they have multiple providers from which to had received education services in a timely choose and that proximity was the main fashion, and nearly two-thirds saying they reason for their choice of provider. Although had been treated “decently” by education the lack of choice declared by patients may providers and charged a reasonable amount mean they cannot afford the cost of private in fees. providers as an alternative to, for example, a The Community Scorecard initiative basic health care facility or établissement de launched by the World Bank in 2010 sur- soins de santé de base (ESSB), the availabil- veyed parents in Egypt’s Ismailia governorate ity of options per se, regardless of cost, is and found significantly lower levels of satis- also likely to be an issue, particularly in faction with education, even thoug h rural areas (World Bank 2013a). 93 percent reported that they were satisfied In the Republic of Yemen, according to the or very satisfied with the headmaster’s per- 2009 Governance and Anti-Corruption formance (Bold and Svensson 2010). Among (GAC) Country Diagnostic Survey, among the parents sending their children for tutor- those using the services over the last six ing, 76 percent reported the need for extra months, less than half of respondents in both academic help, and 21 percent reported that urban and rural areas reported satisfaction inadequate instruction at their children’s with public health care (particularly the avail- school was the main reason. ability and cost of medications), and less than For health services, over 60 percent of 40 percent expressed satisfaction with public respondents to the 2009 GAC survey in education. The survey identified the quality of Egypt reported receiving services in a timely public education as a serious social problem— fashion and at an acceptable cost, and half indeed, more serious than inflation, public reported they felt they were treated decently. sector corruption, and health care quality. Similarly, the 2010 Egypt Health and Governance Study household survey and in- depth interviews indicated general satisfac- Satisfaction with education and tion with health facilities and services, health in Egypt, Morocco, and the including their availability, timeliness, and Republic of Yemen cost. Patients mainly complained about short- Citizens tend to be more satisfied with the ages of specialists, medications, and labora- timely availability of education and health tory tests. services than with their quality. On average, In Morocco, the 2009 QSDS revealed across Egypt, Morocco, and the Republic of general satisfaction with most elements of Yemen in recent years, residents reported health care (on average more than half of 56 TRUST, VOICE, AND INCENTIVES FIGURE 2.22 Patients’ dissatisfaction with aspects of health care: Republic of Yemen, 2010 Availability of medications Cost of medications Infrastructure of the health center Explanations Level of hygiene in facility Waiting time Reception and administration time Time with provider Competencies of medical staff Privacy Attitude of staff Integrity of staff Quality of services provided Treatment by staff Explanations about medications Explanations about health 0 5 10 15 20 25 30 35 40 45 50 % dissatisfied Household survey Exit poll patients Source: QSDS (health), Republic of Yemen, 2010. respondents were at least somewhat satisfied). the Republic of Yemen across both survey The main source of dissatisfaction was the samples. Twenty-nine percent of household cost of prescription drugs, with nearly two- survey respondents and 43 percent of exit thirds of respondents saying they were some- poll patients were dissatisfied with the avail- what or very dissatisfied with these costs, and ability of medications, and 22 percent and only 3 percent thought the costs of prescrip- 41 percent, respectively, were dissatisfied tion drugs were very good. with the cost (figure 2.22). In the Republic of Yemen, the 2009 GAC survey found that on average less than Variation in service satisfaction 30 percent of respondents in both urban and in Tunisia rural areas rated the quality of public educa- tion that their child received during the last Citizens’ satisfaction with services can vary school year as “good,” even as the majority significantly within countries. Surveys from expressed satisfaction with the teachers, syl- Tunisia, for example, indicate a considerable labi, and school access and infrastructure. variation in the perceptions of service quality But significant variation emerged within across sectors (figure 2.23) and localities regions, with the majority of respondents (figures 2.24 and 2.25). Satisfaction levels from Al-Mahwit (80 percent), the capital across sectors are most closely correlated (59 percent), and Mareb (54 percent) satis- between education and health services fied, but few from Amran (7 percent) and (table 2.2). The correlation in satisfaction Lahj (4 percent). In health care, much like between education and health services Egypt and Morocco, the 2010 QSDS, using appears relatively strong compared with both exit interviews and a household survey, other sectors, and also when comparing identified availability and the cost of medica- results across countries using the 2013 Gallup tions as the main sources of dissatisfaction in World Poll data (tables 2.3 and 2.4). T H E S TAT E O F E D U C AT I O N A N D H E A LT H S E R V I C E S D E L I V E RY 57 Similar variation is found in the Republic FIGURE 2.23 Variations in perceptions of service quality: of Yemen. There, the 2013 Yemen Polling Tunisia, 2014 Center survey found variations in the atti- 100 tudes toward service provision among elites in Sana’a, Aden, and Taiz, with respondents 90 in Aden viewing education and health most 80 negatively. By contrast, respondents in Sana’a 70 found social security and electricity to be the 60 Percent most problematic areas, whereas those in 50 Taiz viewed sanitation and water most nega- 40 tively. We further discuss subnational varia- 30 tion in service delivery in chapter 7. 20 10 Corruption 0 Electricity Primary and Security Health Roads Garbage Although citizens’ perceptions of overall secondary facilities collection corruption in MENA countries are compa- education rable with those in other regions, MENA Very poor Poor Good Very good citizens are more likely to refer to corruption when reporting on their experience with Source: Transitional Governance Project, 2014. FIGURE 2.24 Satisfaction with quality of garbage collection, by region: Tunisia, 2014 100 90 80 70 60 % of respondents 50 40 30 20 10 0 Sa azzoiadh K n Ar a a la Bo e r g a i r o sr d o Sid Ha ram en g ka eb e Sb ika r M en h Sid ize Mat ra s t a Be K e a h ala El e S ira i A zou uth um u a Ka Ka ar a eg n El outh J a ri n e zid rth Ou i Bo ine 'S h u s ur o h e Ka sr hs Je u s Th a ba a t M A So stir M nza r ze F t M rz i s B am r Ba on hir l B uz ou No r E l d As l i M e R uj Na Fa l let i d k n l n l lal S i d Ez S o id Sid ou id e r Z E as M ke Bi si El Ezzi E l af n bir lh u ch ch Ha ie un Ch r a n lm e i B rte eu Le dou ba S M o h e Omssin ib Bi H Mel M ak e So Mo out i H haj le u A Am ed l M ui o m be ho l aa all in h rte er i E ffo ar o u M ec b a ua He m B S e i Gh Sa s k B Dj se Sa us So Dissatisfied Satisfied Source: Transitional Governance Project, 2014. 58 TRUST, VOICE, AND INCENTIVES FIGURE 2.25 Satisfaction with quality of education, by region: Tunisia, 2014 100 90 80 70 60 % of respondents 50 40 30 20 10 0 K e ine Fa h er u d o Bi M Kram zid ajla Ka Me ken er e Na Zarz a A a S ala i ka M lah en El nzah eb a E l Mg elh M ffo h Sa ou ra e S ra Gh Jam bira M As ne ’Sa a Sid J oun K Bo ma e R ne E l F th i B Bo uika So h Fe g u a l e d kn h ss S o s Dj Bo Bar s E l s r al i s ir r M m it Th st l B ur be ou r o n uz Ka B e o r Ba S o u r z e He i t j M h el Al l M e l N l Sa Sou Has id i E ho d Sid lha af n i A Sa asti Ha uan rou a z ke h iro n un C h bib ch ran M nch Bi asr u Al t kie rou rte lla Ha ort b ut O u M o iad Le e c h Bi O E z z M hi i n Am Ea ou Sid zzou eri iE u ri a rte te ar m ss s i i i m l a a gr i el ou uh ba m id ze a e He z E ed Sid Sid let M ba se us So Dissatisfied Satisfied Source: Transitional Governance Project, 2014. TABLE 2.2 Correlation between individuals’ evaluations of different services (Spearman’s Rho): Tunisia, 2014 Education Health Roads Electricity Security Garbage Education 1 Health 0.5946 1 Roads 0.3947 0.5233 1 Electricity 0.4306 0.4297 0.4566 1 Security 0.4046 0.4002 0.2716 0.4148 1 Garbage 0.2957 0.3857 0.4568 0.3642 0.3073 1 Source: Transitional Governance Project, 2014. education and health services. According to to education or health is higher than the Transparency International’s 2013 Global global average, reaching 51 percent of Corruption Barometer, more than half of respondents in Morocco who reported pay- respondents in MENA countries believe their ing informal fees in the health sector and 38 education and health systems are corrupt or percent of respondents in Jordan who extremely corrupt. The proportion of MENA reported paying informal fees in the educa- citizens who have paid informal fees related tion sector (figure 2.26). T H E S TAT E O F E D U C AT I O N A N D H E A LT H S E R V I C E S D E L I V E RY 59 TABLE 2.3 Correlation between individuals’ satisfaction with different services: Globally, 2013 Telephone Internet Cellular Transport Road Water Education Health Telephone 1 Internet 0.8093 1 Cellular 0.4529 0.5552 1 Transport 0.3886 0.4238 0.3681 1 Road 0.4713 0.4973 0.3979 0.7378 1 Water 0.4659 0.6002 0.3668 0.6421 0.7045 1 Education 0.1413 0.2154 0.1661 0.6258 0.6582 0.7141 1 Health 0.4396 0.5299 0.2835 0.6455 0.797 0.8014 0.7948 1 Source: Gallup World Poll, 2013. TABLE 2.4 Correlation between individuals’ satisfaction with different services: MENA region, 2013 Telephone Internet Cellular Transport Road Water Education Health Telephone 1 Internet 0.6762 1 Cellular 0.2974 0.5624 1 Transport 0.2309 0.5729 0.3562 1 Road 0.3507 0.6821 0.3454 0.9369 1 Water 0.4618 0.7296 0.2588 0.741 0.8225 1 Education 0.2542 0.6397 0.3168 0.8237 0.8543 0.6623 1 Health 0.428 0.7952 0.4345 0.7279 0.858 0.7633 0.8181 1 Source: Gallup World Poll, 2013. FIGURE 2.26 Percentage of respondents reporting payment of informal fees, education and health care sectors: MENA region and globally, 2013 60 50 40 Percent 30 20 10 0 Algeria Egypt, Arab Iraq Jordan Morocco West Bank Tunisia Yemen, Rep. MENA Global Rep. and Gaza Education Health care Source: Global Corruption Barometer, 2013. Note: MENA: Middle East and North Africa. 60 TRUST, VOICE, AND INCENTIVES FIGURE 2.27 Perceptions of role of corruption in services: Republic of Yemen, 2013 100 90 80 70 % of respondents 60 50 40 30 20 10 0 Corruption at Poor Corruption at Weak Lack of the central supervision the local level infrastructure capacity at the level and lack of local level investment Source: Local Governance Survey, Republic of Yemen, 2013. In the Republic of Yemen, an inventory of not fully aware of the official costs of care, corruption complaints over a three-year which creates opportunities for service period revealed that 240 cases of corruption providers to demand additional payments. were reported in 2005, 358 in 2006, and 558 Surveys on health services delivery and in 2007. Indeed, acts of corruption related to mechanisms that promote transparency in education are some of the more commonly the payment of health services fees, including reported incidences of corruption in the receipts for services and posting of fees, have Republic of Yemen, reaching 22 percent of found some important variations across the cases filed in 2005.7 Health care was the region. Facility-based surveys in Egypt show second most frequent source of corruption that just fewer than 8 in 10 facilities had their cases fi led. In 2005, 10 percent were associ- fees visibly posted (EHGS, 2010). Among ated with the Ministry of Health; in 2006, health service users in Morocco, 84 percent 8 percent; and in 2007, 11 percent.8 Against reported being given a receipt with item- this backdrop, it may not be surprising that i z ed fe e s at t hei r last v isit (Q S D S , the 2013 Local Governance Survey in the Morocco, 2009—see figure 2.28). In the Republic of Yemen suggested that citizens Republic of Yemen, the lack of transparency often blame service delivery problems on cor- is even starker; just 4 percent of patients exit- ruption at both the local and central levels ing a facility reported that fees were visibly (figure 2.27). posted, and only 42 percent of patients In many cases, the lack of transparency in reported they had been given a receipt the cost of services, outcomes, and citizens’ (QSDS, Republic Yemen, 2012). Partly as a rights creates conditions ripe for corruption consequence, different citizens may end up and the selective provision of services . In paying different fees for the same service medical facilities, for example, citizens are (see figure 2.29 based on the EHGS). T H E S TAT E O F E D U C AT I O N A N D H E A LT H S E R V I C E S D E L I V E RY 61 FIGURE 2.28 Transparency in fees at health facilities: Morocco (2009), Arab Republic of Egypt (2010), and Republic of Yemen (2010) 90 80 70 60 50 Percent 40 30 20 10 0 Morocco: patients Egypt, Arab Rep.: Yemen, Rep.: Yemen, Rep.: exiting given receipt with facilities with fees patients given a patients reporting fees visibly posted receipt fees were visibly posted Sources: Morocco: QSDS (health), 2009; Arab Republic of Egypt: EHGS, 2010; Republic of Yemen: QSDS (health), 2010. Reflections in the wider historical FIGURE 2.29 Fees for examination by doctor, by background and institutional contexts characteristics: Arab Republic of Egypt, 2010 The remarkable advancement of MENA 45 Amount paid for doctor’s exam (Egyptian pounds) countries in expanding access to basic 40 education and health services and the remaining challenge of service quality and 35 citizens’ satisfaction have common roots in 30 the unique institutional and sociocultural 25 contexts for service delivery in the MENA 20 region. As we discussed in chapter 1, citizens of the MENA region not only demand better 15 services, but also expect their governments to 10 provide them, reflecting the promises made 5 by Arab leaders at the outset of independence and the subsequent social contract outlined 0 4t qui tile R n 15 le ex st + s Ev oon –6 yrs en ria Fe ale te ing g fia Ri ntile M g 3r qu est 2n Po s ui ile n- d gi l al 65 4 yr ba yr Re ura in in chapter 1 and discussed in more detail in re a e No re ou t M and h q nt m ch d in M d or Af orn en To 40 9 Ur ste rn –3 chapter 4. The postindependence social con- Al tract, however, has eschewed the creation Source: EHGS, 2010. of institutions and accountability mecha- nisms at the political, administrative, and social levels that would motivate providers, underpinnings of the current realities in public servants, and policy makers to deliver service delivery performance in MENA quality services to the poor and other non- countries. privileged populations. Chapters 4 and 5 will As we discuss in chapters 4–9, the citizens examine these historical and institutional of the MENA region are facing a cycle of 62 TRUST, VOICE, AND INCENTIVES performance stuck in a low equilibrium: 3. To measure differences in adherence to pro- institutions and accountability mechanisms tocols between different types of consulta- are failing to promote adequate performance. tions, the EHGS normalized all structural This situation undermines citizen trust, observations to a quality index. The index compiles all components of the observation prompt i ng t hem to adopt for m s of checklist into a score between 0 and 1, in engagement that further erode institutions, which 0 means that none of the elements of a accountability mechanisms, and norms. In checklist was observed and 1 means that the this low equilibrium, the poor fi nd it espe- provider conducted every single aspect of the cially difficult to have their voice heard and consultation according to the checklist. needs addressed. The observed subnational 4. Few dimensions of health care performance variation in performance, satisfaction, and are measured in MENA countries, whereas in trust further confirms the existing system- Canada, for example, performance dimen- wide weaknesses and suggests that they can sions such as acceptability, accessibility, be partly counteracted at the local level. appropriateness, competence, continuity, Indeed, some localities in MENA coun- effectiveness, efficiency, equity, responsive- ness, and safety are measured. tries have successfully counteracted the 5. Empirical and theoretical measures of effi- national cycle of poor performance and ciency are based on ratios of observed output developed local solutions—schools and levels to the maximum that could have been health facilities that motivate their staffs, obtained given the level of input utilization. deliver excellent results in low-capacity set- This maximum constitutes the efficient fron- tings without any special advantages, and tier, which is generally used as a benchmark inspire citizens’ satisfaction, trust, and con- for measuring the relative efficiency of the structive engagement in the community. The observations (http://web.worldbank.org next chapter will describe four examples of /WBSITE/EXTERNAL/TOPICS/EXTDEBT such local successes. DEPT/0,,contentMDK:20297571~menuPK: 64166739~pagePK:64166689~piPK:641666 46~theSitePK:469043~isCURL:Y,00.html). The measurement of efficiency generally Notes requires an estimation of costs, an estimation 1. Evaluations by the Joint Monitoring of output, and a comparison of the two. The Programme conducted by the World Health larger the output is in relation to a given Organization and UNICEF reveal that the input, the more efficient is the activity MENA region is on track to meet the United (Afonso, Schuknecht, and Tanzi 2010). Nations’ Millennium Development Goals for 6. Quality of care offered by the health provider clean water and sanitation. Independent at a basic health care facility in Morocco was analyses, however, indicate that these assess- reported by only 9 percent of patients as the ments failed to take into account water qual- most important factor influencing choice, ity, affordability, and sustainability, resulting whereas quality of care was a bigger factor in in overstated coverage rates for clean water choosing among hospitals, in particular and sanitation (Zawahri, Sowers, and university hospitals: 51 percent of the respon- Weinthal 2011). Moreover, population pres- dents at university hospitals reported choos- sure, increased levels of use, and declining ing that facility because of the quality of the supplies place many Arab countries in danger care provided, compared with 31 percent for of falling under the absolute water-poverty other types of hospitals (QSDS, Morocco, level by 2050 (El Zein et al. 2014). 2009). 2. UNRWA operates one of the largest nongov- 7. Among the corruption cases processed in ernmental school systems in the Middle East. 2005, nearly 20 percent involved education Operating in five areas (the West Bank, Gaza, agencies or services, and an additional Jordan, Lebanon, and Syria), it manages 2 percent were related to higher education nearly 700 schools, hires 17,000 staff, and ministries. When combined, education- educates more than 500,000 refugee students related reports of corruption are more preva- a year. lent than corruption cases related to any T H E S TAT E O F E D U C AT I O N A N D H E A LT H S E R V I C E S D E L I V E RY 63 other services (health, public works, water, El Hassan, K. 2013. “Quality Assurance in Higher etc.). In 2006, education-related corruption Education in 20 MENA Economies.” Higher complaints contributed 9 percent of all Education Management and Policy 24, 2. reported corruption complaints, in third h t t p : / / d x . d o i . o r g / 1 0 . 17 8 7 / h e m p - 2 4 place after finance and agricultural irrigation -5k3w5pdwjg9t. as the specific domains reporting corruption. El-Kogali, S., and C. Krafft. 2015. “Expanding However, when corruption cases related to Opportunities for the Next Generation: Early higher education (4 percent) are added to the Childhood Development in the Middle East general education prevalence, education is and North Africa. Directions in Development: once again more prevalent than any other Human Development.” Washington, DC: specified sector. In 2007, higher education World B a n k G roup. ht t p: //do c u ment s itself constituted approximately 11 percent of .worldbank.org/curated/en /2015/01/23811898 all reported corruption cases, with general /expanding-opportunities-next-generation education contributing about 8 percent of all -early -childhood -development-middle-east reported corruption cases. -north-africa. 8. These data should be interpreted with cau- El Zein, A., et al. 2014. “Health and Ecological tion and not taken as definitive evidence that Sustainability in the Arab World: A Matter of corruption is highest in education and health Survival.” The Lancet 383 (9915): 458–76. care. Because of the salience of education and Ersado, L., and J. Gignoux. 2014. “Egypt: health care to people’s daily lives, it may be Inequality of Opportunity in Education.” that acts of corruption in these areas are more Policy Research Working Paper 6996, World frequently cited than cases of corruption in Bank, Washington, DC. other areas. Gatti, R., et al. 2013. Jobs for Shared Prosperity: Time for Action in the Middle East and North Africa. Washington, DC: World Bank. Grigoli, F. 2014. “A Hybrid Approach to References Estimating the Efficiency of Public Spending Afonso, A., L. Schuknecht, and V. Tanzi. 2010. on Education in Emerging and Developing “Public Sector Efficiency: Evidence for E conom ies.” No. 14 -19, I nter nat ional New EU Member States and Emerging Monetary Fund, Washington, DC. Markets.” Applied Economics 42 (17): ILO (International Labour Organization). 2004. 2147–64. Pathways to Work. Geneva: ILO. Baporikar, N., ed. 2014. Handbook of Research Jabbour, S., and R. Yamout, eds. 2012. Public on Higher Education in the MENA Region: Health in the Arab World. Cambridge, U.K.: Policy and Practice. Hershey, PA: Information Cambridge University Press. Science Reference. Keefer, P., and S. Khemani. 2005. “Democracy, Bold, T., and J. Svensson. 2010. “Baseline Data P u b l i c E x p e n d i t u r e s , a n d t h e Po o r : Report from the Community Score Card Pilot Understand i ng Political I ncentives for in Ismailia Governorate.” I nstitute for Providing Public Services.” World Bank International Economic Studies, Stockholm Research Observer 20 (1): 1–27. University, Stockholm, Sweden. Kelishadi, R. 2007. “Childhood Overweight, Chahine, A., et al. 2014. “Situation Analysis of Obesity and the Metabolic Syndrome in Youth in Lebanon Affected by the Syrian Developing Countries.” Epide miologic Crisis.” http://www.unfpa.org.lb/Documents Reviews 29: 62–76. /Situation-Analysis-of-the-Youth-in-Lebanon Kronfol, N. M. 2012. “Historical Development of -Affecte.aspx. Health Systems in the Arab Countries: A de Onis, M., and M. Blössner. 2000. “Prevalence Review.” Eastern Mediterranean Health and Trends of Overweight among Preschool Journal 18 (11): 1151–56. Children in Developing Countries.” American Mirmiran, P., R. Sherafat-Kazemzadeh, S. Jalali- Journal of Clinical Nutrition 72 (4): 1032–39. Farahani, and F. Azizi. 2010. “Childhood Diwan, I., P. Keefer, and M. Schiffbauer. 2014. Obesity in the Middle East: A Review.” Eastern “On Top of the Pyramids: Cronyism and Mediterranean Health Journal 16 (9): 1009–17. Private Sector Growth in Egypt.” Working ht t p: // w w w. n cb i . n l m . n i h . gov / pub m e d paper, World Bank, Washington, DC. /21218730. 64 TRUST, VOICE, AND INCENTIVES Mokdad, Ali H., et al. 2014. “The State of ———. 2013b. Fairness and Accountability: Health in the Arab World, 1990–2010: An Engaging in Health Systems in the Middle Analysis of the Burden of Diseases, Injuries, East and North Africa. The World Bank and Risk Factors.” The Lancet 383 (9914): Health Nutrition and Population Sector 309–20. S t ra t e g y fo r M E N A ( 2 0 1 3 – 2 0 18 ). Montenegro, C. E., and H. A. Patrinos. 2014. Washington, DC: World Bank. “Compa rable E sti mates of Ret u r ns to Zaky, H. H. M., H. A. S. Khattab, and D. Galal. Schooling around the World.” Policy Research 2007. “Assessing the Quality of Reproductive Wo r k i n g P a p e r 7 0 2 0 , Wo r l d B a n k , Health Services in Egypt via Exit Interviews.” Washington, DC. Maternal and Child Health Journal 11 (3): Mottaghi, L. Forthcoming. Arab Success Stories. 301–6. Spend More, Spend Better: An Efficiency Zawahri, N., J. Sowers, and E. Weinthal. 2011. Frontier Approach in MENA. Washington, “The Politics of Assessment: Water and DC: World Bank. Sanitation M DGs in the Middle East.” Patrinos, H., H. Abul Hamid, J. Reyes, and Development and Change 42 (5): 1153–178. J. Kelcey. 2013. “Learning in the Face of ht t p : // w w w. n cb i . n l m . n i h . g ov/ p u b m e d Adversity.” World Bank, Washington, DC. /22175085. Phillimore, P., et al. 2013. Health System Challenges of Cardiovascular Disease and Diabetes in Four Eastern Mediterranean Countries. Global Public Health 8 (8): Data sources 875–89. Arab Barometer, 2010–11 (Wave II), http://www Rahim, H. F. A., et al. 2014. “Non-communicable .arabbarometer.org/instruments-and-data Diseases in the Arab World.” The Lancet -files. http://www.transparency.org/research 383 (9914): 356–67. /cpi/overview Rijkers, B., C. Freund, and A. Nucifora. 2014. D e m o g r aph i c a n d H e a lt h S u r ve y, A r ab “All in the Family: State Capture in Tunisia.” Republ ic of E g y pt , 20 08 , ht t p: // w w w World Bank, Washington, DC. .dhsprogram.com UNHCR (United Nations High Commissioner for Demographic and Health Survey, Morocco, 2003, Refugees). 2013. “Syrian Refugees Living 2010–11, http://www.dhsprogram.com Outside Jordan Camps.” UNHCR, Geneva. D e mo g raph i c a nd H e a lt h S u r ve y s , I C F ———. 2014. “Health Update — Lebanon.” International, http://www.icfi.com/markets UNHCR, Geneva. /health/surveys-and-data-collection United Nations. 2014. Situation Analysis of E d S t at s , World B a n k , ht t p: //d at atopi c s Youth in Lebanon Affected by the Syrian .worldbank.org/education/ Crisis. New York: United Nations. EHGS (Egypt Health and Governance Study), Veillard, J. H. M., A. D. Brown, E . Barıs ¸, 2 010 , World B a n k , ht t p: //do c u m e nt s G. Permanand, and N. S. Klazinga. 2011. .wo rl d b a n k . o r g /c u r a t e d /e n / 2 010 “Health System Stewardship of National /06/16332545/egypt -management -service Health Ministries in the WHO European - qu a l it y-pr i m a r y-he a lt h- c a re -fac i l it ie s Region: Concepts, Functions and -alexandria-menoufia-governorates Assessment Framework.” Health Policy GAC (Governance and Anti-Corruption) Country 103 (2): 191–99. Diagnostic Survey, World Bank, http://web World Bank. 2007. Municipal Management and .worldb a n k .or g / W B S I T E / E X T E R N A L Loc al Govern ance: A Ser vice Deliver y /TOPICS/EXTPUBLICSECTORANDGOVE Perspective. Washington, DC: World Bank. RNANCE/0,,contentMDK:23399116~page ht t p s: //op e n k nowl e d g e .worldb a n k .or g PK : 210 058 ~pi PK : 210 0 62~t he Site /handle/10986/19616 License: CC BY 3.0 PK:286305,00.html IGO. Gallup World Poll, 2013, http://www.gallup.com ———. 2013a. Challenges in the Health Sector in /services/170945/world-poll.aspx Morocco: Evide nc e from E x pe n diture Globa l C or r upt ion B a rometer, 2013, Tracking and Service Delivery Surveys . Transparency International, http://www Washington, DC: World Bank. .transparency.org/research/gcb/overview T H E S TAT E O F E D U C AT I O N A N D H E A LT H S E R V I C E S D E L I V E RY 65 Local Governance Survey, Republic of Yemen, 8956~piPK:216618~theSitePK:490130,00 2 0 1 3 , h t t p : / / w w w.y e m e n p o l l i n g . o r g .html /consultancy/ SABER (Systems Approach for Better Education PETS (Public Expenditure Tracking Survey), Results), World Bank, http://saber.worldbank World Bank, ht tp: //web.worldbank.org .org/index.cfm / W B SI T E / E X T ER NA L / TOPIC S/ E X T TIMSS (Trends in International Mathematics and SOCIALDEVELOPMENT/EXTPCENG/0,, Science Study), Boston College, ht tp: // contentMDK:20507700~pagePK:148956~pi timssandpirls.bc.edu/ PK:216618~theSitePK:410306,00.html Transitional Governance Project, 2014, http:// PIRLS (Progress in International Reading Literacy transitionalgovernanceproject.org/ Study), Boston College, http://timssandpirls.bc UNICEF MICS (UNICEF Multiple Indicator .edu/ Cluster Sur vey), ht tp: //w w w.unicef.org PISA (Programme for International Student /statistics/index_24302.html Assessment), Organisation for Economic World Development Indicators (database), World Co-operation and Development, http://www Bank, http://data.worldbank.org/data-catalog .oecd.org/pisa/ /world-development-indicators QSDS (Quantitative Service Delivery Survey), World Health Statistics, 2010, World Health World Bank, ht tp: //web.worldbank.org Orga n i zation , ht t p: //w w w.who.i nt /g ho / W BSI T E / E X T ER NA L / TOPIC S / E X T /publications/world_health_statistics/en/ POVERTY/EXTPSIA/0,,contentMDK:20467 Yemen Polling Center, http://www.yemenpolling 190~isCURL:Y~menuPK:1108016~pagePK:14 .org/consultancy/ Local Successes: Satisfaction, Accountability, and Quality 3 at the Local Level • To understand service delivery, one must conduct an in-depth examination. The positive deviance approach can reveal how the elements of local solutions have emerged. • A top-performing rural school in the northern West Bank traces its success to active parental engagement, effective coordination with the school district, and the ability of the school principal to build a motivating and encouraging work environment for its teachers. • In Jordan, an excellent school in a tribal community benefits from a vibrant Education Council trusted by the community for its transparency and inclusive decision making and from school grants allowing some autonomy. • Top health care clinics in Jordan benefit from their partnership with local social institutions and health committees and from formalizing health management procedures at the local level. • The best-performing rural health clinics in Morocco effectively draw on their strong partnerships with local communities as well as positive competition and support devised by the Ministry of Health. • These case studies reveal the importance of building trust, engagement, and accountability in service delivery at the local level. Lessons from these experiences can be part of (though not a substitute for) a broader strategy for promoting systemwide institutional reform. A s described in chapter 2, countries in remarkably high. In chapter 7, we draw on the Middle East and North Africa household surveys to document the nature (MENA) face a challenge in improv- and extent of subnational variation in ing the quality of education and health ser- service provision indicators across selected vices delivery; citizens are not satisfied with MENA countries. In this chapter, we seek the services they are receiving. If the average depth rather than breadth, using case stud- quality of services is poor, however, that ies of four instances of highly effective situation does not prevail always and every- service delivery to explain how regional and where. In some places, the standard of edu- local factors can align to yield positive cation and health care that citizens receive is outcomes. 67 68 TRUST, VOICE, AND INCENTIVES Data drawn from household surveys can and where citizens’ expectations of their provide an abundance of useful information government increases along with rising on where, why, and for whom the quality of wealth, education, and political openness. service delivery varies. But many factors that To demonstrate the nature and signifi- cannot be adequately captured in household cance of these local factors in shaping service surveys, such as the effectiveness of local delivery performance, we present in this leaders, also matter. To more fully appreciate chapter four case studies of success: educa- the significance of the social processes by tion services in the West Bank and Jordan which specific combinations of factors and and health services in Jordan and Morocco. choices shape outcome variation, we need to Each case was selected on the basis of a drill deeper, using different research tools. broader analysis of variation in service deliv- This is especially so if we wish to explain and ery within each economy and because it dem- replicate unusually successful cases, because onstrates a particularly instructive array of some of the key factors driving these out- strategies that have been deployed in response comes will only be discernible (observable) to local (but not atypical) problems. through qualitative methods, and not all of these factors (or combinations of factors) will be reproducible elsewhere. Some communi- Case study 1: Education services ties may do well merely because of political in Jenin, West Bank connections that yield a steady flow of Beyond its intrinsic value, education is an material resources—a factor that cannot be important investment in human capital that readily assessed via a household survey, nor offers people a sense of hope and optimism one that would be invoked as a policy recom- for a better life in a world characterized by mendation for improved performance else- seemingly insurmountable barriers, soaring where. Similarly, it may turn out that an rates of youth joblessness, and deep uncer- idiosyncratic combination of factors drives tainty about what the future may hold for both positive and negative deviance1 in each them. In recent years, the education system setting, making it difficult to identify general in the West Bank and Gaza has achieved con- implications for development policy, which siderable gains and performed well on major necessarily must be articulated at broad levels education outcomes, attaining levels compa- of societal aggregation. rable with those of middle-income countries. Even so, examining cases in which com- In 2011 the literacy rate in the West Bank munities have managed, often despite diffi- and Gaza for adults aged 15 and over reached cult circumstances, to attain extraordinary 95 percent, which is much higher than the outcomes using innovative local solutions to rate in countries with similar per capita prevailing problems can provide many useful incomes such as India and the Arab Republic insights for practitioners and policy makers. of Egypt and similar to that in much richer Crucially, such insights can also be a source neighbors in the region such as Turkey, of hope and inspiration, showing citizens and whose per capita income is seven times that fellow professionals alike that someone, of the West Bank and Gaza, and Jordan, somewhere, somehow has found a better way whose per capita income is three times to do things even though they faced similar greater (Krishnan et al. 2012). Enrollment in obstacles and problems. Building service basic education is universal. In 2012 the delivery systems that recognize and reward enrollment rate for secondary education was improvement in implementation quality is a 81 percent, and, equally important, the central challenge for development in the 21st enrollment rate in tertiary education was century. This is true everywhere, but espe- above 40 percent for the 18–24 age group the cially in the MENA region, where the conse- same year (Krishnan et al. 2012), which is quences of success and failure often feed so high when compared with the rate for immediately into sensitive social tensions, middle-income countries. In addition, LO C A L S U C C E S S E S : S AT I S FA C T I O N , A C C O U N TA B I L I T Y, A N D Q U A L I T Y AT T H E LO C A L L E V E L 69 access to basic and secondary education is classrooms and insufficient financial highly equitable with respect to gender, loca- resources to provide the basic educational tion, refugee status, and household income. materials. In addition, mobility restrictions High enrollment rates and educational on students and their continual exposure to expenditures of 4.9 percent of the gross the psychological trauma arising from the domestic product (GDP) are all the more ongoing conflict have taken a toll on stu- noteworthy considering the difficult eco- dents’ abilities to experience an ordinary nomic and political situation facing the West learning environment in which they can Bank and Gaza (Claussen, Kiernan, and thrive, personally and intellectually. Gramshaug 2013). 2 In contrast to other Beyond the political instability, which countries in the developing world, teacher arguably is the main constraint in improving absenteeism is not a major problem. Teachers education services, the education system in do show up to teach and seem relatively the West Bank and Gaza has long relied on accountable to their clients. 3 In addition, obsolete and outdated pedagogical methods, parents seem to be heavily invested in their characterized by teacher-centered learning children’s education, as demonstrated by with an emphasis on rote memorization and their high levels of in-kind and financial con- basic numeracy. Consequently, students are tributions to schools, particularly in rural not fully engaged, perform poorly, and lack areas. Nevertheless, achieving universal the essential critical thinking and problem- enrollment access is only one step toward solving skills needed for survival in today’s ensuring an educated and productive popula- highly competitive global economy. The costs tion able to compete on the world stage and, to the society emerging from such a system more immediately, to survive the many seem enormously high because high- uncertainties it confronts. Indeed, the focus productivity skills such as innovation and of education policy has recently shifted entrepreneurship are lacking and yet are toward quality of learning because major urgently needed by an economy with high weaknesses continue to prevail in the class- youth unemployment rates, 4 a saturated room (as highlighted by the average student’s public sector, and high population growth of modest performance in national and interna- 2.3 percent (Krishnan et al. 2012). These tional assessments). Students’ outcomes in costs are particularly high for boys, who tend the 2011 Trends in International Mathematics to withdraw from education sooner than and Science Study (TIMSS) were low when girls. Their enrollment rates are more suscep- compared with those of other countries with tible to a weak socioeconomic status, amid similar GDP per capita or per student expen- social norms that focus on supporting educa- diture, and they fall below the international tion for girls and preparation for adulthood. average. For example, the share of eighth- Boys are also under pressure to resume the grade students scoring at least 475 on the breadwinner role when family resources are science assessment (classified as an intermedi- constrained. ate international benchmark) was only 33 percent, whereas the share for the mathe- Kufor Quod Girls’ Secondary School matics assessment was 25 percent (EdStats). The perennial difficulties experienced by The weak overall learning outcomes for residents of the West Bank and Gaza in students in the West Bank mask a high level improving the quality of education stem of variation across the performance spec- largely from the highly complicated political trum, with a few schools scoring well above climate to which they have been subjected for the national average in the TIMSS and over a decade. The region’s economy is highly Tawjihi national assessments. This result sug- dependent on Israel, and its capacity to gests that in these schools, which face con- support schools is highly strained. Schools straints similar to those faced by their peers, in many rural areas suffer from small potentially innovative approaches and local 70 TRUST, VOICE, AND INCENTIVES practices are contributing to education during our visit to Jenin’s education director- quality in ways that could be generalized and ate, one of the 16 field directorates spread thereby serve as a potential source of inspira- throughout the West Bank. The Ministry of tion to schools at the lower end of the Education and Higher Education (MOEHE), spectrum. To this end, we visited one of the through the directorates, provides schools top-performing schools on the 2011 TIMSS, with classrooms, textbooks, and other mate- Kufor Quod Girls’ Secondary School in Jenin, rials, in addition to managerial and technical to try to identify some of the innovative in- supervision.6 In addition, during its prepara- school practices and governance mechanisms tion for the 2011 TIMSS, the school, similar underpinning its performance.5 In doing so, to other schools in Jenin, received rigorous we sought to document some of the ways in support from the directorate.7 The compre- which the different stakeholders—school hensive, structured planning process for the directorate, principal, teachers, students, and TIMSS focused on two efforts. The first was parents—were working together to address forming at the directorate level a committee particular challenges and, in the process, whose sections had different roles and enhance the quality of learning. responsibilities—for example, the supervi- The Kufor Quod Girls’ Secondary School sors’ section designed and developed is in the small village of Kufor Quod, located capacity-building programs and conducted a few kilometers from the city of Jenin in the field visits to provide pedagogical and content north of the West Bank. The school has support to the science and math teachers, and almost 300 students and about 25 full-time the external relations section enhanced the teachers, as well as a principal and a engagement of parents and the local commu- counselor. The school’s performance in the nity by conducting awareness campaigns 2011 TIMSS was relatively high. The average about the importance of the TIMSS. The sec- score for eighth-grade students in science was ond area was encouraging the implementa- 529 and in mathematics 500, placing them in tion of several activities believed to play an the fi rst rank for the highest science average important role in increasing students’ TIMSS score and in the second rank for the highest results, such as developing simulation exer- mathematics average score among all public cises based on questions similar to those in schools in the West Bank and Gaza. Both the TIMSS, conducting frequent competi- scores placed the school above the intermedi- tions at the directorate level (including all ate international benchmark. 144 schools), and investing in social plat- The school, compared with its peers and forms (such as creating a Facebook page) to especially boys’ schools, enjoys a relatively support ongoing discussions between stu- favorable environment that provides a com- dents and provide supporting materials. forting space for learning. Violence and disci- Such thorough planning and preparation pline problems in the classroom are rare. played an important role in raising students’ The number of students in each classroom is TIMSS performances at the directorate level relatively small, enabling teachers to dedicate in general and at Kufor Quod in particular. more time to each student’s individual learn- A sense of genuine pride and accomplishment ing needs. This also enables teachers to move was felt by all involved—staff at the director- vertically across grades (as opposed to teach- ate level, principal, teachers, students, par- ing horizontally all sections within one ents, and the local community—as a result of grade), which could in some ways be advan- having achieved relatively high scores at the tageous to students who might benefit from national level and enjoying the acclaim of having a teacher who is already familiar with being able to represent their country interna- their needs and challenges. In addition, the tionally. The deputy director at the director- school enjoys a high level of parental involve- ate and his staff, who were the key actors in ment and commitment to education—a fac- the major TIMSS preparations, were even tor common to all schools in Jenin, as noted nicknamed “the TIMSS Men.” In all of this, LO C A L S U C C E S S E S : S AT I S FA C T I O N , A C C O U N TA B I L I T Y, A N D Q U A L I T Y AT T H E LO C A L L E V E L 71 the truly remarkable and innovative part was PHOTO 3.1 Abla Habayeb, principal of Kufor that the directorate did not lose track of the Quod Girls’ Secondary School goal—enhancing the education outcomes of students without restricting them to achiev- ing only high TIMSS scores—while also har- nessing the energy and pride gained from achieving high TIMSS scores to push for even higher educational attainment, including in those subjects not formally part of the TIMSS process. The directorate’s effort was particularly successful and most visible in Kufor Quod because of the outstanding leadership exhib- ited by the school’s principal, Ms. Abla Habayeb (see photo 3.1). Over a long career, she has forged remarkably productive and robust relationships with all the parties involved based on reciprocal trust and respect, thereby benefiting from informal networks and tight intercommunity relations. These relationships ensure that accountability mech- anisms work in the right direction and are further harnessed in the pursuit of higher- quality education. In terms of her upward relationship with the directorate, the princi- Source: © Jumana Alaref and Michael Woolcock / World Bank. Further pal’s years of service (29 of them as a princi- permission required for reuse. pal) have earned her an excellent reputation, which she uses in her relationship with the are incorporated into an institutionalized directorate to gain its trust. In turn, she has process that offers other schools guidance, earned the right to receive considerable auton- inspiration, and accountability. To this end, omy to pursue initiatives on her own and seek we now explore some of the broader institu- the directorate’s ongoing support. Her seem- tional processes enabling Ms. Abla to oversee ingly boundless energy and motivation of an effective learning environment. teachers, students, and parents, in addition to her regular consultations with them, further Accountability mechanisms at work increase their commitment and ensure their sustained cooperation as part of a collective Between the school and Jenin’s education quest to improve the school’s learning out- directorate. The midlevel relationships comes and physical environment. between the service providers (schools within In short, Kufor Quod’s impressive aca- the directorate) and Jenin’s directorate are demic performance and dedication to educa- characterized by seemingly effortless direct tion are largely attributed to the school’s communications. In establishing an account- principal, who has made it her personal mis- ability relationship with school principals, sion to aim high and deliver the best educa- the directorate has been able to strike the tion possible to her students, especially in right balance in ensuring that principals are view of the many constraints they encounter accountable for their work and expected to on a daily basis. From a policy perspective, deliver results according to a predefi ned set the challenge remains of setting up a system of clear guidelines, while giving them auton- that ensures that her local practices and inno- omy and space during implementation to find vations are not personality-driven but instead context-specific solutions. Principals are 72 TRUST, VOICE, AND INCENTIVES entrusted to do their work and duties in and helping to formulate a comprehensive innovative ways they themselves deem fit, plan that outlines issues and solutions. For while the directorate supports them consis- example, the science teacher at Kufor Quod tently, enhances healthy competition between Girls’ Secondary School who helped prepare them, and monitors and evaluates them as students for the 2011 TIMSS talked about reflected in students’ overall performance, her experience and participated in training teachers’ feedback, and the level of local sessions with other teachers. In this way, community engagement. During preparation supervisors are an important accountability for the TIMSS, the sense of partnership and channel for teacher monitoring and the provi- collaboration between the two was rein- sion of support in updating teachers’ peda- forced by the participatory approach the gogical practices, bearing in mind that the directorate took through workshops that supervisors themselves still largely adhere to sought the principals’ input on the director- the traditional way of evaluating teachers— ate’s plan and their suggestions for improving that is, their ability to cover all the material and enhancing cooperation. In her efforts to on time—leaving little room and few incen- internally push her staff and students toward tives for teachers to innovate or venture into exerting more effort, Ms. Abla capitalized on different models of teaching. this partnership with the directorate by Although this collaborative framework working hard to earn the trust of both staff does not yet set the stage for complete school- (such as by offering professional develop- based management, it is a refreshing depar- ment) and students (by actively involving ture from a rigid centralized system to one in them in numerous decisions8) in return for which the directorate works with the schools their ongoing support. and their communities to help prioritize school One way in which the directorate is able to needs and determine how best to support stu- support schools is through teacher supervi- dents. This vision was adopted by the director sors. Supervisors (1) assess teachers’ perfor- as a result of her personal conviction and was mances and provide content and pedagogical not implemented at the level of all directorates support; (2) design, develop, and implement by the Ministry of Education. This suggests training programs for teachers on methods of that the success of this framework in support- teaching, classroom management, and mea- ing all schools in Jenin, and its particular ben- surement and evaluation; and (3) provide efits for Kufor Quod Girls’ Secondary School feedback to the Palestinian Curricula Center (mainly because of the principal’s ability to on the newly developed textbooks. The fully capitalize on it), is of high potential value supervisors’ role has recently evolved from a for other directorates in the West Bank and purely authority-based relationship with Gaza seeking to learn from and implement teachers (based on a rigid focus on progress new ideas in their own schools. through the official curriculum) toward a Between the school and its teachers. The more flexible one that concentrates on men- directorate’s philosophy in dealing with its torship, support, and cooperative learning.9 own schools has been echoed by the principal Within Jenin’s directorate, supervisors have in her internal accountability mechanism begun using new criteria in assessing teach- with teachers. The nature of the relationship ers’ performance. These criteria are centered between the principal and teachers is cen- on evaluating the gradual shift in teachers’ tered on continual support, which in turn is pedagogical practices from purely lecturing translated positively into the level of effort to student-centered learning based on discus- exerted by teachers and reflected in students’ sions and debates. In addition, supervisors academic performances. The principal sup- serve as a conduit for transferring effective ports teachers in the following ways: and innovative approaches between class- rooms and schools, inviting well-performing • Continual process of syste matic teachers to provide training for other teachers e n c o u ra ge m e n t a n d n o n m o n e t a r y LO C A L S U C C E S S E S : S AT I S FA C T I O N , A C C O U N TA B I L I T Y, A N D Q U A L I T Y AT T H E LO C A L L E V E L 73 incentive schemes. Despite her limited modules implemented elsewhere that autonomy and lack of ability to reward could be adopted in her school. high-performing and exceptional teach- Meanwhile, teachers’ commitments ers with monetary bonuses and rewards, to constantly improving have fostered the principal ensures that her staff mem- an environment of cooperation aimed bers are constantly appreciated and at enhancing student learning. Teachers receive the incentives they need to sustain often fill in for each other in cases of high-quality work. Teachers reported unforeseen absenteeism. In addition, that such appreciation takes the form of they exchange knowledge by attending fair and equal treatment of all teachers each other’s classes and adopting creative and verbal and public encouragement of teaching techniques such as pairing low- exceptional teachers in front of the entire performing students with high-performing school when test scores are publicized. ones in group homework. The result is a relaxing environment in • Participatory approach to the decision- which teachers are able to perform their making processes at the school level. The job properly, with little rivalry or tension principal frequently consults the teachers between them. when she must make decisions that affect • Professional development and in- service them and the students. For example, the training. As part of their professional school improvement plan is developed development, teachers in Kufor Quod collaboratively by the principal, coun- receive regular training by the Ministry selor, and the math, science, technology, of Education and external entities, mainly and Arabic teachers in order to reflect nongovernmental organizations and students’ needs (such as any particular international donors. The school’s lack weaknesses among students requiring of material resources has not diverted the more support) or staff needs (such as principal from the importance of focus- any necessary training) before the plan is ing on the factors inside the classroom shared with the directorate. Teachers are that have an impact on learning, mainly also consulted on the budget allocation improving teaching pedagogy. Therefore, process before it is sent to the directorate based on her assessment of teachers’ so that the budget reflects their needs. needs, the principal nominates her teach- • Teachers in turn reciprocate the com- ers for training courses and releases the mitment. The framework of partnership time for them to participate, taking into and trust that the principal has estab- consideration the availability of training, lished with teachers has allowed her to its location, and a teacher’s workload. retain the best and most dedicated of However, as argued by teachers during them and provide incentives for everyone our interviews, the training offered may to work harder and be more committed. not necessarily translate into updated She has extended those incentives to, for pedagog y or more ef fec t ive teach- example, the math and science teach- ing, highlighting the need for a more ers who prepared students for the 2011 structured system that is in line with TIMSS and have a reputation (according international standards and that helps to supervisors) for exhibiting special in- teachers upgrade their skills and venture class practices, openness to recommen- into innovative models of teaching. Thus dations and feedback, knowledge of the rather than asking for fi nancial contribu- TIMSS, and a desire to create a student- tions or better technological and material centered environment. In another exam- resources for her school (which is clearly ple of exceptional dedication by teachers a need), the principal requests training to their students, those at the school did courses for her teachers that acquaint not take part in the frequent strikes called them with the successful education by teachers and unions across the country 74 TRUST, VOICE, AND INCENTIVES to demand higher wages. They asserted pursue in their last two years of high on paper that they were on strike, but school—scientific, literary, or commercial.10 in reality—and despite being subject to The principal also uses innovative strategies pressures from the unions—they con- to reach out to parents who are not as actively tinued to teach classes in order to better involved in their child’s education, such as prepare students for their exams. When inviting illiterate parents to attend first-grade asked why they would do such a thing, classes to learn how to read and write. one of the teachers shrugged and said, Engagement by parents and the local com- “Oh, but my conscience would not allow munity extends to issues beyond the class- me to not teach to the best of my capa- room and includes their participation in the bilities.” We then asked, “But imagine decision-making process. Parents, who are you were working in a different school viewed as partners, become heavily invested where you did not feel quite as appreci- in their children’s school. In many instances, ated, would you still do the same thing?” they contribute to the school by providing The teacher replied, “No, I would not. material resources in response to the often- It is only natural to give back what you heard plaint by the Ministry of Education receive in the fi rst place.” that its financial capabilities are limited. Examples of parent involvement include con- Between the school and its students, par- structing the second floor of the school build- ents, and the local community. In a commu- ing and helping to provide a computer lab. nity characterized by tight intercommunity The collaborative environment in the relations, complex political dynamics, and a school has created a safe learning environ- highly constrained government, the support ment for students—one that provides an provided by parents and the local community appropriate platform for students to exert is essential to the principal in sustaining the more effort in the classroom. Students regard quality of her school and pushing forward. their teachers as their “friends” and the prin- To the principal’s advantage, the local com- cipal “as a mother-like figure.” They feel munity consists of highly engaged parents their voices and concerns are heard, and they committed to providing education for their are treated equally, without discrimination, children, often against all odds. Therefore, based on their level of academic perfor- parents’ involvement reflects a very high level mance. Moreover, students with special of awareness of their children’s academic per- needs (such as those with visual or learning formance and behavior in the classroom. The difficulties) receive ongoing support from the principal has formalized such engagement administration and students alike, mirroring through many channels. For example, par- the broader cultivated attitude that everyone ents provide their own feedback and signa- in the school has an equal role to play in the ture on every item of homework to ensure learning process. t h at t h e y m o n i t o r t h e i r c h i l d r e n’s performance. And parents participate in Remaining challenges activities in the classroom. Furthermore, the principal has established a more formal Kufor Quod Girls’ Secondary School is an structure through the school’s parent-teacher exceptional case of what can happen when a association, which she has used as an avenue school’s leadership works with its directorate, to discuss students’ performance with their teachers, parents, and students as one collab- parents before scheduled midterm and fi nal orative unit to optimize the education offered exams and afterward. The Education Career by the school. The question of whether the Council she has established is an attempt to model Ms. Abla (with the support of her supe- stimulate discussions among parents about riors) has been able to implement within the the need to give their children the freedom to boundaries of her school is sustainable choose the academic track they would like to remains, nevertheless, an open-ended one. LO C A L S U C C E S S E S : S AT I S FA C T I O N , A C C O U N TA B I L I T Y, A N D Q U A L I T Y AT T H E LO C A L L E V E L 75 There continues to be room for improvement the end of the academic year, leaving little in having her local best practices “routinized” room for other types of learning activities; into an approach that ensures its continuity (2) the criteria used by supervisors to evaluate after someone else takes over and slowly inte- teachers, which largely rest on teachers’ grating this approach into a structured gover- abilities to teach all the material “on time” nance system that is not personality-dependent. and their ability to recite the information Such a step would enhance the likelihood that accurately to students, leaving no incentives other schools in the West Bank and Gaza for teachers to innovate or venture into differ- would also benefit from her innovations and ent models of teaching that rely more on in turn would be accountable for generating active engagement between students and improvements of their own. teachers; and (3) a reliance on exams that tend Meanwhile, the potential and dedication to merely measure students’ abilities to learn offered by teachers and students demonstrate the information by heart, leaving little room that Kufor Quod has the potential to improve for creative thinking, problem solving, and even further. The TIMSS results achieved by innovation. Limited by the system’s boundar- the school are above the intermediate inter- ies, any professional development practices national benchmark, but they still fall short developed for teachers are thus tailored to of reaching the advanced and high interna- what the system demands that students learn, tional benchmarks. Although the system leaving no chance for promoting new learning established by Ms. Abla may well safeguard paradigms and innovative practices.11 Such a her students’ relatively high achievements and system is particularly costly to those living in prevent them from deteriorating, a lingering the West Bank and Gaza because their politi- question does remain: how does one help a cal reality and nearly crippled labor market school such as Kufor Quod improve even fur- necessitate more than ever finding innovative ther and leapfrog to the standards of modern practices and unconventional methods to education systems that fully prepare their break out of the box in which they find them- students for the global economy? The answer selves and thinking of alternative routes to to that question is not necessarily found in mitigate the effects of the economic and polit- having the administration or teachers under- ical hardships they face. take some new techniques but rather in the broader context of the current rigid educa- Conclusions and implications tion system in the West Bank and Gaza that provides very little room for flexibility, inno- When asked by our team about their dreams vation, or change. It is evident when listening and aspirations for the future, students at to students’ and teachers’ needs that students Kufor Quod did not hold back on giving do not have a sufficiently firm grasp of some impressive answers that revealed their desires relevant subject areas and, most important, to achieve much more than what the world an ability to deconstruct and reconstruct seems to expect from them.12 Ensuring that knowledge and apply it in varying contexts their dreams are not crushed by their harsh and to real-world situations. Unfortunately, circumstances requires a collective commit- students’ abilities to develop critical thinking, ment from policy makers and donors alike to problem solving, and broader skills sets are students acquiring the skills that would somewhat lagging. enable them to achieve what they are capable Understanding these limitations requires of and never give up on their dreams. Not examining the factors that seem to keep the just Kufor Quod students but all students in system locked in place without sufficient the West Bank and Gaza have the right to room for flexibility. These factors include dream and deserve a real chance to translate (1) the use of textbooks that are usually these dreams into reality. loaded with information with the expectation The various lessons and implications for that teachers will cover all the material before policy and practice arising from Kufor Quod 76 TRUST, VOICE, AND INCENTIVES Girls’ Secondary School could be extended to performance), but also ongoing support other schools to reduce the gap between aspi- for the principal by participation in the ration and reality, between the best schools decisions that affect the future of stu- and the rest. Any options for scaling up these dents and their school. successful local practices require a full- • Receiving the surrounding environ- fledged collaboration plan that moves in ment’s trust and support so that the three directions: upward with the director- principal can perform his or her duties ate’s relationship with the school, downward more smoothly and effi ciently. For exam- to the teachers, and then outward to the local ple, Kufor Quod’s development plan— community as part of a binding social com- which consists of ongoing monitoring pact to prioritize school needs and determine and evaluation, regular assessments how best to support student learning of at early stages to detect inefficiencies, increasingly complex issues. Specifically, and the development of remedy plans— several micro-level steps were taken by Kufor requires the collaborative involvement of Quod that other schools and directorates all stakeholders (by incorporating their could consider implementing. These include: input and various assessments) to realize its effective implementation. • Adopting the plan devised by the direc- tor of Jenin’s directorate that set up Beyond these specific insights, however, strong accountability mechanisms with perhaps the most important lesson learned is school principals by giving them suffi - to create space at the local level for schools to cient autonomy in making decisions and find and implement their own solutions to launching innovative initiatives at the their own particular problems. Learning col- school level. Such a plan creates a cul- lectively how to optimize, within the con- ture of partnership and aligns incentives straints the school and its community face, is by cultivating a broader attitude that all the first step to expanding the frontier across stakeholders have roles and responsibili- the education system itself. ties to fulfi ll in return for support, with the ultimate goal of serving students’ learning needs and supporting them as Case study 2: Education services much as possible. in Yarqa, Jordan • Within schools, building effective inter- Jordan is fortunate to have a long-standing action between principals and teachers. commitment to education. Unlike many of its Providing teachers with encouragement neighbors, Jordan does not depend on oil and support and ensuring that they are exports to sustain its national economy, and qualifi ed and regularly participating in thus since its inception as an independent opportunities for professional develop- state in 1946 it has recognized that “invest- ment are essential to motivating teachers ing in people” is the key to sustaining its and to cultivating and sustaining suc- social cohesion, political stability, and eco- cessful accountability mechanisms, not nomic prosperity.13 In a region that scores only with other educators but also with below international averages on educational broader professional norms. performance, we found Jordan to be a posi- • Externally, forging mutually support- tive outlier. Even so, there is considerable ive interaction among teaching staff, variability between schools and between gen- pare nt s, an d the loc al communit y. ders in the quality of the education that stu- Establishing transparent and genuinely dents receive. In the case we describe here, collaborative partnerships ensures not community members of the villages of Ira only ongoing involvement in students’ and Yarqa were able to work together, learning environment in the classroom through formal and informal networks, (which is essential in raising academic to identify and constructively address LO C A L S U C C E S S E S : S AT I S FA C T I O N , A C C O U N TA B I L I T Y, A N D Q U A L I T Y AT T H E LO C A L L E V E L 77 common educational challenges related to Jordan’s education system has long been service delivery. Through this success, their hailed as a star performer in a region that school cluster was able to (1) strengthen the consistently performs below the international citizen provider route for accountability; averages.14 With its advanced education (2) institutionalize a culture of inclusiveness, system regionally, Jordan has an 8 percent transparency, and collaboration in decision illiteracy rate, the third lowest rate in the making; and (3) institutionalize a process of Arab world. Its primary gross enrollment self-review, continual improvement, and ratio increased from 71 percent in 1994 to 99 enhanced services. percent in 2010, and its transition rate to secondary school increased from 63 percent to 98 percent over the same period (EdStats). Zeid Bin Haritha Secondary School Jordan also ensures a high level of gender Past the winding roads leading out of Amman parity in access to basic services. As a result, and deep into the municipality of Al Salt is the it has achieved 95 percent parity in literacy, village of Yarqa, a typical tribal community full parity in primary and secondary enroll- that sits amid the hills of serene rural Jordan. ment, and increased life expectancy for both Zeid Bin Haritha Secondary School, a modest- sexes. In terms of student learning outcomes, sized school located off of a main street, serves on the 2003 and 2007 TIMSS, Jordan’s 192 students from grades 5 to 12. Zeid Bin students outperformed many of their peers. Haritha is one school in a cluster of 11 schools However, most recently, in 2011, student out- in Ira and Yarqa, two small villages with a comes dipped considerably, especially among combined population of 9,400. Like some boys.15 other schools across Al Salt and other munici- Although the dip in student performance palities in Jordan, Zeid Bin Haritha’s build- came as a surprise to Jordan’s Ministry of ings and facilities are outdated and in need of Education and outside observers, the minis- renovation, classes are overcrowded and try has been actively undertaking new and poorly heated in winter, and teacher shortages innovative projects that could maintain are a common problem. At the helm of the improved student performance levels and school is principal Hatim Zaydoun, a tall, curb the recent declines. In doing so, the commanding figure, who projects an aura of ministry, as part of the second phase of the authority even to members of the parent- education development plan that began in teacher association (PTA), a difficult feat that 2009, embarked on a path toward greater has clearly consolidated Hatim’s position school autonomy and enhanced community within the community. engagement as one of its initiatives to improve The PTA of Zeid Bin Haritha (and others student learning outcomes. This process across Jordan) is not a typical parent-teacher includes implementation of the School and association. Its members are retired teach- District Development Program (SDDP), ers, parents whose children graduated long a donor-funded national program that aims ago from school, and practicing teachers. to develop schools and directorates by pro- Here, the PTA is one player in the new moting greater collaboration between schools approach to promoting school autonomy and communities to systematically determine and community engagement. Despite the and address school and local needs. These unconventional formulation of this PTA, efforts seek to foster (1) increased participa- there is no mistaking that education out- tion by local stakeholders in decision mak- comes and quality are at the top of the ing and priority setting; (2) greater efficiency agenda of this association and school. This and transparency in the use of resources; PTA, like many others recently established (3) increased engagement by parents and sup- in Al Salt, is the result of joint efforts by the port through financial or in-kind donations; local community and school to attain better (4) improved student learning and school education outcomes. environment; and (5) improved provision of 78 TRUST, VOICE, AND INCENTIVES education services that result in higher stu- This case study examines the ways in dent learning outcomes. which different groups involved in the day- Under the SDDP, the objectives of increased to-day operation of schools—teachers, school autonomy and community engagement students, parents, communities, and broader are being achieved by a school self-review pro- associational groups—work together (or not) cess that involves input from parents, teachers, to identify and constructively address and students16 in four key areas: teaching and common educational challenges related to learning, leadership and management, com- service delivery. munity engagement, and student environ- ment. Each school appoints a domain leader Challenge: A culture of disengagement for the four key areas, who then collects and analyzes the inputs of stakeholders against Ira and Yarqa are close-knit villages that standardized quality indicators to identify a have defined community structures based on set of priorities. The domain leaders form part tribal lineages and intracommunity relations of the School Development Team, led by the rooted in Jordan’s Bedouin and Islamic school principal, and together they draft a traditions. “We live in a rural small village. School Improvement Plan (SIP). The SIP is It’s like a large family. We hold evening events shared with the Education Council,17 which is all the time, family gatherings and the majlis, responsible for a number of schools within where issues are discussed and we hear from one area, together forming a cluster of about each other and the children about what’s 10 schools. The hope is that the school and happening at school,” explained the father of district improvement plans will allow better a student. Villagers describe their community insight into how communities in which the as “rural,” “tribal,” and “closed.” Here, government provision of social services is tribal and community practices of authority lacking and the lines of accountability are prevail; individuals with predetermined char- broken or blurred because of factors beyond acteristics are given the authority, albeit in a individual control can improve basic services tacit manner, to engage and advocate for the in an equitable manner. village. The close community relations and In Jordan’s rural villages such as Ira and high levels of acquaintance lead to frequent Yarqa, families depend on agriculture and interactions, both formal and informal. the local administrative public sector for their Through these interactions, citizens share income. Community members have main- stories and news from the community, build tained their roots in these villages for genera- ties, and establish a sense of belonging to the tions and enjoy a relatively calm and simple village. The interwoven relations within the life that is family-centered. But they fi nd it villages have made citizens comfortable with difficult to accept change. With the introduc- the pace of rural life in Jordan. Any changes tion of the SDDP and the push toward to this situation are easily detected and are at community engagement in schools through first viewed with cynicism and concern. joint school and community training, Because of the interconnectedness and community members and school principals ease of information flow in these villages, for- were hesitant and uncertain of the benefits mal means of communication at the school to be reaped from changing the current level are overlooked. After all, teachers and processes of managing schools. “We take our principals regularly see the parents of stu- role very seriously, and realize that it’s up to dents outside of school. And yet despite this us to determine what the school needs are informal mode of communication, parents for the upcoming year,” explained Hatim. and other members of the community shy “This is not the way we used to do it in the away from reaching out to teachers or school past. When we were told there needs to be principals on matters beyond those of student more community involvement in school prac- behavior or performance. The lines of author- tices we weren’t sure about that.” ity between the school and community are LO C A L S U C C E S S E S : S AT I S FA C T I O N , A C C O U N TA B I L I T Y, A N D Q U A L I T Y AT T H E LO C A L L E V E L 79 drawn so that teachers and the school princi- to communicate with a parent. However, the pal hold all the decision-making power inter- barriers to engaging with schools that stem nally, with little input and involvement from from socioeconomic factors such as the par- parents—a relationship with which both par- ent’s level of education, employment status, ents and the principal have been comfortable. occupation, social standing, and gender still However, recognizing the positive effects that prevail today. “I’m not afraid of bringing can be reaped from school and community issues up to a school principal or a teacher,” cooperation, educators in Jordan have said Um Amira. “My house is very close to explored options for improved student out- the local girls’ school and I can see a lot of comes that harness this approach to improved what is happening from my window. If any- student outcomes.18 thing is not to my liking or I see misconduct, They must begin, however, by addressing from a teacher or student, I have no problems two challenges: fi rst, how to involve parents reporting it to the principal right away. I am and citizens in the school planning process comfortable doing so because I used to be a when many of them are unsure of how to do school principal. Of course, there are other so, and possibly have no interest in participat- mothers who don’t have this level of comfort ing, and, second, how to convince principals and would not bring issues up because of and schools to accept a shared decision- fears or insecurities. This is common.” making process. As one teacher from the PTA Students are also expected to participate put it, “Our problem in Jordan is that parents in school decisions (photo 3.2). The intention used to be neglectful about their children’s in including students in the decision-making education. This is something we used to suf- process is evident, but in practice it some- fer from. . . . The problem was not with the times contradicts existing notions of the role teacher or student, but it was the environment and expectations of youth in the community. and the lack of cooperation between parents A recent assessment19 found that 44 percent and school.” With the move toward increased of teachers believed that school managers do school autonomy, parents and community not include students in decision making, members are expected to participate and resulting in the exclusion of students from make their voices heard in issues beyond their discussions and leaving principals and teach- children’s learning outcomes and behavior in ers misinformed about everyday problems in school—the traditional reasons for a school the school environment. PHOTO 3.2 Jordanian youth in an after-school setting at an elementary school Source: © Dana Similie / World Bank. Further permission required for reuse. 80 TRUST, VOICE, AND INCENTIVES As a result of implementation of the School head of the table, regardless of whether he and District Development Program and was invited to the meeting, indicating related measures, at the school level, princi- respect and admiration for a man who is pals have had to reconceptualize their role as not only an elder but also the elected chair- school leader from one that is solitary and man of the Education Council for the clus- autocratic20 to one that is inclusive, transpar- ter of schools. ent, and consensus-based, essentially adopt- Activating community engagement with ing a form of distributed leadership. “The schools requires a tactical ability to harness principal is constantly in contact with and create a sense of cohesion and purpose. the parents. He has all our numbers and he’s Parents have to think beyond their children, the center of contact as he is aware of all the and teachers and principals have to think school problems,” explained one parent. beyond their own schools, because under Hasan, a teacher who used to be a student at the SDDP, schools are placed into a cluster the same school, recounted, “In the past, the that in theory establishes a network of sup- way we used to engage with families was at port and collaboration if those relations the end of the school year, there was a school function as designed. A lthough many celebration that parents were invited to, in schools across Jordan are still struggling order to learn about the past school year and with regulating their cooperation under the student performance. Other than that there school clusters, Ira and Yarqa are happily was no communication with parents unless showcasing the benefits of their experi- there was a problem with their child.” These ences. The strategic maneuvering of the vil- comments illustrate the difficulty encoun- lages could be attributed to the training tered in enforcing new behavior and attitudes received through the project, as well as hav- toward engagement and community involve- ing identifi ed a strong Education Council ment in which principals are being asked to chairman, who in that capacity represents give up their absolute authority in all school- the local traditions and cultures of the com- related matters, but they are given relatively munity, while having a forward-looking few incentives to do so and have little under- vision of what the schools and citizens standing of the expected results. These diffi- together can accomplish as he was an edu- culties intensify in contexts that are otherwise cator himself for over 20 years. 21 His tolerant of autocratic leadership styles in knowledge of the education sector and the schools and suspicious of external challenges trust local members have in him have made to authority. it possible for him to mobilize and influence local representatives and leaders and con- vince them of the benefits of functioning Partnerships working toward change under a cluster of schools with increased Abu Muhanad evokes the classical image of parental and community engagement. “We a tribal leader. His traditional headdress, work as one team: the local community and the hatta , a white and black fabric, is the school,” said Abu Muhanad as he draped over his gray hair, and it frames his de s c r ib e d t he ne w at t it ude s towa rd face to reveal the deep lines etched by years increased school autonomy. of experience and wisdom. His knowledge However, realizing a common vision and of Jordan’s history and its education system commitment to the effective delivery of edu- is quite comprehensive, and he can recount cation entails in every country forging a sus- it with great accuracy and detail. No mat- tained complementarity between informal ter the occasion, he wears the traditional and formal systems. Communities every- thoub over which he adds a wool sweater where have a prevailing approach to educat- and a sports coat to protect against the ing and socializing their young members; it is winter chill. When he enters a room, all only relatively recently that this approach welcome him and make space for him at the was supplemented by a more structured and LO C A L S U C C E S S E S : S AT I S FA C T I O N , A C C O U N TA B I L I T Y, A N D Q U A L I T Y AT T H E LO C A L L E V E L 81 uniform institution called schooling that is prevailing informal system, Abu Muhanad is overseen (if not necessarily provided) by the also progressive in his views of and vision for state. 22 Whether and how the prevailing education. “You always have to keep reading informal system accommodates and supports and staying up to date with what is happen- the formal system play a large role in shaping ing in America, Canada, Russia, or Korea. the overall quality of education that children Otherwise you are left behind in this con- receive.23 stantly changing world,” explained Abu Muhanad. With a genuine commitment to education and improving student outcomes, Managing power relations he has, with the support of school principals, As implementation of the SDDP began, com- been able to build a process of democratic munity members organized themselves in a decision making that includes the voices of manner that institutionalized and legitimized teachers, parents, and, to a small extent, the informal instantiations of community students—a difficult accomplishment consid- and tribal leadership through school PTAs ering that it has required changes in the atti- and local Education Councils (LECs). “At the tude, behavior, and beliefs of community meeting yesterday, three community repre- members. “The decisions we make here are sentatives attended who were not invited,” done collaboratively,” said a teacher in refer- said a school principal about a PTA meeting. ence to the Education Council. “We used to “We cannot stop them from attending and run on a dictatorial system . . . [and] now sometimes their presence steers the conversa- that has changed. We have consensus and we tion away from the original topics of discus- work as one school cluster.” This reconceptu- sion.” In principle, educational reformers alization, largely a response to the SDDP, has advocating increased community engagement created a sense of internal accountability that at the school level seek school cooperation extends beyond that of school to parents; it is with a community council comprising a now one of parents and community to school. broad cross section of society. In Ira and Of course, some impediments to progress Yarqa, the reality is that the prevailing norms still exist. Describing the current impedi- of tribal authority largely determine who ments to community cohesion, one principal does and does not sit on such councils, observed, “Once we passed the first obstacle thereby ensuring that tribal and community of convincing citizens to engage with the control over how schools educate and ulti- schools we faced another challenge. There mately socialize their students is maintained. are some people who created obstacles on The villages of Ira and Yarqa, under Abu purpose, for personal reasons. . . . Essentially Muhanad’s leadership, were able to reconcile they don’t want to see a principal who has some of those effects by respecting and been leading successfully to progress. If a cer- accommodating the informal social struc- tain school principal is doing good work and tures throughout the process of change. To has a strong reputation, some people might that end, larger community buy-in to the worry that that school principal can be a vision of how their schools should be operat- competitor for a school council position that ing was achieved, and community gains they want.” became plausible. To address some of these challenges, edu- Indeed, the remaining challenge is manag- cators have devised innovative solutions to ing the negative aspects of these informal mitigate the undue influence of powerful mechanisms, which tend to appear in the community members. For example, Hatim, a form of capture by one or more groups who principal, created the Friends of the School share the same vested interests over the Council, an informal but diverse group of interests of the majority. Although he sym- parents to provide independent input into key bolizes the traditional culture and as an elder decisions.24 He took this step out of concern is likely an influential member in the that an elected body such as the PTA would 82 TRUST, VOICE, AND INCENTIVES mainly represent local elites who received These demonstrated efforts to improve the votes because of their influence and family quality, diversity, and frequency of interac- affiliation. Although the PTA is an important tion between the school and community body, it cannot address some of the tribal stand in stark contrast to the old ways of issues that arise. By creating a separate body communication and are now bearing fruit— with no formal leadership structure and with for example, attendance at previous PTA a representative from every family in the meetings had averaged 5 persons, but it was community, the principal was able to elicit now attracting more than 37. As one teacher the information he needed and secure more remarked, “The school used to be closed; equitable community buy-in to reform now it is open for the community. We as processes without letting the formal PTA teachers and a school can work with parents meetings become a venue at which select to address the needs of students. We have a community leaders felt challenged or only common understanding now between the advocated for their own interests. This type school and parents.” of maneuvering was possible because the Soliciting feedback from parents via a principal has been in this community for sev- questionnaire as part of the input to the SIP eral years, making him adept at managing was at first regarded by parents as a “test.”26 local politics. Establishment of the informal The result was low initial response rates, but Friends of the School Council thus achieved over time (and with further encouragement similar programmatic ends without it being from teachers, principals, and leaders such as perceived as a rival of the existing power bro- Abu Muhanad), parents realized that such kers. The school development plan was there- information provided a useful and rational fore created through a joint meeting with the basis on which difficult decisions could be PTA, students, community members, and the made. “Now the engagement with the com- Friends of the School Council. Nevertheless, munity is different: it’s regular, they are passive forms of resistance can still be exerted involved in decision making, they oversee on a daily basis against anyone perceived as and support the school,” observed Hasan, receiving too much attention. the teacher and former student at Zeid Bin Haritha School. “They provide support to the teachers and school by their active Working together and holding all participation.” accountable The increased interaction was then trans- Thus in the villages of Ira and Yarqa, based lated into higher parent expectations, leading on a belief in the benefits of community and them to seek stronger school initiatives, activ- school engagement, the daily process of edu- ities, and standards of teaching. These expec- cation was transformed into one that was tations were in turn reciprocated by the responsive, agile, and relevant to the local schools, which—having seen and, for some, context. Greater engagement among the having harnessed the positive outcomes of schools, community, Education Council, field community involvement—were happy to directorate, and Ministry of Education cooperate with the local community to resulted in stronger accountability among address challenges to service delivery. schools, parents, and students. Establishing “I stopped solving school problems on my PTAs, student councils, and a student sugges- own,” explained the principal of a girls’ tion box 25 allowed groups to identify their school. “I now wait for the LEC meeting, and interest in education, and, through participa- share my problems and there I fi nd multiple tion in the school self-review process, they solutions. Principals now offer support to realized their right to contribute to the school each other, and we come up with ideas that assessment, question school operations and I would have never thought of.” An example management, and not limit their interven- is the responses to the electrical issues facing tions to exam scores and behavioral matters. her school: “The school was built in two LO C A L S U C C E S S E S : S AT I S FA C T I O N , A C C O U N TA B I L I T Y, A N D Q U A L I T Y AT T H E LO C A L L E V E L 83 stages; when they built the second phase of Functioning as a cluster the school they didn’t install enough fuse One of the incentives for principals to relin- boxes and with the operation of the other quish some of their singular authority and part of the school our electricity would con- support this new process of community stantly be cut off. Computers in the lab were engagement was the recognition bestowed on burnt as a result of this. When I brought this them as leaders in the local Education up, I found lots of help. Community members Council and the creation of a dependable spoke to the electrical company; others came support system. The LEC facilitated and for- to the school to see what they could offer malized interaction between school princi- in terms of solutions until they installed pals that had not taken place previously (or new boxes.” had occurred much less frequently). It also With clear, positive outcomes from com- became a platform for cooperation and prob- munity engagement, friendly competition lem solving for the cluster, thereby alleviating and rivalry began among local schools, and the burden of a sole operator, the principal, even parents, to provide better student out- to address school challenges. As one principal comes and better school activities that would pointed out, “When the schools were placed boost pride and prestige for the community. in a cluster and we began to meet and follow It also helped parents and the community up on student scores, we realized there was a rally around common issues that their chil- joint problem with students’ writing skills in dren might be facing but that had not been Arabic in the early grades, second, third, and addressed by the education system such as fourth. The teachers decided to study and student learning difficulties and services for analyze student results, after which they put students with special needs. For example, the together a plan for development and met with local community together with the adminis- the cluster where it was agreed to implement tration organized doctor’s visits to schools to the plan across the schools. Now in the sec- identify students with vision impairments. ond year of implementation we see improved After the visit, parents pooled their funds to student learning.” As a cluster, the communi- print textbooks in larger font sizes for stu- ties and the schools they encompass have dents with severe visual issues. And the become a unit that has a vested interest in school worked with mothers of students with seeing that all schools succeed because one learning difficulties to organize a workshop school’s success can be shared with the neigh- on the special teaching practices and meth- boring school or community. ods needed for preparing diagnostic plans B eyond the immediate communit y- to assess the learning challenges such to-provider relations, midlevel accountability students face. within the cluster between the field director- This direct citizen-provider interaction, ate and the schools is largely mediated by the called the short route to accountability in LEC. At the directorate level, every cluster is the World Development Report 2004: represented by the head of its LEC at the Making Services Work for Poor People Educational Development Council. 27 This (World Bank 2004), is also a response to the representation enables the field directorate to realization of schools, parents, and commu- stay informed about all the local develop- nity members that their direct engagement ments and challenges that the schools are fac- with the Ministry of Education has largely ing within each cluster under the directorate. proven to be ineffective. School leaders and In turn, the directorate reviews the School community members in Ira and Yarqa are Improvement Plans and sets priorities and convinced that the Ministry of Education is plans based on the common needs, which in not capable of supporting the day-to-day turn are based on the actual realities faced by needs and unpredictable occurrences that all schools in the directorate. Thus the field deter the provision of quality services in directorate has to ensure that schools in Ira their villages. 84 TRUST, VOICE, AND INCENTIVES and Yarqa have updated SIPs and are com- In an education system that is segregated by municating their progress. gender after the fourth grade, the net result is In addition to the LEC acting as an inter- that boys often receive instruction from less mediary entity between the schools and field talented and less committed male teachers, directorate, school advisers (formally known creating a negatively reinforcing cycle in as inspectors) have been able to provide which the educational experience of boys schools with guidance and support on issues only further diminishes the likelihood that related to curriculum, instructional matters, the best of them will later seek a career as a and school resources. The school advisers teacher. Even if young men were to start their are, to the extent possible, members of the career as teachers, they probably would not local community or neighboring communi- remain in that profession should other ties, and so they are familiar with the context employment opportunities come along. and social structures in which these schools Moreover, because of the familiar disci- must operate. School advisers also have, at pline issues associated with managing all- some point in their careers, been principals male classrooms, the pedagogical styles in and have faced many of the same difficulties boys’ schools tend to be autocratic (even current principals have to address. In the quasi-militaristic), in stark contrast to the villages of Ira and Yarqa, the community, more interactive, engaging teaching styles which welcomes the advisers, has placed characterizing most girls’ schools, with great confidence in the advisers’ abilities to their more extensive exchange of knowl- solve problems related to student behavioral edge, more critical thinking, and more issues and to matters of teacher shortages. advanced teaching methods. Historically, The close cooperation among school, adviser, most trainee teachers in Jordan have been and community has resulted in some given no opportunity to gain practical expe- instances in advisers circumventing the long rience in classroom management, and once route through the Ministry of Education in placed in a school they receive little in the favor of finding local solutions to educational way of in-service training by the Ministry challenges. For example, the shortage of a of Education to upgrade and refine their biology teacher in one school was remedied skills. This is now steadily changing, but it by sharing a biology teacher from a neighbor- has yet to be incorporated into standard ing school. This solution was organized by professional practice. the adviser, who, with the schools, recog- nized that the process followed by the School and community problem solving Ministry of Education to appoint a teacher was too slow to solve the shortage. Parents, teachers, and students are aware of these challenges and make no attempts to hide or deny the effects they have on the The pursuit of change teaching and learning process. But the mem- Acquiring and retaining good teachers are a bers of the community and schools in Ira and central preoccupation of both parents and Yarqa together devised local solutions to principals in Ira and Yarqa, and across these problems. Problem solving became a Jordan. One key challenge arising from the community activity in which everyone was structure of the labor market is the availabil- given an opportunity to pitch in, make a sug- ity of high-quality male teachers. Whereas gestion, or lend his or her voice to the ongo- teaching is a relatively desirable family- ing conversation. Schools resorted internally friendly, high-status profession for women, to their teachers for support and externally to thereby attracting a steady stream of the parents, community members, and the school brightest graduates,28 teaching is a path taken cluster. Community and school problems by men only when other more prestigious, could range from how to utilize the school more lucrative options are not available. grounds to how to address a student’s LO C A L S U C C E S S E S : S AT I S FA C T I O N , A C C O U N TA B I L I T Y, A N D Q U A L I T Y AT T H E LO C A L L E V E L 85 learning disability or to identify the in-kind those opportunities are appreciated, further services a parent can provide. efforts toward comprehensive teacher devel- By leveraging social and community rela- opment are needed. Teachers and supervisors tions, the schools in Ira and Yarqa solved agree that even though they do get a degree local problems through transparent proce- of support from their field directorate, more dures that were accepted locally. Some of the can be done to improve teaching and learn- outcomes of the collective decision-making ing practices. To address deficiencies, teach- process included parents volunteering to read ers have been working together within the stories to students (a parent fluent in written school cluster to address their needs. “If a Arabic read to fi rst- and second-grade stu- new teacher is appointed, as teachers we give dents, another provided a similar service in our time and tell him or her to sit in on our English); pairing high-performing students classes so that they can learn,” explained one with low-performing ones; planting a citrus teacher. garden on the school grounds and giving the The shift to focusing additional resources harvested fruits to students as a snack; and on teachers continues to evolve. Teachers and printing textbooks in a larger font size for principals in Ira and Yarqa clearly under- students with visual disabilities. stand that quality education is established through quality teaching, but there is still a desire to demonstrate quality through mate- Empowering teachers to focus on rial resources at the school. Initially, coopera- student learning outcomes tion through community engagement and the “The role of the teacher has changed,” LEC were centered on bridging the shortages explained Um Sumaya, a teacher at a girls’ in materials by securing in-kind and financial school. The surrounding group of peers from donations for materials such as smart boards, her school nodded in agreement with her white boards, and computers. This is not statement. “In the beginning I used to just unlike many other communities and regions write on the board all class while students in Jordan and elsewhere that equate quality copied the work. Now I just lecture for five with physical and material resources—after minutes and let the students participate in all, without basic infrastructure, quality edu- practical learning activities.” The consensus cation cannot be delivered. However, among the teachers demonstrated the change research and practice have demonstrated that not only in the style of cooperation between the greatest determinant of improved educa- the school and community, but also in the tion services is the relationship between internal processes of a school from the devel- teacher and student (photo 3.3). Thus invest- opment of a SIP to the pedagogical styles ing in teacher training and quality is an used in the classroom. important step toward providing better edu- Um Sumaya and her peers readily admit- cation services. ted that although their teaching practices Because of the persistent constraints in were changing, there was still room for investing in teacher quality, it is not sur- improvement. In fact, they would eagerly prising that some parents hire tutors to accept any opportunity for further profes- work with their children af ter school sional development. As one teacher said, hours, and that teachers themselves often “When we started implementing the SDDP, become these tutors as a way of supple- our school created a plan based on outcomes menting their modest incomes. However, of the surveys of the teachers, parents, and this practice creates a negative cycle in students. From the survey we saw that which the basic incentives for teachers (and teacher training is a priority. So we sent a let- students) to do their job in class time are ter to the directorate of education informing reduced, with the children from families them of this need. And from there we least able to afford tutoring fees suffering received training opportunities.” Although the most. 29 86 TRUST, VOICE, AND INCENTIVES PHOTO 3.3 Young Jordanian practicing her arithmetic at an elementary school Source: © Dana Similie / World Bank. Further permission required for reuse. Responding to large- and medium-scale our school to help in that effort. The two interruptions in service delivery schools kept cooperating together through- out the process, along with the other schools Perhaps the most well-known example of in the cluster.” Because of this support, the intercom munit y cooperation was the damaged school was indeed able to open on response to a fire at the Yarqa Secondary time. School for Boys that had rendered the school Other (less dramatic) examples of commu- unsafe. Several classrooms were severely nity cooperation include parents offering to damaged, and the school’s archives were fi x a water tank, plant a garden, construct a destroyed.30 The school happened to be quite fence, and donate a stove and ladder, and the near Zeid Bin Haritha. Principal Hatim provision of workshops and lectures on civil described his school’s response: “In the sum- defense, technology, medicine, and teaching mer, the PTA held an urgent meeting, and we in collaboration among the local community, brainstormed on how to help in this situation schools, and local universities. “The Ministry so that the school could be operational by the of Education is too slow,” said a teacher. “We start of the academic year. We raised funds, know we can’t depend on them for all our and the PTA cooperated with families and problems. Now we know when we need the student council to restore the neighboring support we turn to the cluster and we ask school.” Cooperation even extended to shar- students’ parents and other community mem- ing teachers because the damaged school had bers for help. Together we are able to get the a shortage of teachers; temporarily accepting resources we need.” Perceptions such as these some students to lighten the burden; and and the poor or failed responses from the restoring the damaged archives. “The docu- Ministry of Education have pushed schools ments in the school were totally burned; we and community members closer together had to bring specialists in to recover some of because they recognize that efforts to involve those documents. We even sent teachers from the ministry in every interruption of LO C A L S U C C E S S E S : S AT I S FA C T I O N , A C C O U N TA B I L I T Y, A N D Q U A L I T Y AT T H E LO C A L L E V E L 87 education services are not effective, even such problems are regularly addressed. though at times it is necessary. The Ministry of Education’s statements that it had no financial resources to spare and ask- ing that students be transferred to the neigh- Enduring challenges boring schools suggest that much remains to These encouraging examples of community be done to enhance the resources of the min- cooperation demonstrate the ways in which istry as well as strengthen its own relation- collective action can be harnessed in pursuit ships with the communities it serves. of the common good. 31 As many observers Another challenge is to ensure that stake- have noted, however, close communities also holders are aware of their level of agency and have a downside in that they can exert undue their ability to engage with leadership teams. influence over decisions that, with broader The more educated and economically better input, might have yielded superior outcomes. positioned parents in Ira and Yarqa were giv- In addition, powerful community leaders can ing their feedback and engaging with the sometimes feel threatened by external initia- schools, but this relationship was not as tives that they perceive as a potential (or apparent for illiterate community members actual) challenge to their authority and or those with lower incomes. The general status. But even the most functional of com- consensus among the local Education munities can only do so much; without ade- Council and PTA was that these community quate public resources the quality of school members were represented by either their buildings and resources will inevitably children (students at the schools) or by decline. The schools we visited showed visible ack nowledg ment of t hei r ha rdsh ips. signs of aging, and classrooms were very However, this could not be determined. small for the number of students they were A fi nal enduring challenge, at least from expected to hold. School facilities were also an assessment point of view, is discerning lacking—for example, the bathrooms had to when and how the “words” of stakeholders be changed to accommodate students. The about local education reform (deftly using old teachers’ bathroom was destroyed, and a familiar development phrases) and their larger shared student-teacher bathroom was “actions” (incorporating new behaviors into created. However, the bathroom is far from everyday practice) actually begin to align. For the main school structure, making it hard to example, although the PTA recognizes that monitor what happens there. there should be greater involvement between It is unreasonable to expect positive com- the school and community and that it repre- munity relations, as important as they are, to sents a difference in the past relationship, the plug resource gaps of this magnitude. (In this tone and terms of discussion still seem to fall sense, all our case studies are instances of along the same lines: how to fi x low scores optimizing within otherwise suboptimal and address behavioral problems. PTA mem- structural circumstances.) A lack of proper bers can be very good about voicing their facilities such as a sports center can create concerns in “development-speak” and calling problems, especially for boys who feel for the school and community to “work hand constrained because they have no space to in hand.” In practice, however, when it comes express themselves and expend youthful to establishing the structured means for energy. Following the fire, the disconnect ongoing contact, there is clearly room for with the Ministry of Education in terms of improvement. Because the SDDP is still fi nancial support was partially fi lled by the something relatively new to the villages of Ira local community, who care deeply about and Yarqa, stakeholders refer to the changes their students attending the school—indeed, they have experienced as processes that hap- they fi xed the fi re-damaged school by them- pened within that project. Internalization of selves in two months. But such “social actions is still forming, presenting the risk solutions” cannot be the basis on which that should the project end, community 88 TRUST, VOICE, AND INCENTIVES engagement might end as well. And yet there accountability works. Moreover, the LEC is also reason to be optimistic that further was established as part of compliance with improvements can build on what has been program guidelines, and formal regulations attained thus far. for its mandate were only recently passed. Meanwhile, community engagement and accountability relationships may be improv- Reflections and implications ing in “successful” cases such as Ira and The villages of Ira and Yarqa have success- Yarqa, but other communities have not been fully used an external program to mobilize as responsive to the program. Fourteen out of social relations, which are a defining feature 42 field directorates do not have an Education of everyday life in Jordan. Communities are Council thus far, but implementation is structured around close family ties and slowly moving forward, and so there is much singular tribal affiliations (a key basis of iden- for everyone to learn from successful cases as tity, informal authority, and dispute resolu- the rollout of the program continues. tion), which means there was already a social platform on which to build. Discerning ways in which to harness the many positive aspects Case study 3: Health care of these relationships—and, where necessary, services in Jordan to curb their undue influence—became the The quality of health care in Jordan is notably central task of school principals seeking to uneven. On the one hand, Jordan has a repu- implement the SDDP in ways beneficial to tation in the MENA region for high-quality students, teachers, parents, and the wider specialty care, prompting some 200,000 community. National education reforms are patients to travel to Jordan from other coun- successful to the extent they are coherently tries each year seeking treatment. On the and legitimately mapped onto prevailing local other hand, anecdotes point to the lower norms and governance mechanisms and, quality of primary health care (PHC), leaving when necessary, incrementally curb the undue many Jordanians—particularly the poor, the influence of powerful local elites. Realizing vulnerable, and those living in rural areas— these types of reforms is a collective social without access to high-quality essential task that, like any such task anywhere, health services. The problem facing Jordan is requires persistence, learning, and a tolerance not a dearth of knowledge or limited public for occasional setbacks. Building the capabil- spending on health care, but rather ineffi- ity for such tasks and ensuring the legitimacy ciency coupled with inequity in financing and of the reform process itself entail ensuring the provision of services to meet citizens’ that the details informing such strategies pri- needs. This said, the availability of quality marily emanate from citizens themselves (as services depends not only on providing mate- opposed to external development “experts”’). rial and human resources, but also on estab- Although commitment to the LEC seems lishing good governance procedures and high in Ira and Yarqa, as well as awareness practices to use these resources effectively. and appreciation of the possibilities for col- Thus governments, development partners, laboration it offers, there is some room for and scholars have increasingly emphasized improvement in formalizing avenues such as the role that voice and participation, trans- PTAs and translating everyone’s willingness parency, accountability, monitoring and to work hand in hand into actual actions that enforcement, and rules and regulations play guarantee participatory approaches in reach- in determining good governance with down- ing decisions. A more structured system of stream impacts on the quality of service offering incentives and support (professional delivery. Collectively, these facets of gover- development, rewards), whether for teachers, nance interact to shape how efficient, effec- students, or principals, is needed to enhance tive, and responsive the health system is to the learning environment and ensure that citizens’ needs (photo 3.4). LO C A L S U C C E S S E S : S AT I S FA C T I O N , A C C O U N TA B I L I T Y, A N D Q U A L I T Y AT T H E LO C A L L E V E L 89 PHOTO 3.4 Sign greeting patients at a health committees. Questions focused on the quality center in Jordan: “Welcome to our center. We of services delivered, management and promise to present you with the highest quality of administrative systems, and governance pro- health services.” cedures and relations. Visual inspection of the facilities and administrative records was also conducted to assess the condition of the existing infrastructure and equipment, the availability and control of pharmaceutical stocks, and the management of patient health records and other health and administrative information. The results of this exploratory study are not readily generalizable to all PHCCs and CHCCs in Jordan, but they point to struc- tural constraints that limit the ability of health facilities to provide equitable access to high-quality essential services. At the central level, the Ministry of Health plays an impor- tant and dominant role in supporting PHCCs and CHCCs, but the current system lacks incentives to promote sufficient staffing, par- ticularly of female doctors and staff in less desirable posts such as in rural and remote areas. In addition, although the system of using PHCCs as an entry point for doctors trained abroad helps provide PHCCs with a Source: © MENA study team / World Bank. Further permission required for steady stream of newly minted doctors, most reuse. leave the clinics after a short period to return to specialized medical training. Indeed, the Six clinics in four governorates majority of doctors employed in the clinics To identify governance factors that may visited had taken their degrees outside of affect the quality of service delivery in public Jordan and had been employed in their clinic primary health care clinics (PHCCs) and for a very short period. Resource limitations comprehensive health care clinics (CHCCs) thus create a system dominated by less- in Jordan, we visited six clinics in 4 of experienced staff and foster a high turnover, Jordan’s 12 governorates (‘Amman, Al Balqa’, which in turn hinders communication and Jarash, and ‘Ajlun). The clinics were selected the formation of trust between communities in collaboration with the Ministry of Health and health clinic staff, as well as among staff and from referrals by the Health Care within facilities and with the Ministry Accreditation Council (HCAC) based on the of Health. clinics’ notable improvements in the quality Other important structural constraints of their services over the preceding three to that result in inefficiencies and preclude more five years and based on their positive perfor- proactive clinic management and effective mances compared with clinics in the sur- service provision are mainly related to the lack rounding areas and in regions with similar of transparency and systematic monitoring. socioeconomic profiles. Both Ministry of Health and clinic staff During the site visits, open-ended qualita- suggested a lack of transparency in staff tive interviews were conducted with chief placement, promotion, and training opportu- medical officers, medical and administrative nities and in budget processes. The lack of staff, patients, and members of local health transparency prompts communities to make 90 TRUST, VOICE, AND INCENTIVES extensive requests, and it undermines facili- Formalized decision-making and manage- ties’ resourcefulness and their abilities to set ment procedures foster better health care. and achieve long-term goals, ultimately forgo- The accreditation process requires establish- ing midrange changes that could improve ing and adhering to clear administrative and facilities, programs, and staff quality. This clinical procedures, including patient file sys- situation hinders planning, and particularly tems, personnel fi les, committees for clinic the ability of facilities to make informed management, and guidelines for access to choices that increase the quality and efficiency pharmacy stocks. Such procedures set clear of health care services. Furthermore, inade- expectations and facilitate information shar- quate resources for effective monitoring ing and problem solving between patients appear to compound the problem because and staff and among staff. Moreover, as a facility-level issues often go undetected. This, staff member in one of the PHCCs visited (Al coupled with the fact that facilities have little Mastaba) explained, such guidelines, along or no control over their own budget and its with engagement in the accreditation process, execution, limits the ability of clinics to help to establish a “culture of quality” and improve overall service quality based on per- encourage staff to do their best. ceived community needs. Efforts to improve transparency were Despite these constraints, the site visits clearly visible in many of the clinics visited. uncovered drivers that appear to promote Prices and patients’ rights were visibly posted good service delivery. These drivers are in every clinic, and in some, such as the related to facility management and quality Sakhra CHCC, the results of patient surveys assurance. They are also related to the rela- as well as responses to patient suggestions tionships between facilities and their commu- and complaints were prominently displayed nities, as well as the social institutions and on a bulletin board. Such transparency cre- ties that shape positive engagement between ates and reinforces a culture of respectful the community and providers, among facility patient-centered care. Practices instigated by staff, and between the facility and commu- the accreditation process also created and nity and the broader regional and national formalized opportunities for patients’ and levels. These drivers are discussed in the sec- staff members’ voices to be heard. Posting of tions that follow. hotline numbers for patient complaints, the provision of suggestion boxes (photo 3.5), and periodic patient surveys were practices Facility management and quality assurance PHOTO 3.5 Suggestion and complaint box at a Effective administrative procedures can health care center in Jordan improve the quality of service delivery through more transparency and greater voice and par- ticipation. Based on the findings from this case study, an accreditation process provides one mechanism for improving facility admin- istration and the attendant quality of health care. The Jordanian accreditation process, which is administered by the HCAC, 32 sets rigorous standards to be met every three years. The procedures and requirements help establish clear rules and regulations, increase transparency for clients and staff, develop more effective staff monitoring, and give the staff in facilities as well as communities a Source: © MENA study team / World Bank. Further permission required for greater voice and higher level of participation. reuse. LO C A L S U C C E S S E S : S AT I S FA C T I O N , A C C O U N TA B I L I T Y, A N D Q U A L I T Y AT T H E LO C A L L E V E L 91 instigated by the accreditation process. multiply the positive outcomes of these Accreditation standards also support voice reforms. The chief medical officer (CMO) of and participation among staff members the Sakhra CHCC recognized the value of through regularly held staff meetings and procedures and extended them beyond the committee participation to address issues minimum requirements. For example, as part such as infection control, quality improve- of personnel management, the CMO estab- ment, patient safety, and security. lished a training manual for new staff Local health committees provide a venue members, reviewed it with them, and required for aggregating and expressing community them to sign a paper agreeing to its conditions. needs and preferences and mobilizing In addition, the CMO searched for and dis- resources to address them. The terms of refer- tributed standard procedures on diseases ence established by the Ministry of Health for beyond Jordanian guidelines and worked local health committees, combined with the with the health committee to establish a role of the coordinator, provide an organized Facebook page where clinic activities and dis- approach to assessing and communicating cussion of clinic performance would be community needs. An annual survey of com- posted online. However, administrative mittee members determines priorities for reforms are less effective under weak leaders. action, and motivated committee members, For example, a CMO in one clinic noted that chosen from key community leaders, leverage staff meetings were not held because people their talents and networks to mobilize worked on different days and the assumption resources to address community health con- was that “everyone knows their job.” cerns. The health committees act primarily as a “board of directors,” helping to identify key Social ties issues and mobilize their resources and networks to enhance outreach and implemen- Social networks play an important role in tation. For example, a community health facilitating voice and participation, and screening organized by the Zay CHCC health potentially in improving the extent and qual- committee uncovered previously unrecog- ity of services. Personal ties, whether among nized high levels of diabetes (32 percent) and family or friends and neighbors, are particu- hypertension (25 percent), thereby facilitating larly valuable for establishing priorities, patients’ turn to the clinic for chronic care. extending public health outreach in the com- Although the nature of blood ties and munity, and mobilizing resources to support other informal networks were unchanged clinic activities and development. Interviews before and after the establishment of the with the staff and members of local commit- committees, interviewees in clinics and on tees at the health care facilities visited indi- the health committees agreed that the for- cated that shared identity, especially a malization of the process generated through common tribal affi liation, was particularly the accreditation process made the health valuable in rural areas but somewhat less rel- committees more effective. Factors contribut- evant in urban centers, where people from ing to this included members’ pride in being diverse regions and backgrounds intermingle given a formally recognized role, enhanced and many residents do not come from the financial and administrative support, require- major Trans-Jordanian tribal families. ments regarding regular meetings and estab- Indeed, when members of the local health lished procedures (such as voting for priority committees in rural facilities and local resi- public health issues that the health council dents hail from the same tribes, the staff and would address in the community), and governing board of a facility have a built-in requirements for broad participation. channel through which to reach the commu- Leadership enhances the impact of admin- nity with vaccination and other health cam- istrative reforms associated with the accredi- paigns and to encourage greater compliance tation process. Strong, proactive leaders can with medical instructions. For example, 92 TRUST, VOICE, AND INCENTIVES in the Zay CHCC, located in a village near were reluctant to undergo exams. Ties with Al Salt, members of the health committee felt other local actors such as school officials also that their personal and professional connec- facilitate the work of health centers. At the tions helped to mobilize the community to Alyazedyah Health Center in Al Balqa’ gov- participate in health care campaigns such as ernorate, the school principal offered to ask screening for noncommunicable diseases and students to help the staff with refurbishment family planning initiatives. of the center. The head of the local health Among the key factors driving the per- center gratefully accepted, and a group of ceived increased participation in these events students painted and cleaned the facility. were shared blood ties; committee members Strong social ties may facilitate health care claimed that their center’s health initiatives improvements in some instances, but the were relatively successful because the mem- tight-knit nature of the community can also bers came from the same families as the complicate the work of the health center. For targeted beneficiaries. In their estimation, this example, it may make sensitive issues such as enhanced their positive contributions to the domestic violence more difficult to address. work of the local health committee. Similarly, Indeed, one clinician explained that domestic at the Sakhra CHCC in ‘Ajlun governorate, violence was not “seen” in the clinic, noting staff members noted that local tribal relations that because everyone in the village is related, facilitated their work in the community. no one would talk and the providers thus do The Al-Moumani family, a prominent Trans- not ask. Where strong social ties are driven Jordanian family, is the dominant group in by blood ties and where they promote inter- the village, cementing its presence in the com- marriage and consanguinity, they can also munity through high rates of intermarriage. increase health risks. Of the 24 members of the center’s local health committee, 17 hailed from this tribe. These Conclusions family ties enable the staff and committee members to reach local residents relatively The Jordanian health system stands out easily and to encourage them to visit the cen- within the MENA region for its relatively ter and to comply with medical advice. comprehensive coverage of the population in Beyond family ties, social networks and basic health services. Furthermore, some of linkages to important local social institutions the staff members at the facilities visited had help health workers to accomplish their distinguished themselves by their dedication tasks. Mosque and religious leaders play an and commitment to their work, which clearly important role in both urban and rural areas, has tangible effects on the quality of care particularly for issues that can be understood delivered. The comparative success of the within religious teachings such as contracep- health system in Jordan appears to continue tives, vaccinations, drugs, and smoking. even now in the face of the serious strains Directors of youth centers, school principals imposed on the system by the Syrian refugee and teachers, and other established commu- crisis. nity members were cited as lending similar However, the challenges to sustaining the weight to initiatives. Staff members empha- gains are numerous. For one thing, although sized that the support of the local religious administrative reforms and the accreditation authorities was vital for their health process can generate multidimensional campaigns. For example, the cooperation of improvements in the quality of care, only a the local mosque and the inclusion of the small percentage of health centers in the pub- imam on the health committee helped to lic health system have been accredited, and boost the success of family planning initia- many are not even eligible because of deficien- tives among local residents in Sakhra. Before cies in their infrastructure. Furthermore, the the establishment of this working relation- future of the HCAC is uncertain because the ship with local religious institutions, women U.S. Agency for International Development LO C A L S U C C E S S E S : S AT I S FA C T I O N , A C C O U N TA B I L I T Y, A N D Q U A L I T Y AT T H E LO C A L L E V E L 93 (USAID) program that funded its activities as diabetes and hypertension, which have and fully subsidized the accreditation process high prevalence rates in the area and through- for participating public facilities ended in out the country. Thus an especially high vol- 2013. Unless the facilities seeking accredita- ume of patients were waiting to be seen by tion are willing and able to share the financial the one of the center’s two full-time doctors. burden of the process, sustainability of the On average, about 90 patients visit the center HCAC is uncertain. Moreover, the adminis- each day, almost all of whom come in the trative reforms at the facility level can result morning. The center’s catchment area is in significant improvements, but until home to more than 18,500 people. The cen- upstream problems are solved, thereby reduc- ter has an attractive garden, which was cre- ing the high turnover, inexperienced staff, ated and is tended to by the center’s number of weak budgets, and other problems, employees, and a ramp from the parking lot the improvements in health care will be to the entrance of the center provides handi- limited. Similarly, although social institutions capped patients with access. This health cen- can facilitate better health outcomes, these ter is one of the more than 2,600 public are facilitators of, not replacements for, effec- primary health care centers that mainly serve tive governance. the urban and rural poor in Morocco.33 There is much to be learned in order to Over the last decade, the Moroccan gov- develop governance mechanisms that facili- ernment has made a concerted effort to tate quality service delivery. The evidence improve public health as part of a larger linking governance to service delivery quality strategy to boost human development and is still sparse and focused on clinic operations reduce poverty. Despite decreases in mortal- and not on health outcomes. It is essential to ity rates during earlier decades, Morocco’s identify the key governance levers that will maternal, child, and infant mortality rates— improve quality throughout the system, espe- especially in rural areas—are among the cially among facilities that are unaccredited highest in the MENA region. According to and unlikely to undergo accreditation in the the national statistics, in 2011 about near future, which currently is the vast 10 percent of Morocco’s 13.4 million rural majority of facilities. It is also essential to residents and 3.5 percent of its urban resi- explore the linkages between blood ties (espe- dents fell below the national poverty rate. cially tribal diversity), social institutions Meanwhile, several important institutional more broadly, and the delivery of social ser- reforms were adopted during the 2000s. In vices. These need to be more systematically particular, the government focused on examined through rigorous research based expanding health care coverage through the on comparisons of subnational units. To the development of medical insurance schemes. extent that political and social institutions In 2002 the government enacted a law that are associated with improvements in services precipitated the creation of two insurance delivery, it is important to explore why and programs: a compulsory public health insur- how they are linked in order to identify areas ance system (Assurance Maladie Obligatoire i n w h i c h r e for m i s m or e l i kely to or AMO), which covers public sector and for- be effective. mal private sector employees, and a medical assistance scheme targeting the poor (Regime d’Assistance Maladie aux Economiquement Case study 4: Health care Démunis or RAMED), which was extended services in Morocco across the country in 2012. Located in one of the poorest provinces in The government bodies closest to primary Morocco, the public health center we visited health centers, the provincial delegations, on a hot morning in May was packed with manage the budgets of the primary care patients. The day of our visit was reserved for facilities, which have on average about 15,000 patients suffering from chronic diseases such people in their catchment areas. Each of the 94 TRUST, VOICE, AND INCENTIVES 83 delegations has a head doctor, who is in In return for enrolling in the CQ, employees charge of ambulatory care at the provincial have greater access to supplemental training level, or a medecin chef du SIAAP (Service programs, and centers may become eligible to d’Infrastructures d’Action Ambulatoires receive new equipment or even to receive Provincial), who oversees all of the primary funds for renovation. The CQ and other care facilities in a province. In 2011, as part reforms in the health system have made of the national administrative decentraliza- important advances in improving access to tion process, 16 regional directorates were and the quality of primary health care in established. These directorates work closely Morocco. with the provincial delegations to manage The track record of the CQ and a spate of local health priorities and issues in their research in health policy and management respective regions. suggest that governance at multiple levels—at In 2007 the Ministry of Health launched the facility itself as well as at all administra- the first round of the Concours Qualité (CQ) tive levels—has an important effect on the to improve the quality care at public hospitals quality of health care. To explore these link- and primary care facilities. 34 The CQ con- ages, we visited six different public health sists of competitions between health facilities centers outside of major urban areas of to achieve the highest level of care within Morocco. About half of the selected facilities their respective provinces. The process entails had participated in the CQ at least once. a self-assessment in which staff at participat- Because of the limited size of the sample and ing centers complete standardized question- the fact that it does not represent the full naires that measure a range of quality range of variation in health centers in the indicators, as well as an audit by peers. Based country, it is impossible to make generaliza- on the logic that competition and recognition tions based on our findings. Nevertheless, of good work motivate people to seek our visits yielded some compelling observa- improvement, the CQ is designed to encour- tions that deserve further investigation using age the participating facilities to upgrade more systematic research methods. care. The heads of provincial delegations Our visits to urban and semiurban pri- select centers to participate based on the mary care centers revealed substantial varia- motivation of the team, their openness to tion in the quality of care, even in this limited change and willingness to adopt new proce- sample. A consistent array of factors were dures, and their prospects for winning. observed at facilities that performed well, including larger patient loads, reasonable wait times despite high demand, good man- PHOTO 3.6 A well-managed and well-stocked pharmacy in one of agement and good availability of stocks, con- Morocco’s primary health care centers sumables, and equipment (photo 3.6), and detailed and regular maintenance of patient medical records. For example, at a center in a small provincial town we noted that the vac- cines and other medications were stored properly and carefully, the pharmacy was well managed and stocked, and all personnel were following the proper protocols for dis- posal of medical waste. Medical staff mem- bers kept consistent medical records and followed clear procedures for storing and restocking medications and consumables and maintaining careful records on the upkeep, repair, and replacement of equipment. Staff Source: © Melani Cammett / World Bank. Further permission required for reuse. members claimed that the adoption of these LO C A L S U C C E S S E S : S AT I S FA C T I O N , A C C O U N TA B I L I T Y, A N D Q U A L I T Y AT T H E LO C A L L E V E L 95 procedures was leading to shorter wait times This is especially important in the context of and improving the monitoring of patients’ tight budget constraints and the consequent health status. The center’s management had chronic understaffi ng of facilities. The head also devised clever ways to ensure that doctors at the high-performing centers we patients formed lines when waiting for ser- visited spoke with evident pride of their vices to avoid crowding. For example, accomplishments and placed great value on through a small window near the exit of the their roles as stewards of the quality improve- facility, the pharmacy staff dispensed the ment process. medications prescribed to patients. To access Ultimately, a successful facility requires the window, patients had to pass through a far more than strong leadership, particularly corridor created by a half-wall designed to to maintain its good practices over time. The accommodate one wheelchair and no more innovations and procedures introduced by than one person at a time. The center is in the visionary and energetic leaders must be process of constructing a similar structure so institutionalized and routinized so they can that patients line up single file outside the outlive the tenure of a single person. facility while waiting to enter during the Furthermore, in rural areas it is all the more morning rush. At a center in another provin- challenging to attract competent physicians cial city, a nurse stationed at the entrance with good leadership skills. In a facility in a greets patients and issues them colored tick- semirural area, the head doctor was often not ets that correspond with color-coded signs present at the facility, in part because chronic indicating the different departments in the understaffi ng obliged him to rotate among facility. The system ensures that illiterate several different centers in the province. patients know where to go even if they can- not read the signs posted in the center. Team spirit and shared mission Interviews with doctors and other medical staff as well as observations during the site A sense of a shared mission and a collabora- visits suggested that at least five factors are tive ethic help to motivate staff and ensure associated with improvements in the quality that they all know and fulfill their responsi- of health care at Moroccan PHCCs: bilities. These principles may even encourage (1) leadership; (2) team spirit and a shared personnel to propose ways to improve the mission; (3) participation in the Concours delivery of services. Like leadership, team Qualité; (4) effective coordination with local spirit and a relatively flat organizational cul- and regional officials from the Ministry of ture are especially important in the context Health; and (5) partnerships and community of resource scarcity in which staff members relations. These factors are described in the are required to make do with less. As the del- sections that follow. egate from the Ministry of Health in one province noted, “It is vital to have a team ethic. If a center has problems in interper- Leadership sonal relations, then it does not work. . . . The presence of dynamic, energetic, and Motivation among the personnel is vital, visionary leadership at multiple levels is a sine especially given the lack of human resources qua non for the establishment of well- in the health sector.” Doctors at high- managed, high-performing PHCCs. Effective performing centers have internalized the leaders can motivate staff members to carry importance of team spirit. The head doctor out their duties competently and thoroughly, of a busy health center in a provincial city introduce new procedures and management emphasized that a strong ethic of teamwork systems, institute a “culture of quality” among and will (volonté) is shared by all the staff. staff members, inspire confidence in the com- To promote a team-based approach and to munity and local government officials, and facilitate the effective functioning of the attract additional resources for the facility. center, the head doctors of health centers 96 TRUST, VOICE, AND INCENTIVES convene monthly staff meetings. At these and When asked what changed in her health other supplemental meetings, the employees center after participation in the CQ, the head discuss the problems they encounter while doctor of a center in a semiurban area carrying out their jobs and report on their claimed that wait times plunged, regular staff activities. Some facilities also hold monthly meetings were institutionalized, more man- training workshops for the staff. For exam- agement and administrative protocols were ple, in one center a nurse showed us a docu- introduced, posters informing patients of the ment describing a recent training workshop center’s policies and providing important on quality improvement that was conducted health information were displayed around by the head doctor of the provincial SIAAP. the facility, and sanitary procedures were An earlier training workshop in this center greatly enhanced, potentially reducing infec- focused on increasing and reinforcing trans- tion rates. The head doctor at another semi- parency and good record keeping (tracabilité) urban center noted that participation in the in the facility. In a semirural facility, the head CQ greatly improved relations among staff doctor discussed the lack of sense of team- members, who now share team spirit. work in the center and the importance of The competition inherent in the CQ process overcoming obstacles to better interpersonal also encouraged her staff members to adopt relations among the staff members. and adhere to many administrative changes. In particular, she noted that the stocking and storage procedures for medications improved Participation in the Concours Qualité substantially after participating in the CQ. Meetings with staff members from facilities At another health center in a provincial town, that had and had not competed in the national the head doctor said that since her center CQ revealed that the program clearly has a enrolled in the CQ it has been far more struc- positive effect on the management and tured and organized and staff members feel administration of the participating PHCCs. more valued. She noted that staff members The mere act of enrolling in the program gen- already possessed dynamism and will, but erates a significant transfer of knowledge and they subsequently learned about specific pro- procedures to the managers of health centers. cedures and systems that would improve the The head doctors and nurses of all participat- operation of the center. “We wanted to work ing PHCCs unanimously agreed that system- but didn’t know how,” said the head doctor. atic procedures for maintaining medical These observations from centers that par- records, monitoring stocks of medications, ticipated in the CQ contrast sharply with consumables, and equipment, and tracking those by staff at facilities that did not enroll vaccinations and other health campaigns in in the program. For example, in a semirural the community were developed in preparation facility, staff reported that basic sanitary for the CQ competition and maintained practices were deficient, and the examination thereafter. As a result, participating centers rooms did not have separate medical waste instituted better systems of record keeping, containers. At another nonparticipating facil- managing stocks and equipment, and dis- ity in a small provincial town, the medical posal of medical waste. Staff members also staff could not report on the system for man- agreed that these changes resulted in better aging drug stocks, and they were bewildered tracking of patient health status and shorter when asked to report on the basic manage- wait times. The senior staff members of facili- ment and administrative procedures used in ties participating in the program now sub- the facility. scribe to a set of management principles based The clearly positive impact of the CQ is on teamwork, openness to innovation, and testament to the value of the program. The ongoing quality improvement, and are more greatest challenge for the program, however, open to pursuing partnerships to improve is its sustainability. When asked whether she access to medical care in their communities. intends to enroll the facility in the upcoming LO C A L S U C C E S S E S : S AT I S FA C T I O N , A C C O U N TA B I L I T Y, A N D Q U A L I T Y AT T H E LO C A L L E V E L 97 competition, a doctor at a high-performing doctors to be entrepreneurial in establishing center that had won a prize in a recent round local partnerships. of the program replied that she would not participate in future rounds because she Partnerships and community relations could not commit the time to prepare for the competition while meeting the needs of a Partnerships with nearby groups, organiza- heavy patient load. In fact, she highlighted an tions, and prominent individuals help PHCCs irony of participation in the CQ: participat- to meet the needs of their surrounding com- ing centers attract more and more patients, munities more effectively. Staff members even from beyond their catchment areas, as described initiatives with local mosques, the news of their improved quality of care women with religious educations who serve spreads. As a result, high-performing centers as religious guides (murshidat), the women may face heavier strains on their material and who work in the local hammams (public human capital than other centers or heavier baths), and the local civil society organiza- strains than they experienced prior to tions working on the environment, women’s enrollment in the program. Furthermore, a i s s u e s , yout h , a n d i n o t h e r a r e a s . lack of personnel may hamper the ability of Collaborative projects focus on a range of some facilities to implement or maintain the topics such as family planning, HIV/AIDS systems they instituted after participation testing, breast cancer, and chronic diseases. in the CQ, particularly in understaffed Partnerships with these local actors help rural centers. PHCCs to target populations with health awareness campaigns, motivate patients to follow medical protocols, and encourage Effective coordination with local and openness in the community to health- regional officials from the ministry of promoting practices that might otherwise be health viewed as violations of culturally conserva- For PHCCs, effective coordination between tive norms. For example, with the support of the head doctor and local health officials is the murshidat, women in some communities critical to meeting the needs of the popula- began to use intrauterine devices at a higher tions in their catchment areas. Regular rate and agreed to undergo HI V/AIDS exchanges between the administrators of testing. facilities and officials from the provincial del- Formal and informal partnerships with egation helps to ensure that stock-outs of local groups are all the more important in the medications and equipment do not occur, context of low resources. Health officials in that facilities receive resources when avail- one province emphasized that the lack of able, and that local solutions are developed financial resources leads to chronic under- for local problems. In addition, open chan- staffi ng, in part because doctors who retire nels of communication aid in the dissemina- are not replaced. In this context, the delega- tion of good administrative and management tion of the Ministry of Health in one prov- practices, usually emanating from the local ince we visited is now trying to promote delegations and regional directorates down partnerships. For example, the head doctor at to the administrators of PHCCs. For exam- one facility negotiated lower prices for ple, a provincial health official noted, “When women at various private providers in the provincial delegates are close to the staff and area. Zakat funds 35 were mobilized to work closely with them, they are more moti- support the medical needs of these women, vated.” Furthermore, the outlooks of local offer home visits by doctors, and provide officials affect the possibilities for innovation. clothing for local children. The local delega- In some provincial delegations, officials tion also approached local private institu- actively encourage partnerships with other tions, asking them to sponsor material aid local actors, granting more freedom to programs within hospitals in the province. 98 TRUST, VOICE, AND INCENTIVES In one center in a provincial town, the (see chapter 7), and a key basis for searching resourcefulness of the staff coupled with for locally legitimate solutions to these chal- strong community engagement led to tangi- lenges is to map, explore, and explain this ble improvements in the facility. The center’s variation, to share the strategies and (espe- benches, outdoor fencing, garden, and inte- cially) the ideas driving positive outcomes, rior painting and décor were all obtained or and to encourage those elsewhere to do the carried out with the help of local formal and same. informal groups and community members. Such an approach alone, however, cannot Local community-based groups also helped bring about broader systemic reform, and it by advertising the center’s vaccination cannot compensate for the absence of the campaigns. basic levels of financial, human, and material resources needed to make even a minimally Case studies to share ideas and effective system function. It must be part of, inspire change not a substitute for, institutional change. But this will take a long time, and waiting for Many interesting and important findings such change to happen means consigning the emerge from these four case studies. At one present generation to a bleak future. Change level, these findings may not seem surprising. has to begin somewhere, and one place to After all, good leadership, effective manage- start is to recognize and reward those fea- ment systems, close engagement with the tured in the cases described here—that is, local community, mutual accountability those who are already making it happen. mechanisms, high expectations, aligned If there are transferable lessons, they lie in incentives, and strong support from middle reiterating the important aspirations (no mat- tiers of the civil service (that is, not just “pol- ter how attained) of structuring meaningful icy makers” at the top and frontline staff at incentive schemes, diligently managing the bottom) all combine to enable teams to human resources, and fostering both moti- provide services that meet high professional vated leadership and genuinely engaged staff. standards and satisfy citizens’ demands, In order to realize these goals, however, often in the face of very difficult circum- domestic policy analysts can conduct ana- stances. The cases examined are veritable lytic case studies along the lines provided islands of success in seas that all too often here, identifying local instances of where, cannot or will not deliver. As we discuss in how, and by whom effective responses to more detail in later chapters, these cases these organizational bottlenecks in everyday reveal how cycles of performance that posi- services provision have been found and tively reinforce one another have been created implemented. and sustained at the local level. The primary point of presenting such suc- cess stories, however, is not to elicit an item- Notes ized list of factors or practices that should be 1. The concept of (and methodology underpin- adopted elsewhere. It may be that some can ning) positive deviance as a learning tool for indeed be replicated elsewhere, but at the frontline service delivery reform was formu- level at which actual solutions need to be dis- lated by Pascale, Sternin, and Sternin (2010). cerned, the combination of factors required is Methodologists might instinctively be con- cerned that such an approach commits likely to be highly idiosyncratic—lots of dev- the error of “selecting on the dependent ils lurk in lots of details. Above all, these variable”— that is, of failing to recognize that cases demonstrate that in spite of deep and the same factors seemingly driving success pervasive challenges confronting service may elsewhere have no (or even the opposite) delivery providers in the MENA region, effect. Ordinarily, this is a legitimate concern, someone somewhere has figured out how to but the positive deviance approach (1) pre- make things work. There is always variation sumes some prior professional knowledge of LO C A L S U C C E S S E S : S AT I S FA C T I O N , A C C O U N TA B I L I T Y, A N D Q U A L I T Y AT T H E LO C A L L E V E L 99 the extent of the variation characterizing the held with the director of Jenin’s education outcome variable of interest (such as quality of directorate, math and science supervisors, the education) and the main (or expected) factors principal of Kufor Quod Girls’ Secondary responsible for it (such as wealth, teacher School, teachers, students, parents, and the quality); (2) seeks to explore the forms and local community. The case study was inspired sources of variation within settings that are in part by an earlier study carried out by the geographically proximate and demographi- Assessment and Evaluation Department cally similar, thus “controlling” for broad fac- (AED) of the Ministry of Education and tors that might otherwise be responsible for Higher Education (MOEHE), which com- observed variation such as proximity to roads pared high-performing schools with low- and public transport; and (3) makes only mild performing schools to uncover differences in causal claims regarding the formal identifica- school- and classroom-level factors influenc- tion of factors shaping local outcome varia- ing student learning. tion and instead strives to inspire those 6. The city of Jenin is an agricultural center for working elsewhere by showcasing specific the surrounding towns. Because the city has instances of how local innovators have crafted no major industry, its citizens have shifted effective solutions to difficult problems. The their attention to education as their only classic case is nutrition among the poor in the option to elevate their financial standing. slums of Vietnam, where a positive deviance This fact is indeed mirrored in the city’s approach discovered wide variation in stunt- yearly performance in the state exams, ing among infants, but learned that some chil- Tawjihi. In 2013, 7 high schools from Jenin dren living in exactly the same desperate alone were among the 10 highest-performing circumstances were nonetheless doing rela- schools across the West Bank and Gaza. tively well because their mothers defied the 7. All schools in Jenin underwent this rigorous prevailing community norms, opting to feed preparation, but Ms. Abla was particularly their infants small amounts several times a day able to capitalize on it and use it to her advan- (rather than larger amounts twice a day) and tage, as we explain shortly. to supplement their diets where possible with 8. For example, students selected the dates for nutrients. These simple but nonobvious differ- their exams and even the colors of the cur- ences were social innovations by the mothers tains in the classroom. themselves, and thus enjoyed considerable 9. Supervisors were previously known as local legitimacy when shared with others. “inspectors,” but the title was recently abol- Importantly, such discoveries are comple- ished in the spirit of this new concept. ments to, not substitutes for, broader changes 10. This initiative is of particular importance in health policy (and expansions in health because girls tend to choose the literary policy infrastructure and resources) that might stream in order to later enter careers that con- be required to bring about society-wide form to the traditional definition of what improvements in child nutrition. constitutes an acceptable profession for a 2. Claussen, Kiernan, and Gramshaug (2013) girl. also report that education accounted for 11. On the broader importance of and challenges 15.7 percent of the total public expenditure associated with forging such education sys- in 2012, at approximately US$680 per stu- tems in the 21st century, see Pritchett (2013). dent per year. All dollar amounts are U.S. 12. The students’ stated career aspirations dollars unless otherwise indicated. included dentist, lawyer, aeronautical engi- 3. As we note shortly, however, strikes can neer, and teacher. cause teachers to miss many days of instruc- 13. Interview with the chairman of the local tion over the course of a school year. Education Council, March 20, 2014. Even 4. In 2000 youth unemployment was 4 percentage so, Jordan remains heavily dependent on points over the already high rates of overall donations from the Gulf countries and unemployment. By 2009 this gap had increased donors. to 10 percentage points (World Bank 2011). 14. As assessed by international assessments such 5. Field visits to Jenin were conducted March as the TIMSS and the Programme for 23 and March 24, 2014. Semistructured International Student Assessment (PISA). interviews and focus group discussions were 15. TIMSS, 2003, 2007, and 2011. 100 TRUST, VOICE, AND INCENTIVES 16. A preset questionnaire is developed for all 25. Students did seem to actually use the sugges- four key areas and distributed to the stake- tion box, but even if they did so only infre- holders based on a random sample of the quently, the box’s very existence conveyed an school’s database. important symbolic signal that student input 17. The Education Council (EC) is composed of to decision making was desirable and useful. all school principals within the cluster, stu- 26. Completing a questionnaire obviously requires dent representatives, community representa- literacy, and it seems the database of parents tives, and the head of the EC, who is an only included those who were literate, leaving elected community member. The EC approves it unclear (to us) how the views of illiterate the SIP and shares it with the Field Directorate. parents were obtained. The directorate is responsible for monitoring 27. The Educational Development Council at the progress and reporting to the Ministry of directorate level consists of a president, a dep- Education. uty, and three members: the head of the LEC 18. The integration of decision making and within the school cluster, members of the field improved cooperation between school and directorate development team, and a student community is one approach detailed in newly team consisting of two members. Its main adopted Act 6/L of Ministry of Education responsibilities include (1) participating in Law Number 3 (1994) and its amendments. activities and competitions; (2) strengthening 19. The most recent SDDP assessment was con- the mutual understanding of the community ducted under the Monitoring and Evaluation partnership by encouraging schools and oth- Partnership (MEP) project and published in ers to share their experiences and success sto- May 2013. MEP is a four-year (2010–15) ries; and (3) discussing challenges during project funded by the U.S. Agency for implementation of the SIPs of school clusters. International Development and implemented 28. As measured by student scores (specifically by World Education with the aim of strength- Tawjihi). ening the technical capacity of the National 29. Certain reforms were recently implemented Centre for Human Resources Development by the Ministry of Education in which teach- (NCHRD) and providing financial support ers receive higher salaries for teaching stu- for a series of program quality evaluations of dents for longer hours to deter them from Jordan’s Education Reform for Knowledge resorting to private tutors. Other parental Economy (ERfKE II) program. perspectives on this were that (1) many teach- 20. “As a school principal all the authority and ers are simply not qualified or trained to decision making power was with me,” teach properly (which forces parents to seek explained one principal. “Community mem- tutors); (2) teachers are not paid or given the bers and parents didn’t get involved in the incentives to teach well, and so they do not; planning process.” and (3) the situation has perpetuated stu- 21. The chair of the local Education Council is dents’ belief that they must take private elected by the community and receives train- classes to receive higher grades, regardless of ing for the role through the SDDP project. how good the teacher is. 22. Most countries, of course, have both private 30. The perpetrators of the fire remain at large, and public schools, but even if some students although it is suspected that youths were are receiving education in private schools, responsible. That “no one knows” who com- that sector itself is still regulated (and in some mitted the crime seems implausible in a small cases actively funded) by the state. community, not least because certain aspects 23. The importance of the complementary rela- of how the fire was spread appear quite stra- tions between informal and formal systems tegic, but for now we take it as given. has a long history in social science, but in 31. Technically, we cannot formally verify that recent years it has been articulated most these positive examples of school-community prominently by Nobel Laureate Douglass interaction would not have happened any- North (1990). way (that is, we do not have a “counterfac- 24. The Friends of the School Council is com- tual”). One could argue that local culture and posed of one influential member of every fam- Bedouin traditions dictate that one cares for ily who sends their child to the school at hand. one’s neighbor and community, and thus that LO C A L S U C C E S S E S : S AT I S FA C T I O N , A C C O U N TA B I L I T Y, A N D Q U A L I T Y AT T H E LO C A L L E V E L 101 these examples cannot be attributed to the / r a m a m o h e / P E R% 2 0 E d u c a t i o n% 2 0 reforms. Even so, we are not trying to make a Palestine%20 -%20Final%20Report%20 causal empirical claim. Because of the region’s 10%20September%202013.pdf. reputation as a place where positive collective Krishnan, N., T. Vishwanath, A. Thumala, and action is relatively rare, we are using these P. Petesch. 2012. Aspirations on Hold? Young examples as illustrative evidence of what did Lives in the West Bank and Gaza. Washington, take place within the auspices of a developed DC: World Bank. education council and of what positive col- North, D. 1990. Institutions, Institutional Change lective action can achieve in the sphere of and Economic Performance. New York: enhancing public education. Cambridge University Press. 32. The Health Care Accreditation Council Pascale, R., J. Sternin, and M. Sternin. 2010. The (HCAC) was established in 2007 as a private, Power of Positive Deviance: How Unlikely not-for-profit organization that oversees and Innovators Solve the World’s Toughest implements the accreditation of health care Problems. Cambridge, MA: Harvard Business facilities in Jordan (see http://www.hcac.jo). School Press. 33. In 2010 Morocco had 2,689 primary health Pritchett, L. 2013. Schooling Ain’t Learning: The care centers (World Bank 2013, 7). Rebirth of Education. Washington, DC: 34. The sixth edition of the CQ was completed in Center for Global Development. 2014. World Bank. 2004. World Development Report 35. Zakat is form of obligatory alms giving paid 2004: Making Services Work for Poor People. by Muslims to other poor Muslims. It is one Washington, DC: World Bank. of the five pillars of Islam, and all practicing ———. 2011. Coping with Conflict? Poverty and Muslims who have the financial means are Inclusion in the West Bank and Gaza. expected to pay the alms. Washington, DC: World Bank. https://open k n o w l e d g e .w o r l d b a n k . o r g / h a n d l e /10986/2774. References ———. 2013. Fairness and Accountability: Claussen J., M. Kiernan, and R. Gramshaug. Engaging in Health Systems in the Middle 2 013. Pu bl i c E x pe n d i t u re R e v i e w of East and North Africa. Washington, DC: Education in Palestine. September. http:// World Bank. w w w. m o h e . g o v. p s / ( S ( l o p s a q i k f f t g q v 45 s r j b t r 55 ) A (s C V M l O1 E z w E k A A A A YjNi Z TQ4OTMtYmY1YS0 0OW FhLTg0 Data source Ym E t M G Z i Y j I 2 N D g x M z Q 4J 5 2 b 6 e E d S t at s , World B a n k , ht t p: //d at atopi c s DpOUkbJvo71Nxl8J T5W hM1))/ Uploads .worldbank.org/education/ The Historical and Institutional PART II Drivers of Performance P art I revealed that although local successes exist, the economies in the Middle East and North Africa (MENA) are not meeting their potential in providing their citizens with the education and health services they believe the state should deliver. Why does service delivery fall short of potential in the MENA region? The answer to this question lies in the “cycle of poor performance” that has emerged in much of the region. State institutions lack both internal and external accountability, in part because of the shortage of information on performance that could guide centralized decisions and in part because of the lack of incentives toward establishing accountability mechanisms for performance in the delivery of public sector services. When institutions are weak, service delivery policies are not imple- mented successfully. Citizens’ engagement: formal and informal Institutions: Cycle of political performance at administrative, Citizens’ trust in the national and and social public institutions local levels Performance: effort and ability to meet citizens’ needs 103 104 TRUST, VOICE, AND INCENTIVES Low trust in institutions undermines bottom-up pressures for improving service delivery. Citizens can provide incentives for public services delivery through choice (using public services), voice (giving feedback to providers), and voting (choosing political leaders who support service delivery systems). However, in the face of weak institutions, poor performance, and low trust, people often disengage. They turn instead to local nonstate actors and institutions for services. When they do demand services from the state, citizens tend to do so through informal channels and seek piecemeal, selective solutions to their individual problems. Chapter 4 explains how historical experience has led citizens to value education and health, has fostered their dependence on the state, and has limited the state’s responsiveness. Chapter 5 provides a detailed picture of the political, administrative, and social institutions that affect service delivery, setting the scene for a more detailed look at service delivery performance in chapters 6–9. MAP II.1 Government Effectiveness Index values for MENA, 2013 IBRD 41534 | MARCH 2015 Black Sea Ca s pi an Sea M edite rran ea n Sea ATLANTIC Me SYRIAN TUNISIA MALTA di te OCEAN rrane an Sea LEBANON A.R. WEST BANK AND GAZA IRAQ ISLAMIC REP. MOROCCO OF IRAN JORDAN KUWAIT A L G E R I A L I B Y A ARAB REP. BAHRAIN OF EGYPT SAUDI QATAR UNITED ARABIA ARAB Government Effectiveness EMIRATES OMAN Index, 2013 Re d Se More than 1.0 a 0.1 – 1.0 REP. OF -0.4 – 0 YEMEN en Arabian -1.0– -0.5 Ad MENA average -0.36 f of Sea Gul Less than -1.0 DJIBOUTI Non-MENA/non-OECD average -0.25 No data available OECD average 1.31 Data source: Worldwide Governance Indicators, World Bank. Source: World Bank (IBRD 41534, March 2015). Historical and Cultural Roots of Citizens’ Attitudes and 4 State Performance • Drawing on their historical and cultural roots, citizens of the countries in the Middle East and North Africa value education and health services and expect the state to provide them. • Colonial and postcolonial rules have reinforced the political salience of subnational identity groups, including tribes and large families, regions, and ethnic and sectarian divisions. • The political economy of the region has brought about weak institutions, often incapable of meeting citizens’ demands. A s we saw in chapter 1, the citizens of entrenched interests. The agreement that was the Middle East and North Africa finally struck called for the states to provide (MENA) place a high value on educa- citizens with education, health care, and tion, health, and other services, and they employment in return for acceptance and expect the state to provide such services. But little resistance. At first, this arrangement this has not always been true. In earlier years, resulted in more state-provided services, providing these services was the responsibil- including a significant expansion of student ity of religious communities and private enrollment, the establishment of new hospi- households. Later, the state began to provide tals, and other improvements in public service services, seeking to protect its own military delivery. And yet the political compact and commercial interests. Even as public edu- thwarted the creation of the mechanisms that cation systems emerged, however, the provi- would ensure voice and accountability, ulti- sion of education was primarily elite-driven.1 mately contributing to heightened popular Only in the postindependence period did expectations, social divisions, and the atro- the widespread provision of public services phy of political institutions. As revealed in become part of a new social compact. It arose the coming chapters, the result has been a in many cases from the political battle raging huge demand for services, widespread disap- between the newly emerging elites and pointment in their provision (particularly in 107 108 TRUST, VOICE, AND INCENTIVES nonoil states), diminishing trust in public and eighth centuries, but access to services institutions, and disengagement. was la rgely re st r ic ted to t he el ite s. The Abbasid Caliphate (750 –1258 A.D.) became a seat of knowledge and advance- Education and health as elite ment, with the Abbasids establishing the privilege House of Wisdom (Bayt al-Hikma) that The three monotheistic religions that have brought together scholars from across the shaped the MENA region hold education and world to translate and expand on research health in high regard. For example, in the and teachings from Europe to China. Islamic Hadith, as narrated by Al-Tarabni, Scholars made important advances in science, the Prophet Muhammed encourages educa- technology, and medicine. The fi rst hospital tion, saying, “Seeking knowledge is obliga- was established in Baghdad in 805 A.D., and tory on every Muslim.” The Qur’an (5:32) at one time as many as 60 hospitals are said states that “if anyone saved a life, it would be to have existed in the capital city (Nagamia as if he saved the life of the whole people, 2003; National Library of Medicine 2011). thereby placing great importance on the role However, access to hospitals in the region’s of doctors.”2 So, too, in Judaism one finds in major cities was limited primarily to elites. the Talmud that “whoever saves a life from For the first 500 years of the Ottoman Israel, the Scripture considers it as if he saved Empire’s long rule (1299–1922), successive an entire world” (Babylonian Talmud, caliphates believed health services, hospitals, Sanhedrin 37a). Christianity highly values and clinics were the responsibility of charita- the extension of education and progressive ble organizations, and education was the medicine to local communities. Thus in the responsibility of the family, who could teach 18th and 19th centuries, Christian mission- children at home or seek the help of scholars aries were among the pioneers in establishing in the kuttab. local schools, universities, and clinics.3 In the 19th century, however, Ottoman Because Arab societies initially viewed rulers turned to Western education in the education and health as private affairs, fami- wake of rising security threats from the lies and communities took responsibility for recently modernized European powers. As providing these services. In education, for Gasper explains, in the 1820s, Mohammed example, parents chose whether and how to Ali, the Turkish commander who ruled educate their children. Often, both boys and Egypt, at first under Ottoman command, girls4 were taught how to read and were inau- “established modern schools, sent promising gurated into the fields of Arabic poetry and students abroad to complete their studies, the art of oration. There were also close ties and brought in foreign advisers and experts between religion and education. In Saudi to train military officers and teach at new sci- Arabia, for example, “the child’s education entific and technical institutes” (Gasper began at home. As soon as he could speak it 2013, 13). Reforms focused on education, but was the father’s duty to teach him ‘the word,’ access to education was initially limited to a La ilah illa Allah (no God but Allah). When small segment of the population. Similarly, six years old, the child was responsible for Moroccan rulers and the Ottoman Empire the ritual prayer. It was then that his formal introduced a “New Order” (Nizam Al-Jadid) education (consisting of religious studies, lit- and the “Auspicious Reordering” (Tanzimat, eracy and light mathematics) began. The ele- 1839–1909) in the hope of matching the mentary school (kuttab) was an adjunct of development of their European neighbors the mosque, if not the mosque itself” (Trial, and conveying the appearance of European- Winder, and Bayly 1950, quoting Hitti). style statehood (Anderson 1987). These The elites began to recognize the value of reforms began to change the nature of education and health care, particularly dur- citizen-state relations in which the state inter- ing the Golden Age of Islam in the seventh acted more closely with its citizens and H I S TO R I C A L A N D C U LT U R A L R O OT S O F C I T I Z E N S’ AT T I T U D E S A N D S TAT E P E R F O R M A N C E 109 became involved in new domains. And yet better locations, and were better financed Moroccan rulers and the Ottoman Empire than the civil schools. In addition, military focused primarily on expanding the army, schools were free of charge, whereas civil economy, and state bureaucracies that would schools charged high tuition fees, which support the creation of a middle class that undoubtedly had an effect on student enroll- could compete in the areas of commerce and ment (for example, by 1900 the civil prepara- foreign trade. The military and provincial tory school in Baghdad had attracted administration were restructured, standing 96 students, whereas the military school had armies were established, and a new adminis- enrolled 256 students). Children from elite trative staff with greater technical skills was families dominated the civil schools, and fostered, all accompanied by higher taxes the families continually lobbied for more. (Anderson 1987). The sultan hoped education would bind the Ottoman rulers wanted to import Western people of the empire together, and so the education models because they perceived expansion continued—from 1893 to 1894 such an education to be a means to acquiring more than 50 military middle and prepara- military power and improving economic tory schools were constructed across the productivity. During the late Ottoman empire. Yet even as the empire ended in 1922, period, Sultan Abdul Hamid II encouraged public education was far from universal. the development of education by aspiring not This was even truer outside the Ottoman only to Western attitudes toward education, Empire. The Arabian Peninsula and the cur- but also to Western structures of schooling rent Gulf countries persisted in their nomadic that separated children by age and placed and tribal traditions. In Saudi Arabia, for them within a single institution. These example, Bedouins were mainly illiterate, changes in education began with the intro- although they had deep knowledge of their duction of military schools in 1834 (Provence country’s geography and history. Poetry and 2011). By 1845 the government had decreed the Qur’an were of utmost importance; that a military preparatory school should be Bedouins could easily recite pages of folk built in all provincial capitals with an army poetry and knew the Qur’an by heart (Trial, corps headquarters. In addition to military Winder, and Bayly 1950). Public education, schools, missionaries established and oper- however, was unknown. ated schools across the empire in cities such Similarly, only a minority of the popula- as Beirut, Istanbul, and Damascus, including tion enjoyed modern health care. The role of a missionary college that was founded in the missionaries in the health industry was less mid-1860s (Provence 2011). However, recog- contentious than it was in education, and so nizing that the notion of sending Muslim they were allowed to operate and contributed children to missionary schools did not appeal significantly to expanding health care, par- to elite families, the sultan issued an act that ticularly in the Levant (Bourmand 2008; introduced public education. Chiffoleau forthcoming). And yet most state The new Education Act of 1869 estab- efforts to expand health care focused on lished a multitiered civil education system securit y, and especially the militar y. with two options, military (askariyye) school Sanitation reforms and quarantines were or civil (mulkiyye) school. The law dictated implemented to protect the army from the that elementary schools (ibtidaiyye) were epidemics that were widespread at that time. to be built in each village, middle schools Municipal hospitals were built in the prov- (rusdiyye) in each town, and an Imperial inces, and the posts of civilian sanitary doc- Sultani Lycee (idadiyye) preparatory school tors were multiplied to contain the spread of in every provincial capital. This was an ambi- infectious diseases. However, graduates of tious endeavor, and so implementation was the medical schools in Cairo and Istanbul, slow. Military schools were given priority. the most prestigious institutions at the time, They were constructed fi rst, were placed in were quickly absorbed into the army in an 110 TRUST, VOICE, AND INCENTIVES effort to keep the military strong and able to Eastern population gained access. Those who protect its territory. Before the reforms, had the means traveled abroad to Europe and Ottoman rulers paid little attention to health the United States. For the most part, how- care. The sale of food and medicine was reg- ever, the majority did not pursue a higher ulated as part of general commercial regula- education. Illustrative of the limited access to tions in order to prevent economic fraud and education was Algeria, which in the 1950s ensure the adequacy of provisions for the had an indigenous population of 10 million. capital city. Only 7,000 were in secondary school, and only 600 attended a university (Richards and Waterbury 1996). From privilege to state building Those elites who gained access to educa- The decline of the Ottoman Empire altered tion were given positions in colonial bureau- the process of state formation and adminis- cracies, thereby expanding the manpower trative development and consequently the used to modernize the state. This new civil provision of education, health, and other service was needed to manage roads, rail- services. The impact was felt at different roads, ports, and power; to run the mail and times for different countries. For example, telegraph services; to identify and tax the Algeria moved from Ottoman to French population; to staff all echelons of the local control in 1830 and would remain so until administration; and to assume all the inter- 1962, and Tunisia became a French protec- mediate positions between the colonial torate in 1881. But Lebanon, Syria, and other authorities and the population. Thus by countries in the Levant remained under essentially creating a new upper class of man- Ottoman rule until the end of World War I. agers and officers, middle class of techni- Imperial powers also exercised power differ- cians, and lower class of workers (Polk 1965), ently across territories (Owen 1992), ranging the colonizers were able to sustain their from direct colonies to mandates, protector- control and enforce policies that increased ates, condominiums, and direct treaties. services and finances for their own benefit. There were also differences based on the Similarly, the colonial interests determined degree of rule (direct or indirect), type of health care policies. Sanitation policies, for government (monarchy or republic), and the example, were used to protect colonizers’ political importance of the European settler military and economic interests—and in community. All these factors determined how some cases as a tool to “civilize” populations power was exercised within the newly created that the colonizers considered backward, states, and they would ultimately influence ignorant, and dangerous (Camau, Zaiem, the extent to which populist welfare regimes and Bahri 1990; Arnold 1993; Rivet 1995; would emerge in the independence period, Mitchell 2002). The outbreak of smallpox in mobilizing support on the promise of public Palestine in 1921 proved to be such a case services. (Davidovitch and Greenberg 2007). The In general, colonizers concentrated public health reforms under the Tanzimat largely services in capitals and large urban areas, missed Palestine, and thus disease was thereby limiting access to a small elite group. widespread; malaria and trachoma were Education, in particular, was used as a tool common ailments. The British rulers exerted to win favors, control segments of the popu- much effort to rid Palestine of infectious lation, and staff colonial administrations. diseases, installing new sewage systems, cre- Control was achieved by limiting access to ating swamp drainage projects, and conduct- secondary education to elite families in order ing hygiene education campaigns. They to win their loyalty. Cairo University and also introduced new regulations and poli- American University of Beirut were the only cies for licensing health care professionals, two functioning universities in the region, pharmaceutical and food regulations, and and just a small segment of the Middle quarantine measures. H I S TO R I C A L A N D C U LT U R A L R O OT S O F C I T I Z E N S’ AT T I T U D E S A N D S TAT E P E R F O R M A N C E 111 Such programs had limited results, how- social and economic development across ever. Many of them were only partially populations (Abu-Laban and Abu-Laban implemented because the British investments 1976). They also sought to mobilize popula- were limited. And local populations often tions that were hugely frustrated with the feared modern medicine. For example, when socioeconomic gaps between the colonial vaccinations were introduced, many locals elite and the rest of the people. In the heady resisted the injections by hiding in caves and days of the post–World War II period when remote areas when physicians visited their state-led development and populism were villages. Overall, the colonial powers did not gaining strength globally, nationalist leaders pay much attention to the health needs of the throughout much of the region used the locals except when the health of the local promise of services to gain support of the populations was impeding colonial economic masses. Countries whose militaries were ori- production or enterprises. ented toward the lower classes in order to The colonial powers not only left vast overthrow the incumbent elites as well as segments of the population without adequate socialist-oriented republics with large education and health care, but also reinforced welfare states began to emerge. Examples social divisions and thwarted development were the Arab Republic of Egypt, Iraq, of effective administrative institutions. and the Syrian Arab Republic. In other According to Owen (1992), three policies countries such as Jordan and Morocco, underlay colonial rule: (1) to favor certain which managed to hold off such threats, ethnicities (such as Arab over Berber), conservative forces remained in power, often regions, and tribes; (2) to create alliances in monarchies. Yet they too needed to with large landowners, who for their loyalty respond to popular pressures, in part by received tax exemptions and the legal author- promising public services (Waterbury 1970; ity to rule over their peasants; and (3) to Yoav 2007; Lust-Okar 2008, 2009a, 2009b). manage economies in such a manner that left Finally, in the Gulf states, where indepen- the Middle Eastern economies open and sub- dence came later and oil rents were avail- ject to influence from the colonial rulers and able, the ruling families founded their other markets (for example, states were not regimes on a promise of services for loyalty allowed to have their own central bank and (Anderson 1987; Beblawi and Luciani 1987; remained relatively constrained in their abil- Crystal 1990; Herb 2005). ity to raise revenue). As a result, tribes, fami- Eventually, a social compact emerged to lies, and other groups gained political bind governments to their citizens. The new salience; wide chasms in society fostered pop- leaders promised to improve living standards ulist demands; and state institutions and equality and to introduce mass access to remained underdeveloped (Tibi 1990; public services such as education and health. Shryock and Howell 2001; Hourani 2010). Postindependence constitutions echoed the role of the state as the institution responsible for social, political, and economic policies; in Independence and the new essence, the new constitutions introduced the social compact: Services, a right state as an agent of public welfare. Policies for all fortifying the social compact included reli- Arab countries shifted their provision of ance on state planning in setting economic public services after they gained indepen- priorities, nationalization of private and dence (for most of them, after World War II). foreign assets, food and fuel subsidies, and Leaders wanted to break away from the poli- the overall centralization and control of polit- cies set by the European colonizers and ical parties, unions, and other associations focus instead on industrialization, better liv- in an effort to limit the political sphere to ing standards, higher incomes, and the one that conformed to the unity of the expansion of services that would improve state (Yousef 2004). These policies instilled 112 TRUST, VOICE, AND INCENTIVES in citizens the expectation that the state represents a 13 percent annual growth rate would provide services. (Abu-Laban and Abu-Laban 1976). By 1970 In addition, the discovery of oil and six Arab states (Jordan, Kuwait, Lebanon, expansion of the oil industry in some coun- Libya, Syria, and Tunisia) had nearly univer- tries made it possible to offer generous wel- sal educational systems at the primary level fare, leaving citizens to anticipate that (A bu - L aba n a nd A bu - L aba n 19 76). services would be their slice of the oil booty. The expansion of schools was accompanied In the Gulf states, education, health, and by efforts to improve student access and other social services were introduced in the enrollment. In 1971 the average net enroll- 1950s as part of a welfare state in which local ment rates in the MENA region were populations would receive services free of 59 percent, compared with 90 percent in charge or pay a small fee to gain access member countries of the Organisation for (Bahgat 1999). In Kuwait, the discovery of oil Economic Co-operation and Development in 1946 led to the establishment of the fi rst (OECD). By 1975 the MENA average had public hospital. The number of medical pro- jumped to 66 percent, and in 1990 it reached fessionals grew from 3,840 in 1970 to 13,616 82 percent. By 2010 MENA countries had a in 1986 (Salid 1991). In the United Arab net enrollment rate (95 percent) that was com- Emirates, primary health facilities increased parable to that of the OECD countries from 45 clinics in 1977 to 105 health centers (96 percent). The female enrollment rate, by the end of 1999. which in 1971 was only at 44 percent of pri- Public education was also introduced to mary school–aged girls, had by 2010 reached create a skilled workforce to support the mod- 93 percent in the MENA countries. The num- ernization process. To promote and encour- ber of university students rose from 20,000 in age families to send their children to school, 1945 to about 400,000 in 1971 (Issawi 1982). education was offered free of charge, and During the early period of independence, some enrolled students received books, uni- health facilities still reflected the era of elitism forms, and monthly allowances (Bahgat that preceded that period. They were situated 1999). In Saudi Arabia, development of the oil only in large cities, were ill-equipped, and industry was accompanied by Western influ- were under the administration of religious ences, which also seeped into the field of edu- missions. Meanwhile, the medical workforce cation. An American boys’ school opened in was limited—for example, in 1952 Egypt had Saudi Arabia’s Eastern Province, teaching one physician per 2,700 persons (Chiffoleau English, basic mathematics, and other courses 1997), and in 1955 Syria had one physician in vocational education. Those who gradu- per 4,000 persons (Camau, Zaiem, and Bahri ated with these skills were able to enter the oil 1990). However, some strides were being industry and earn salaries that were three made in curbing disease and raising health times higher than those of their untrained indicators. For example, the spread of vacci- peers. Following the discovery of oil in nation centers across the region helped reduce Kuwait, it expanded its public schools, and the incidence of childhood diseases as well as enrollments grew from 51,090 in 1961 to infant mortality rates. 364,412 in 1986 (Salid 1991). Countries across the MENA region thus saw major changes in the education sector. The expan- Institution building in newly sion of public education translated into the independent states secularization of schools and the formaliza- The expansion of public services was impor- tion of education by standardization and set- tant in building national support for the ting a curriculum. The number of primary newly established regimes and promoting school teachers in 13 Arab states combined social and economic development in the for- more than tripled between 1950 and 1971, mer colonies, but it also emerged in the con- rising from 103,004 to 382,477, which text of, and contributed to, weak political H I S TO R I C A L A N D C U LT U R A L R O OT S O F C I T I Z E N S’ AT T I T U D E S A N D S TAT E P E R F O R M A N C E 113 and administrative institutions. The adminis- a key component of efforts to strengthen and trative institutions that emerged were stabilize political regimes. However, the intended to implement redistributive policies, political logic that drove policies and imple- but most Arab states emerging from the era mentation resulted in bloated bureaucracies, of colonization had weak administrative inefficient state-owned enterprises, and capacities and competence. Countries such as higher deficits (Nabli 2007; Cammett and Egypt and Tunisia, which had gained early Diwan 2013). autonomy from the Ottoman Empire and Incumbents also used access to choice were subjected to less direct colonial rule, positions in large bureaucracies and state had functioning national administrations by resources to reward political allies. Often, the early 20th century, while countries such they favored members of certain tribes, sects, as Libya and Lebanon had weak administra- regions, or other social groups, exploiting tive penetration and ill-fitted bureaucracies identities that had been reinforced during the with no experience, making full territorial colonial era. In countries such as Saddam control difficult. Other states were still strug- Hussein’s Iraq, Jordan, the Republic of gling with the social structures and econo- Yemen, and Saudi Arabia, state elites used mies that were bunched together via the tribes as the “building blocks of the modern arbitrary territorial lines drawn by the state” (Alon 2007, 7—also see Ayubi 1995; Europeans. This was an acute problem for Jabar 2001; Sakai 2001). As Shryock and the nomadic tribal families in Iraq, Jordan, Howell (2001, 266) explain about Jordan, and the Gulf states. the Hashemites “built a political system that In addition, most newly independent states corresponds to, addresses, and depends on lacked the funding needed to support exten- these houses [Jordanian tribes] in fundamen- sive social services (Polk 1965). The rich oil- tal ways: as targets of incentive, punishment, producing countries, particularly the Gulf and reward; as sites and methods of recruit- states, were an exception. They generally had ment to public office; and as a means of the economic resources to support an expan- exclusion from power.” Elsewhere, most sive public services program, although they notably in Lebanon, sectarian divisions were were affected by the instability in oil prices. the basis of the political order. However, the They were also able to develop positive rela- use of social identities in establishing political tions with their citizens, who increasingly order did not translate into political inclu- came to see themselves as minorities in sion; pacts often remained at the elite level, largely expatriate communities, thereby fur- with few benefits trickling down to constitu- ther deepening support for their leaders. ents (Cammett 2009; Corstange 2010). Nor Often, however, countries that promised did this strategy foster political compromise. their citizens free public services, such as Indeed, as Waterbury (1970) argued about Egypt, Libya, the Republic of Yemen, and Morocco, managing confl ict across groups Syria, struggled to finance those efforts. served to both help ruling elites maintain Minor alarms were raised early on about power and reinforce these identity groups. the sustainability of this route, and in some instances, budget deficits occurred, resulting in inadequately supplied facilities Economic crises, neoliberal (Saleh et al. 2014). policies, and the failure to Despite these challenges, and in the face of reconstruct the social compact the initial stages of instability that included By the early 1980s, the situation in much of military coups, the Arab states continued to the region had become unsustainable, but it expand services. They promised universal was difficult to reverse. Populist politics, health care, public education, and even jobs combined with falling oil revenues, increas- to all graduates. Governing elites saw such ingly large young populations, and ineffi - policies as crucial for development, as well as cient bureaucracies, strained economies. 114 TRUST, VOICE, AND INCENTIVES In particular, those regimes that did not countries varied greatly over the period 1995– have oil revenues were forced to implement 2012, but there was no significant decrease. austerity programs. Consumer subsidies The levels of spending in low- and were reduced, and agricultural and indus- middle-income MENA countries on both trial prices were liberalized. This led to a education and health care are similar to those sharp increase in food and kerosene prices. in countries with similar levels of gross For example, the price of flour and flour domestic product (GDP) per capita, although products increased by 50 – 67 percent in the Gulf countries tend to spend less when Egypt in 1977, by 40 percent in Morocco in compared with OECD member countries (see 1981 and another 34 percent in 1984, and figures 4.1 and 4.2). For example, in terms of more than doubled in Tunisia the same the primary education expenditure per stu- year. In 1989 the Egyptian government dent as a percentage of GDP per capita, in waived state subsidies for flour a second 2012 Jordan spent 12 percent and the Islamic ti me, and t he Jordan ian gover n ment Republic of Iran 16 percent, whereas Kuwait reduced subsidies for fuel and cigarettes. spent 17 percent and Oman 14 percent Meanwhile, Algeria and Egypt reduced (slightly lower percentages than countries state expenditures on welfare programs, with similar GDPs per capita). Figure 4.3 especially health care (Bienen and Gersovitz shows that in 2011 Jordan, the Islamic 1986; Seddon 1989; Sadiki 2000; Rivlin Republic of Iran, Kuwait, and Oman had and Even 2004; Devlin 2010). similar spending levels on secondary educa- Citizens vehemently resisted cuts in these tion, whereas the Republic of Yemen was areas (Prasad 2014, 15–18) . The urban spending only 11 percent of its GDP on sec- middle and lower classes took to the streets ondary education per student—a percentage protesting austerity programs. Subsidy cuts that is low when compared with those of and higher prices led to demonstrations countries with similar GDPs per capita. and riots in Morocco in late 1983, Tunisia However, citizens have also seen their out- in 1984, Sudan in 1982 and 1985, Algeria of-pocket expenditures for education and in 1988, and Jordan in 1989 (Richards health care increase, even as governments and Waterbury 1996, 268). Governments have failed to reduce their fiscal burdens. responded in many cases by restoring subsi- Weak and captured institutions, as discussed dies. They also opened up opportunities in in the next chapter, have led to inefficient use the private sector, called elections, (re)con- of funding and the poor-quality services out- vened parliaments, allowed the establish- lined in chapter 2. The result is that citizens ment of new political parties and civil society who can increase their spending receive the organizations, and attempted to reform services they need. Trends in primary school public services. enrollment indicate that households are And yet despite budgetary pressures, the spending more out of pocket for private MENA countries generally did not reduce schooling, particularly in the Gulf countries spending on education and health. According where families are more likely to have dispos- to data from the United Nations Educational, able income (figure 4.4). On the other hand, Scientific and Cultural Organization citizens of the Gulf countries tend to spend (UNESCO) Institute for Statistics, these coun- less out of pocket on health than their peers tries increased their per capita spending on pri- in the MENA countries. In the Republic of mary and secondary education throughout the Yemen, for example, out-of-pocket expendi- 2000s; only Jordan and the Islamic Republic of tures make up 72 percent of total health Iran saw a slight decrease in spending in 2011. expenditures, whereas they make up only Similarly, according to data from the Global 9 percent of total health expenditures in Health Observatory Data Repository at the Qatar (World Development Indicators, World Health Organization (WHO), the per 2012). Today, households in MENA coun- capita expenditures on health by the MENA tries pay on average 6 percent of their total H I S TO R I C A L A N D C U LT U R A L R O OT S O F C I T I Z E N S’ AT T I T U D E S A N D S TAT E P E R F O R M A N C E 115 FIGURE 4.1 Per capita government expenditure on health: MENA and other countries, 2012 Per capita government expenditure on health 5,000 (PPP-adjusted), current international $ 4,000 3,000 2,000 12 1,500 10 1,000 15 13 14 85 9 6 7 500 1 2 4 11 0 0 00 00 00 0 0 0 00 00 50 ,00 ,00 ,00 1,0 2,5 5,0 0,0 0,0 10 25 50 10 15 GDP per capita (PPP-adjusted), current international $ 95% CI Fitted values Low-income MENA countries Middle-income MENA countries GCC countries OECD Other countries Sources: World Development Indicators, 2012; Global Health Observatory Data Repository, 2012. Note: PPP = purchasing power parity; GDP = gross domestic product; CI = confidence interval; MENA = Middle East and North Africa; GCC = Gulf Cooperation Council; OECD = Organisation for Economic Co-operation and Development. Low-income MENA countries: 1 = Djibouti, 2 = Republic of Yemen; middle-income MENA countries: 3 = Islamic Republic of Iran, 4 = Iraq, 5 = Jordan, 6 = Lebanon, 7 = Libya, 8 = Tunisia, 9 = Algeria; GCC countries: 10 = Kuwait, 11 = Oman, 12 = Qatar, 13 = Saudi Arabia, 14 = United Arab Emirates, 15 = Bahrain. household expenditure on health care, with the following decade (Diwan 2013). At the most of this spending used on medications, same time, promises of democracy never physician visits, and diagnostic services materialized. Parliaments, political parties, (Elgazzar et al. 2010). Lower-income coun- and civil societies remained weak, and state tries such as Egypt, Morocco, the Republic of institutions lacked the accountability, trans- Yemen, and Syria have higher out-of-pocket parency, and enforcement mechanisms spending, placing the poor at a disadvantage needed to translate citizens’ demands into that could push them deeper into poverty or effective policies. force them to forgo health care. Moreover, after the fall in 2011 of Egyptian president Hosni Mubarak and High expectations, weak Tunisian president Zine El Abidine Ben Ali, institutions, and low evidence emerged that the small minority of performance politically connected benefited dispropor- Despite weak institutions and broken prom- tionately from the political reforms estab- ises, citizens’ expectations that the state lished in the wake of the economic crises of would play a major role in providing educa- the 1980s. As a result, the gap between the tion, health, jobs, and other services did not rich and poor widened. 5 Meanwhile, as dissipate.6 However, populations lost trust in Cammett and Diwan (2013) argue, the alli- the ability of governments to provide quality ance between the middle classes and the state services that addressed their needs, and that had emerged in the 1990s eroded during they saw corruption and lack of freedom 116 TRUST, VOICE, AND INCENTIVES FIGURE 4.2 Per capita government expenditure on primary education (percentage of GDP per capita): MENA and other countries, 2011 60 Expenditure per student (% of GDP per capita) 50 40 Iran, Islamic Rep. 30 20 Yemen, Rep. Kuwait 10 Oman Jordan 0 0 00 00 00 0 0 0 0 0 50 ,00 ,00 ,00 ,00 ,00 1,0 2,5 5,0 10 20 40 60 80 GDP per capita (PPP-adjusted), current international $ 95% CI Fitted values Low-income MENA countries Middle-income MENA countries GCC countries OECD Other countries Source: World Development Indicators, 2011. Note: GDP = gross domestic product; PPP = purchasing power parity; CI = confidence interval; MENA = Middle East and North Africa; GCC = Gulf Cooperation Council; OECD = Organisation for Economic Co-operation and Development. and opportunities as the major reasons for point that the 2013 expenditures were 300 failure. Better education, health, and other percent higher than those in 2010 (Africa services were among the demands of people Report 2014). Influenced by developments in during the Arab Uprisings that began to the MENA region, in Sudan, opposition to a spread across the Middle East in late 2010 cut in fuel subsidies led to unrest, leaving and early 2011. more than 50 dead (Zaid et al. 2014). This is Most governments found it difficult to not to say that subsidies are never drawn reduce subsidies or to limit the public sector back: The Islamic Republic of Iran replaced wage bill, especially in the face of the Arab subsidies to energy products with targeted Uprisings. Under mounting economic bur- cash transfers to the population in 2010, and dens, governments seeking to reduce subsidies Egypt managed to do so in 2014, with and limit the public sector wage bill to ease President Abdel el-Sisi exploiting his honey- budget deficits saw their efforts often pushed moon period in office to make an unpopular back. The governments of Egypt, Jordan, the decision. In general, however, such cuts are Republic of Yemen, and Tunisia attempted to politically difficult because the public contin- cut subsidies in 2011, 2012, and 2014, but ues to expect that the state will provide such reversed those measures in the face of mount- support. ing opposition (Al-Khalidi 2012; Buck 2012; Revised constitutions in Egypt and Tunisia BBC 2014; IRIN 2014). Indeed, from 2010 to have further strengthened citizens’ rights and 2013 the Tunisian government responded to the state’s obligation to ensure appropriate public demands by increasing subsidies to the funding for education. Governments across H I S TO R I C A L A N D C U LT U R A L R O OT S O F C I T I Z E N S’ AT T I T U D E S A N D S TAT E P E R F O R M A N C E 117 FIGURE 4.3 Per capita government expenditure on secondary education (percentage of GDP per capita): MENA and other countries, 2011 50 Expenditure per student (% of GDP per capita) 40 Iran, Islamic Rep. 30 20 Oman Kuwait 10 Yemen, Rep. Jordan 0 0 0 00 00 00 0 0 0 0 ,00 50 ,00 ,00 ,00 ,00 1,0 2,5 5,0 60 10 20 80 40 GDP per capita (PPP-adjusted), current international $ 95% CI Fitted values Low-income MENA countries Middle-income MENA countries GCC countries OECD Other countries Source: World Development Indicators, 2011. Note: GDP = gross domestic product; PPP = purchasing power parity; CI = confidence interval; MENA = Middle East and North Africa; GCC = Gulf Cooperation Council; OECD = Organisation for Economic Co-operation and Development. the region have developed new strategies, pol- kept will largely depend on the ability of icies, and reforms in the education and health countries to orient the incentives of public sectors that demonstrate a commitment to servants, providers, and communities toward improving services. Education reforms their implementation. include improvements in curricula, in teacher Successfully improving service delivery policies (as demonstrated by the Systems depends on more than simply announcing Approach for Better Education Results, rights and policies. The ability of institutions SABER),7 and teacher training. Similarly, in to perform, citizens’ evaluations of such per- the health sector, governments have intro- formances and trust in public institutions, and duced policies and reforms for universal cov- their willingness to engage are essential. And erage and quality in health care. As part of yet achieving a positive cycle of performance the global movement toward universal health requires overcoming some of the problems coverage, Egypt, Morocco, Tunisia, and the that emerged through the postcolonial period: Gulf Cooperation Council countries are striv- citizens’ high expectations, deep social ing to increase health care coverage, and divisions, and weak institutions. As we shall countries have aligned with the World Bank see in the coming chapters, these problems around its regional health strategy with diminished performance, fostered frustration, fairness and accountability at its core. and shaped engagement in ways that often Whether such strategies, policies, and reforms undermined the development of stronger, can succeed and whether promises can be more effective systems. 118 TRUST, VOICE, AND INCENTIVES FIGURE 4.4 Private primary school enrollment as a percentage of total primary enrollment: MENA region, selected years, 1985–2010 80 70 60 50 Percent 40 30 20 10 0 Syrian United Bahrain Jordan Kuwait Morocco Qatar Arab Tunisia Arab Republic Emirates 1985 10.5 6.9 25.9 3.4 19.6 4.5 0.5 24.6 1995 17.9 27.3 31.1 3.7 34.1 3.9 0.6 41.2 2005 23.6 30.4 33.0 6.6 45.3 4.4 1.1 61.1 2010 31.1 33.1 39.1 10.8 55.5 4.2 2.1 71.7 Source: EdStats. Note: The bars represent the difference in private primary school enrollment as a percentage of total primary enrollment between 1985 and 2010. Notes reforms, state formation solidified, and thus administrative positions increased and were 1. In general, elites are those in a society who soon handed over to a new generation of enjoy a disproportionate amount of social, well-educated bureaucrats who performed political, or economic power. In the MENA like their Western counterparts. As the Arab region, there has been some variation in the states gained independence, state administra- nature of elites over time. During the Ottoman tion and state-society relations developed and rule, elites were those individuals who served so did the nature of the state’s beneficiaries, the sultan (usually through the collection of the elites. Across the region, elites were for- taxes), who embraced Islam, and who assimi- mulated across sects, class, tribes, and land- lated into Ottoman social practices, including owners, leading to variation, for example, speaking the Ottoman language. As the central among the landowner elites of Morocco, the power of the sultan began to weaken in the elitist tribesmen of the Republic of Yemen, reform and modernization periods, the nature and the Ba’ath party members of Iraq. of the elites also shifted as local households 2. So, too, a Hadith, as narrated by Usamah Sbin within the ruling classes began to gain politi- Shareek, established the legitimacy of medi- cal and economic control, soon becoming cine: “I was with the Prophet, and some Arabs provincial notables who secured military and came to him asking, ‘O Messenger of Allah, administrative functions. Under the Tanzimat should we take medicines for any disease?’ H I S TO R I C A L A N D C U LT U R A L R O OT S O F C I T I Z E N S’ AT T I T U D E S A N D S TAT E P E R F O R M A N C E 119 He said, ‘Yes, O You servants of Allah take / a r t icle / 2012 /11 / 2 2 /us-m ideast- su m m it medicine as Allah has not created a disease -jordan-idUSBRE8AK11120121122. without creating a cure except for one.’ They Alon, Y. 2007. The Making of Jordan: Tribes, asked which one. He replied ‘old age.’ ’’ Colonialism and the Modern State. London: 3. This had complex implications for devel- I. B. Tauris. opment as well as West-East relations. See Anderson, L. 1987. “The State in the Middle Makdisi (2008) and Dog ˇ an and Sharkey East and North Africa.” Comparative Politics (2011). 20 (1): 1–18. 4. Girls usually joined young boys during the Arnold, D. 1993. Colonizing the Body: State early years of education, but once they reached Medicine and Epidemic Disease in Nineteenth- puberty their education was discontinued. C e ntur y India . B erkeley: Universit y of 5. For a discussion of the disproportionate gains California Press. by crony capitalists, see Diwan, Keefer, and Ayubi, N. N. 1995. Over-Stating the Arab State: Schiffbauer (2014) and Rijkers, Freund, and Politics and Society in the Middle East. Nucifora (2014). London: I. B. Tauris. 6. Evidence of this can be found in public opin- Bahgat, G. 1999. “Education in the Gulf ion polls. When asked about whether the state Monarchies: Retrospect and Prospect.” should play a large role in the economy, respon- International Review of Education 45 (2): dents in Egypt and Tunisia were nearly united in 127–36. demanding that it do so. When asked what the BBC (British Broadcasting Company). 2014. “Shia most important components of democracy are, Houthi Rebels and al-Qaeda Clash in South in Egypt, nearly 70 percent of respondents saw Yemen.” BBC News Middle East, October 15. democracy in economic terms—either narrow- http://w w w.bbc.com /news/world-middle ing the gap between rich and poor or provid- -east-29627772. ing basic necessities (Transitional Governance Beblawi, H., and G. Luciani, eds. 1987. The Project, 2012). More than 32 percent in Rentier State. London: Croom Helm. Tunisia agreed (Transitional Governance B ienen , H . S ., a nd M . G ersov it z . 1986 . Project, 2012, 2014) and 12 percent in Libya “Consumer Subsidy Cuts, Violence, and (Transitional Governance Project, 2013). Of Political Stability.” Comparative Politics the three countries in transition, only in Libya 19 (1): 25–44. was there support for a limited state role in Bourmand, P. 2008. “Ya Doktor, Devenir the economy (or capitalist economy) and was medecin et exercer son art en Terre Sainte, Une democracy viewed in terms of civil liberties experience du pluralism medical dans l’Empire and turnover of government through elections Ottoman fi nissant.” PhD diss., Université de (Transitional Governance Project, 2013). Provence, Aix-en-Provence, France. 7. SABER is an initiative to produce compara- Buck, T. 2012. “Protests in Jordan after Fuel tive data and knowledge on education policies Subsidy Cut.” Financial Times Online , and institutions, with the aim of helping coun- November 14. http://www.ft.com/intl/cms tries systematically strengthen their education /s/0/066b952e-2df0-11e2-8ece-00144feabdc0 systems. .html#axzz3GMYcrl00. Camau, M., H. Zaiem, and H. Bahri. 1990. Etat de santé, besoin medical et enjeux politiques References en Tunisie. Paris: CNRS Editions. Abu-Laban, B ., and S. Abu-Laban. 1976. Cammett, M. 2009. “Democracy, Lebanese- “Education and Development in the Arab Style.” Middle East Research and Information World.” Journal of Developing Areas 10 (3): Project, August 18. 285–304. Cammett, M., and I. Diwan. 2013. “Conclusion: Africa Report. 2014. “Tunisia: Democracy After The Political Economy of the Arab Uprisings.” A rab Spring.” Januar y 27. ht tp: //w w w In A Political Economy of the Middle East, .theafricareport.com/North-Africa/tunisia 3rd ed., edited by Alan Richards and John -democracy-after-arab-spring.html. Waterbury, 407–37. Boulder, CO: Westview Al-Khalidi, S. 2012. “Jordan Subsidy Cuts to Press. Spur A id, Grow th: M inister.” Reuters , Chiffoleau, S. 1997. Medecines et medecins en November 22 . ht tp: //w w w.reuters.com E g y pt e , C o n s t r u c t i o n d ’u n e i d e n t i t e 120 TRUST, VOICE, AND INCENTIVES professionnelle et projet medical. Lyon-Paris: IRIN. 2014. “Yemen Fuel Subsidy Cuts Hit Poor Maison de l’Orient Mediterraneen- Hardest.” IRIN Humanitarian News and L’Harmattan. Analysis, August 25. http://www.irinnews.org ———. Forthcoming. “De la peste d’Orient a /report/100535/yemen-fuel-subsidy-cuts-hit l’OMS. Une historie de la santé publique -poor-hardest. internationale.” Centre national de la recherche Issawi, C. 1982. An Economic History of the scientifique, Paris. Middle East and North Africa. New York: Corstange, D. 2010. “T he Parliamentar y Columbia University Press. Election in Lebanon, June 2009.” Electoral Jabar, F. A. 2001. “Sheikhs and Ideologues: Studies 29: 276–96. Deconstruction and Reconstruction of Tribes Crystal, J. 1990. Oil and Politics in the Gulf: under Patrimonial Totalitarianism in Iraq, Rulers and Merchants in Kuwait and Qatar. 1 9 6 8 –1 9 9 8 .” I n Tr i b e s a n d Po w e r : New York: Cambridge University Press. Nationalism and Ethnicity in the Middle East, Davidovitch, N., and Z. Greenberg. 2007. “Public edited by Faleh Abdul-Jabar and Hosham Health, Culture and Colonial Medicine: Dawod, 69–109. London: Saqi. Smallpox and Variolation in Palestine during Lust-Okar, E. 2008. “Competitive Clientelism in the British Mandate.” Public Health Reports Jordanian Elections.” In Political Participation 122 (3): 398–406. in the Middle East, edited by Ellen Lust-Okar Devlin, J. C. 2010. Challenges of Economic and Saloua Zerhouni. Boulder, CO: Lynne Development in the Middle East and North Rienner. Africa Region. Singapore: World Scientific ——— . 20 09a. “L eg islat ive E le c t ions i n Publishing Co. Hegemonic Authoritarian Regimes: Diwan, I. 2013. “Understanding Revolution in Competitive Clientelism and Resistance to the Middle East: The Central Role of the Democratization.” In Democratization by Middle Class.” Middle East Development Elections: A New Mode of Transition, edited Journal 5 (1): 30–60. by Staffan I. Lindberg. Baltimore: Johns Diwan, I., P. Keefer, and M. Schiffbauer. 2014. Hopkins University Press. “Private Sector Growth and Cronyism in ———. 2009b. “Reinforcing Informal Institutions Egypt.” World Bank, Washington, DC. http:// through Authoritarian Elections: Insights from siteresources.worldbank.org / EXTABCDE Jordan.” Middle East Law and Governance, / R e s o u r c e s / 7455 6 76 -14 0 1 3 8 8 9 0 1 5 2 5 3–37. /9576629-1401389020127/ Private_sector M a kd isi , U. 20 0 8. A r till e r y of He a ve n : _ grow th _ and _ cronyism _ in _ Eg ypt _ June A me ric an M i ssion aries an d the Failed _2014-Presentation.pdf. Conquest of the Middle Eas t. Ithaca, NY: Dogˇ an, M. A., and H. J. Sharkey. 2011. American Cornell University Press. Mission aries and the Middle E a st: Mitchell, T. 2002. Rule of Experts: Egypt, Foundational Encounters. Salt Lake City: Tec hn o - politic s , Mod e r nit y . B erkeley: University of Utah Press. University of California Press. Elgazzar, H., F. Rooad, C. Arfa, A. Mataria, Nabli, M. K. 2007. Breaking the Barriers to N. Salti, J. Chaaban, D. Salehi-Isfahani, Higher Economic Growth: Better Governance S. Fesharaki, and M. Majburi. 2010. Who and Deeper Reforms in the Middle East and Pays? O ut- of- Pocke t He alth S pe n ding North Africa. Washington, DC: World Bank. and Equit y Implic ations in the Middle ht t p s: //op e n k nowl e d g e .worldb a n k .or g East and North Africa . Washington, DC: / bit st re a m / h a nd le /10986 /6914 World Bank. /439690PUB0Box310only109780821374153 Gasper, M. 2013. “The Making of the Modern .pdf?sequence=1. Middle East.” In The Middle East, 13th ed., Nagamia, H. 2003. “Islamic Medicine History edited by Ellen Lust, 1–71. New York: Sage. and Cu rrent Practice.” Jour n al of the Herb, M. 2005. “No Representation without International Society for the History of Ta x a t i o n ? R e n t s , D e v e l o p m e n t , a n d Islamic Medicine 2 (4): 19–30. Democracy.” Comparative Politics 37 (April): National Library of Medicine. 2011. “Islamic 297–316. Culture and the Medical Arts: Hospitals.” Hourani, A. 2010. A History of the Arab Peoples. http://www.nlm.nih.gov/exhibition/islamic Cambridge, MA: Harvard University Press. _ medical/islamic_12.html. H I S TO R I C A L A N D C U LT U R A L R O OT S O F C I T I Z E N S’ AT T I T U D E S A N D S TAT E P E R F O R M A N C E 121 Owen, R. 1992. State, Power, and Politics in the Universal Health Coverage in the A rab Making of the Modern Middle East. London: Uprising Countries Tunisia, Egypt, Libya and Routledge. Yemen.” The Lancet 383 (9914): 368–81. Polk, W. 1965. “The Nature of Modernization: S a l i d , K . O. 1 9 91. “ K u w a i t : Po l i t i c a l The Middle East and North Africa.” Foreign Consequences of Modernization, 1750–1986.” Affairs (October). http://www.foreignaffairs Middle Eastern Studies 27 (1): 46–66. .com /a r t icles / 23741 / w i l l ia m-r-pol k /t he Seddon, D. 1989. “Riot and Rebellion in North -nature-of-modernization. Africa: Political Responses to Economic Crisis P ra s ad , N . 2 014. B roke n R e l a t i o n shi p: in Tunisia, Morocco and Sudan.” In Power Social Policies, the Arab States and the Middle and Stability in the Middle East, edited by Class. Beirut: United Nations Economic and Berch Berberoglu, 114 –35. London: Zed Social Commission for Western Africa. http:// Boock. www.escwa.un.org/information/publications Shryock, A., and S. Howell. 2001. “‘Ever a / edit /upload / E _ ESCWA _ EDGD_14_TP-2 Guest in Our House’: The Emir Abdullah _E .pdf. Shaykh Majid al-Adwan, and the Practice of Provence, M. 2011. “Ottoman Modernity, Jordanian House Politics, as Remembered by Colonialism, and Insurgency in the Interwar Um m S u lt a n , t h e W idow of M ajid .” Arab East.” International Journal of Middle International Journal of Middle East Studies East Studies 43 (2): 205–22. 33 (2): 247–69. Richards, A., and J. Waterbury. 1996. A Political Tibi, B . 1990. “T he Si mu lt a neit y of t he Economy of the Middle East. Boulder, CO: Unsimultaneous: Old Tribes and Imposed Westview Press. Nation-States in the Modern Middle East.” Rijkers, B., C. Freund, and A. Nucifora. 2014. All In Tribes and State Formation in the Middle in the Family: State Capture in Tunisia. East, edited by Philip S. Khoury and Joseph Washington, DC: World Bank. http://www Kostiner, 127–52. Oakland: University of - wd s .w o r l d b a n k . o r g / e x t e r n a l /d e f a u l t California Press. / W DSContentServer/ W DSP Trial, G. T., R. Winder, and R. Bayly. 1950. / IB/2014/03/25/000158349_201403250929 “Modern Education in Saudi Arabia.” History 05/Rendered/PDF/WPS6810.pdf. of Education Journal 1 (3): 121–33. R ivet, D. 1995. “Hygienisme colonial et Waterbury, J. 1970. The Commander of the medicalization de la societe marocaine au Faithful: The Moroccan Political Elite, temps du protectorat francais: 1912–1956.” In A Study in Segmented Politics. New York: Santé medicine et societé dans le monde Cambridge University Press. Arabe, edited by E. Longuenesse, 105–28. Yoav, A. 2007. The Making of Jordan: Tribes, Lyon- Paris: Maison de l’Orient Colonialism and the Modern State. London: Mediterraneen-L’Harmattan. I. B. Tauris. Rivlin, P., and S. Even. 2004. “Political Stability Yousef, T. M. 2004. “Development, Growth and in A rab States: E conom ic Causes and Policy Reform in the Middle East and North Consequences.” Memorandum No. 74, Jaffee Africa since 1950.” Journal of Economic C enter for St rateg ic St udies, Tel Aviv Perspectives 18 (3): 91–116. University. http://www.tau.ac.il/dayancenter Zaid, A., H. Sherry, M. El-Badrawi, and J. Haber. /commentary/memo74.pdf. 2014. “Arab Uprisings and Social Justice: Sadiki, L. 2000. “Popular Uprisings and Arab Implications of IMF Subsidy Reform Policies.” Democratization.” International Journal of New America Foundation, February 27. Middle East Studies 32 (1): 71–95. http://newamerica.net/publications/policy Sakai, K. 2001. “Tribalism and the State: Remarks /arab_uprisings _social_justice. on the Army, Cabinets, and the National Assembly, 1980–1990.” In Tribes and Power: Nationalism and Ethnicity in the Middle East, Data sources edited by Faleh Abdul-Jabar and Hosham E d S t at s , World B a n k , ht t p: //d at atopi c s Dawod, 136–64. London: Saqi. .worldbank.org/education/ Saleh, S. S., M. Alameddine, N. M. Natafgi, Global Health Observatory Data Repository, A. Mataria, J. Nasher, M. Zeiton, S. Ahmad, World Health Organization, http://apps.who and S. Siddiqi. 2014. “The Path towards .int/gho/data/node.main.78?lang=en 122 TRUST, VOICE, AND INCENTIVES Tra nsit iona l G over na nce Projec t , ht t p: // /indicator/NY.GDP.PCAP.PP.CD; http://data transitionalgovernanceproject.org/ .worldbank.org/indicator/SE.XPD.SECO.PC UNESCO Institute of Statistics, http://data.uis .ZS; http://data.worldbank.org/indicator / SE .unesco.org/index.aspx?queryid=190 .XPD.PRIM.PC.ZS World Development Indicators, World Bank, World H e a lt h S t at i s t i c s , World H e a lt h http://data.worldbank.org/indicator/SH.XPD Orga n i zation, ht t p: //w w w.who.i nt /g ho .OOPC.TO.ZS; http://data.worldbank.org /publications/world_health_statistics/en/ Institutions Influencing the Cycle of Performance 5 • In most countries in the Middle East and North Africa, political institutions are highly centralized and not inclusive, and citizens and frontline service providers have little influence on policy formation and implementation. • Political institutions lack accountability mechanisms, with citizens unable to obtain information, to voice demands, or to give policy makers and public servants the incentives through formal channels to improve services. • Administrative institutions suffer from highly centralized bureaucracies and weak management systems, and so they are unable to allocate human and material resources efficiently and manage performance. • Social institutions emphasize obligations to members of social networks over national welfare, resulting in the widespread practice of wasta, a form of clientelism, as well as informal payments in return for services. • Weak political institutions and the lack of accountability contribute to a cycle of poor performance. W hy have decades of capacity build- (World Bank 2004). Indeed, the World Bank ing contributed little to service (2003) has reported that enhancing gover- quality in the Middle East and nance in the MENA region rests on two core North Africa (MENA) region? Why is infor- values: inclusive laws and regulations that mation not properly collected throughout treat all members of society indiscriminately1 the delivery chain and followed up on later? and mechanisms of internal and external For many political economists and country accountability that ensure that those laws experts, the answers to such questions lie in and regulations are respected. When these understanding the incentives that political institutions are weak and accountability is institutions in the MENA countries gener- lacking, providers lack incentives to perform ate. World Development Report 2004: well and citizens are more likely to suffer Making Services Work for Poor People poor-quality services—that is, the cycle of made the compelling argument that better performance for service delivery is stuck in a public services depend on better governance low-level equilibrium. 123 124 TRUST, VOICE, AND INCENTIVES This chapter explores the institutions that that create clear expectations of providers’ shape service provision in the MENA region, efforts and sanctions for those who fall focusing on three interrelated arenas: politi- short. National-level institutions also cal, administrative, and social (see figure 5.1).2 affect the clinic’s services by means of trans- Institutions include both the formal (legislated parent, responsive policy making, ratio- rules) and informal (norms) rules and proce- nalized budgetary support, and effective dures that outline the responsibilities of enforcement of rules and mechanisms play- actors, the costs and benefits associated with ing important roles. alternative choices, and the legitimacy of These institutions shape accountability actions. Political institutions are the norms mechanisms, helping to create incentives for and rules in the state sphere; social institu- actors across the service delivery chain to tions encompass the formal institutions accept and respond to their responsibilities. within society (such as tribal councils and Accountability generally requires that those formal religious institutions) as well as social who hold providers and policy makers norms; and administrative institutions are the responsible are informed about providers’ rules and regulations that structure bureau- and policy makers’ roles and responsibilities cratic processes. as well as performance (transparency), and Institutions exist and influence actors at they are able to punish low performers different levels, from national policy makers (enforcement). These mechanisms can be to frontline service providers. In a public based on political institutions (such as infor- health clinic, for example, providers’ efforts mation acts, elections), social institutions are shaped by the nature and strength of the (such as norms of transparency, social sanc- administrative and social institutions within tions), or administrative institutions (such as the clinic. Examples of factors that affect ser- clear information-sharing procedures, effec- vices are whether there is a well-implemented tive enforcement). Institutions may be less human resources policy that sets clear per- directly related to soft accountability—the formance standards, calls for annual reviews, internalized norms of responsibility. And yet ensures that results are transparently even here institutions may play a role in help- recorded and reviewed with the clinician, ing form professional associations that foster and lists sanctions for those who fail to per- professional norms, or they may reduce cor- form, and whether there are social norms ruption and create norms of responsibility. FIGURE 5.1 Political, administrative, and social institutions affecting service delivery Political institutions Administrative institutions Social institutions External accountability Internal accountability Social identity, mechanisms mechanisms values, and norms INSTITUTIONS INFLUENCING THE CYCLE OF PERFORMANCE 125 Political, administrative, and social institu- effect on service delivery. Unfortunately, tions are interrelated. As shown in chapter 4, as discussed in chapter 4, most MENA strong social institutions—that is, the social countries have developed regimes in which norms and tribal, sectarian, and other institu- state institutions are politically captured, tions that reinforce identity groups—have highly centralized, and weak. State institu- affected the development of political institu- tions thus limit the ability of citizens and tions. Indeed, at times they have even incor- political actors to demand change, implement porated tribal or sectarian norms and policies, or hold those in the service delivery mechanisms formally into political systems chain accountable. (Shryock and Howell 2001; Weir 2006; Maktabi 2013). State institutions also can Influence of authoritarian regimes create or reify identity categories and social groups, thereby influencing social divisions Authoritarian regimes dominate the MENA (Lieberman and Singh 2009; Lieberman and region. Polity measures find that, as a region, Singh n.d.). Similarly, captured political the MENA countries have the least con- regimes can undermine the development of strained executives, the weakest parliaments, rational administrative systems. and among the least independent judiciaries This chapter examines institutions in the in the world (see figures 5.2, 5.3, and 5.4).3 MENA region with an eye toward their Certainly, one can debate whether authori- impact within the cycle of poor performance tarianism always undermines development on accountability and, subsequently, perfor- and the provision of services,4 but at least in mance in service delivery. The fi rst section the MENA region it has skewed the distribu- examines how political institutions hinder tion of resources away from those in need policy development and implementation. The and has led to a leakage of resources. As dis- second looks at how administrative institu- cussed in chapter 4, both monarchs and tions in the public sector hamper account- presidents have increasingly relied on their ability mechanisms in the compact among ability to reward key constituencies—such as policy makers, administrators, and service the military, the intelligence services providers by, for example, limiting the flow (mukhabarat), key party loyalists, and tribal of information and possessing limited author- leaders—in order to stay in power. But this is ity to impose sanctions or reward good effort. not unique to the MENA region; elsewhere, The third section then explores how social authoritarian regimes also maintain power institutions at both the national and local by granting privileged access to services levels shape service delivery. Unfortunately, based on political connections and loyalty in much of the region today, interlinked polit- (Olivier de Sardan 1999; Desai, Olofsgard, ical, administrative, and social institutions and Yousef 2009). In the MENA countries, across the service delivery chain impede the this practice has been reflected at times in the provision of education, health, and other establishment of institutions (for example, services. military health systems, elite educational facilities) that funnel a disproportionate amount of resources to key constituents and Political institutions provide them with better services. In general, Political systems affect service delivery. The the success of long-standing authoritarian 2004 World Development Report character- regimes in the region has relied on and con- ized the “long route of accountability” as that tributed to weak accountability. by which citizens hold providers accountable Political capture extends across the service indirectly through exercising pressure on pol- delivery chain. As demonstrated by the iticians (World Bank 2004). Political actors Global Integrity Index (2009), the MENA also can engage directly in administrative region lags behind other regions in terms of reform, initiating changes that have a direct civil servant appointments and evaluations 126 TRUST, VOICE, AND INCENTIVES FIGURE 5.2 Institutional constraints on the decision-making power of the executive: All regions, 2014 7 6 5 4 Score 3 2 1 0 MENA SSA SEA EA World SA EE AU LA WE-NA Source: Polity IV Annual Time-Series 1800–2013. Note: Score: 1 (unlimited executive authority) to 7 (executive parity or subordination). AU = Australia and Pacific; EA = East Asia; EE = Eastern Europe; LA = Latin America; MENA = Middle East and North Africa; SA = South Asia; SEA = Southeast Asia; SSA = Sub-Saharan Africa; WE-NA = Western Europe and North America. FIGURE 5.3 Degree of legislative autonomy: All regions, 2014 9 8 7 6 5 Score 4 3 2 1 0 MENA SEA SSA SA EA World LA EE AU WE-NA Legislature’s influence over the executive Legislature’s institutional autonomy Legislature’s specified powers Source: Fish and Kroenig 2009. Note: Legislative influence, autonomy, and power are measured as the number of powers of the legislature over the executive, the extent of institutional autonomy of the legislature vis-à-vis the executive, and the number of specified powers held by the legislature. Score: 0 (least powerful) to 9 (most powerful). AU = Australia and Pacific; EA = East Asia; EE = Eastern Europe; LA = Latin America; MENA = Middle East and North Africa; SA = South Asia; SEA = Southeast Asia; SSA = Sub-Saharan Africa; WE-NA = Western Europe and North America. based on professional criteria, the actions that they distort the incentives that public (hiring, fi ring, etc.) of civil service manage- sector employees have to work efficiently and ment, and publication of the number of honestly and impair the ability of citizens and authorized civil service positions along management to hold them accountable. with the number of positions actually fi lled Political capture is particularly pro- (figure 5.5). The result of weak procedures is blematic because the MENA regimes are INSTITUTIONS INFLUENCING THE CYCLE OF PERFORMANCE 127 FIGURE 5.4 Regional averages of judicial independence: All regions, 2014 7 6 5 4 Score 3 2 1 0 an sia a ia sia ca t vie ric As be fri tA hA Af So nA st ib as ut rth Ea st- ar he ra So eC Po No ha ut th d So Sa nd an d b- ta an pe Su as ica ro eE Eu er dl Am rn id ste M tin Ea La Source: Bertelsmann Transformation Index, 2014. Note: The y-axis reflects an expert-coded index on a 7-point scale, regarding the question, “To what extent does an independent judiciary exist?” An independent judiciary is defined as one that has the ability and autonomy to interpret and review existing laws, legislation, and policies, both public and civil; pursue its own reasoning, free from the influence of political decision makers or powerful groups and individuals and from corruption; and develop a differentiated organization, including legal education, jurisprudence, regulated appointment of the judiciary, rational proceedings, professionalism, channels of appeal, and court administration. highly centralized. As a 2007 World Bank zoning, and other local issues, but they have report noted, “With few exceptions, MENA no influence over education, health, or other countries still feature remnants of their inher- services. ited colonial past where the sharing of power between the central and local levels of gov- Lack of information and transparency ernment is still heavily skewed toward the former” (World Bank 2007, 1). Directly Lack of information further undermines elected local governments are found in only a accountability. Freedom of information small minority of countries in the region, and and public disclosure laws and practices where they do exist, councils have limited that would allow citizens and intermediar- budgets and responsibilities. With the excep- ies to monitor government activities and tion of the Arab Republic of Egypt, Morocco, pressure state actors for better performance and the West Bank and Gaza, local council are scarce. In some countries, Access to budgets account for less than 5 percent of Information (ATI) laws are on the books, but total public expenditures, far behind the their provisions are vague. Limited institu- world average of 38 percent for federal sys- tional capacity and resources constrain their tems and 22 percent for unitary ones (World implementation, and governments prohibit Bank 2007, iii). Most local councils have the public disclosure of a wide range of infor- some control over solid waste management, mation. For example, Jordan passed an ATI 128 TRUST, VOICE, AND INCENTIVES FIGURE 5.5 Performance of MENA region compared with other regions in effectiveness of laws governing civil service administration, 2009 80 70 60 50 Score 40 30 20 10 0 In practice, civil servants are In practice, civil service In practice, the government appointed and evaluated management actions (such as publishes the number of according to professional hiring, firing, promotions) are authorized civil service positions criteria not based on nepotism, along with the number of cronyism, or patronage positions actually filled MENA E AP ECA LAC SAR SSA Source: Global Integrity Index, 2009. Note: Score: very strong (90–100); strong (80–89); moderate (70–79); weak (60–69); very weak (below 60). EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin America and the Caribbean; MENA = Middle East and North Africa; SAR = South Asia; SSA = Sub-Saharan Africa. law in 2007, but the information council information (Almadhoun 2012). Civil ser- responsible for the implementation and vants who provide the general public with promotion of access to information never information can be sanctioned under vague received funding from the state budget; the legislation governing the treatment of official scope of information designated as “classi- data and information, as in Egypt. Further fied” remained broad; the personal data undermining the use of information are legal required to request information appeared provisions on libel and slander that too often excessive; and a large swath of the popula- are used against citizens who criticize gov- tion, including officers in 50 percent of ernment officials. Such treatment is a strong Ministries and 40 percent of journalists, had disincentive to filing complaints related to the no knowledge of the law as late as 2010. 5 poor delivery of services. Lebanon, Morocco, and Qatar lack the rele- Indeed, the MENA region lags notably vant legislation. In Lebanon, a law proposed behind the rest of the world in transparency. in 2009 remains stalled. In Morocco’s 2011 On the 2009 Global Integrity Index, the constitution, Article 27 states that citizens region ranked among the lowest on the public cannot be denied access to public informa- access to information indicator (figure 5.6), tion except when access threatens national the legal right to access information, and security or private life, but no legislation fur- whether the right of access to information is ther defines or guarantees public access to effective. So, too, the Open Budget Index INSTITUTIONS INFLUENCING THE CYCLE OF PERFORMANCE 129 FIGURE 5.6 Public access to information: All regions, 2009 90 80 70 60 50 Score 40 30 20 10 0 SSA MENA EAP ECA SAR LAC Source: Global Integrity Index, 2009. Note: Score: very strong (90–100); strong (80–89); moderate (70–79); weak (60–69); very weak (below 60). EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin America and the Caribbean; MENA = Middle East and North Africa; SAR = South Asia; SSA = Sub-Saharan Africa. ranks the MENA region acutely low on par- and penetration rates, whereas the other ticipatory budget indicators and does not MENA countries are struggling with build- rank the region at all on the citizens’ budget ing these connections. The challenges to indicator, which reflects the existence of a digital development vary across the region. publicly available, simplified budget docu- Lebanon, for example, has the human capi- ment that uses nontechnical language and tal and capacity to implement digitized gov- accessible formulas to facilitate citizens’ ernment services, but it lacks the fi nancial understanding and engagement (figure 5.7). capability and political stability needed to The Open Data Barometer (2013) found that adopt and implement digitization initiatives. most MENA countries ranked below the Elsewhere, human capital is in short supply. world average on their readiness to imple- The MENA countries also face issues such ment reforms (figure 5.8) and give citizens the as the human and resource capacity of their rights and freedoms to use data to hold gov- Ministries, their information and communi- ernments accountable. It also found an cations technology infrastructure, data absence of strong rights to information for protection, consumer rights, and informa- citizens (figure 5.9). Meanwhile, the region tion technology security standards in order also falls behind the world average on the to build greater trust between users and availability of data on the performance of service providers.6 health and primary and secondary education The MENA countries thus vary signifi- services (Davies 2013). cantly in their e-governance readiness The MENA governments have made pos- and implementation. The United Nations’ itive strides toward recognizing that connec- e-government development index (EGDI) is a tivity and digital development can facilitate composite measure of three important dimen- faster and more direct access to services sions of e-government: provision of online and information, but their levels of digital services, telecommunication connectivity, development vary widely across and within and human capacity. According to the EGDI, MENA countries. High-income countries Bahrain is among the top 25 EGDI world have achieved high levels of connectivity leaders, ranking 18th out of 144. Bahrain is 130 TRUST, VOICE, AND INCENTIVES FIGURE 5.7 Open budget index: All regions, 2013 60 50 40 Score 30 20 10 0 MENA SSA EAP LAC ECA SAR Source: Open Budget Index, 2013. Note: EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin America and the Caribbean; MENA = Middle East and North Africa; SAR = South Asia; SSA = Sub-Saharan Africa. FIGURE 5.8 Readiness subindex: Selected MENA countries and world average, 2013 60 50 40 Score 30 20 10 0 ira rab ain a co r an . ge ep ta bi oc rd ra Qa hr ,R ra s Em d A te e Jo or iA Ba en av ite M ud m ld Un Ye Sa or W Source: Open Data Barometer, 2013. Note: The value for the Republic of Yemen is zero. The readiness subindex has three main components: (1) government capacity and the presence of government commitments to open data; (2) citizen and civil society freedoms and engagement with the open data agenda; and (3) available resources for entrepreneurs and businesses to support economic reuse of open data and to catalyze intermediary actions (Davies 2013). followed by other member countries of Morocco are 75th, 80th, and 82nd, respec- the Gulf Cooperation Council (GCC): the tively, and toward the bottom are Libya, United Arab Emirates, 32nd; Saudi Arabia, ranked 121st, and Algeria, 136th. All coun- 36th; Qatar, 44th; Oman, 48th; and Kuwait, tries in the region, with the exception of the 49th. The other MENA countries are ranked United Arab Emirates and Algeria, have closer to the center: Tunisia, Egypt, and moved up in the rankings, demonstrating INSTITUTIONS INFLUENCING THE CYCLE OF PERFORMANCE 131 FIGURE 5.9 Right to information law index: All regions, 2013 70 60 50 40 Score 30 20 10 0 Europe Americas Asia-Pacific Africa Middle East and Central Asia Source: Open Data Barometer, 2013. their commitment to improving e-government stakeholders in the labor market. Another services across the region. In other assess- example of digitalizing government services ments of digitized government7 services, the in the region is Morocco’s development of an United Arab Emirates ranked third and e-consultation platform through the website Saudi Arabia ranked fifth among the top 10 of the Secrétariat Général du Gouvernement. leading countries providing digitized services. Citizens can access legislative texts online, The high ranking of the United Arab read and download them, and post their Emirates is attributed to initiatives such as comments and concerns. the Emirates Government Service Excellence Although some e-government services are Program, which was established to standard- provided in the MENA region, the lack of ize service quality, ensure consistency in information needed to assess and monitor customer experience across service centers, local service provision remains a problem, and promote cost efficiency. Based on citizen hindering the ability of civil society organiza- satisfaction surveys conducted by Accenture tions and the media to play watchdog and (2014), the United Arab Emirates also ranked advocacy roles. Indeed, in the MENA region, as the top leader in citizen satisfaction civil society organizations and the media do and engagement. The citizens of the United not enjoy the freedom and the ability to exer- Arab Emirates are aware of the importance cise real pressure on the state. Citizens’ rights of digital channels in improving service to protest and demonstrate, and to voice quality and outcomes, and, as a result, they demands through the media, are restricted by have become confident of their government’s law and in practice. Despite the exponential ability to meet the challenges that may arise increase in civil society organizations in the in the process. In Saudi Arabia, a number Middle East following liberalization over the of public services are available through last two decades, the ability of associations to digital channels. For example, the Ministry mobilize and voice critical demands remains of L abor and the Hu man Resou rces limited. They continue to operate under Development Fund developed a Virtual highly restrictive legal frameworks that allow Labor Market ecosystem that serves all considerable scope for government entities to 132 TRUST, VOICE, AND INCENTIVES interfere in their activities. Egypt is an exam- complaints for both service users and service ple of severe restrictions, and even in Qatar, providers. A household survey conducted in which enjoys a particularly positive rating on Jordan in 2012, the Statistical Survey on the corruption indexes, Law 12 (2004) prohibits Volume of Demand of Legal Aid Services, civil society organizations from focusing on found that 6 percent of civil legal problems corruption (Business Anti-Corruption Portal involve access to health care, primarily 2014). There are important exceptions, of related to medical negligence. Those respon- course, such as Al-Bawsala in Tunisia and the dents in the lowest two expenditure quar- Leaders of Tomorrow in Jordan, but, in gen- tiles, which represent the poor and much of eral, civil society is weak and disconnected. the near-poor, were more likely to report As Jordan’s director of civil society organiza- legal problems related to health care. tions at the Ministry of Political Development Citizens do file complaints in attempts explained in 2010, “apart from having some to hold service providers accountable. For scattered operations and state sponsored example, in Jordan in 2010, the Ombudsmen clubs, they pretty much have nothing else.”8 Bureau reported that the Ministry of Education received the second highest number of complaints for maladministration Lack of external accountability of any public sector entity, just behind the mechanisms Civil Service Bureau. The Ministry of Health Justice sector services, when implemented received the sixth highest number of com- properly, provide accountability tools for citi- plaints. More than 44 percent of complaints zens to hold service providers accountable for about the Ministry of Education were related poor or undelivered services. These tools can to the hiring, promotion, and assignment of be used in three primary ways: (1) for citizens teachers and administrative staff. In 2010 the fi ling complaints for poor or nondelivery of Ministry of Education also received 31 com- services; (2) for individual or group com- plaints, including through its complaints plaints from service providers (such as doc- hotline, related to the physical and verbal tors, nurses, or teachers) about the poor abuse of students. However, according to the working conditions under which services are National Centre for Human Rights, no effec- provided; and (3) for public interest litigation tive actions were taken against the education to force governments to enforce constitutional staff alleged to have committed abuses. The and other rights to education and health care. National Centre for Human Rights received Public interest litigation—bringing cases only 41 complaints in 2010 and 44 in 2009 against government that affect a high number related to access to health care services—a of beneficiaries “for the public good”—could relatively small number in view of the magni- be used to force governments to improve ser- tude of health care services. Meanwhile, vice delivery frameworks when constitutional complainants are subjected to some risk. In rights are involved. But such actions in the 2010, 38 teachers protesting for better work- MENA region remain rare because citizens ing conditions were temporarily suspended have not yet benefited from litigation, unlike by the Ministry of Education. those of other countries such as India and Executive authorities also have a tradi- South Africa (Brinks and Gauri 2014). Indeed, tion of appointing politically dependent in the MENA countries, the justice system judges and continually meddling in the juris- fares poorly in terms of cost, ease of access, dictions of courts (World Bank 2003). As a speed proceedings, and corruption. result, the judicial branch is an instrument Comprehensive data on complaints from the executive can use to legitimize its politi- citizens about the delivery of education and cal ambitions (Jabbour and Yamout 2012). health services are scarce, but the limited Independent audit agencies, inspectors- official data and anecdotal evidence suggest general, anticorruption commissions, and that both sectors are considerable sources of ombudsmen lack sufficient resources, INSTITUTIONS INFLUENCING THE CYCLE OF PERFORMANCE 133 authority, and autonomy, and their assess- Authority, Administrative Prosecution ments are not always disclosed to the public Authority, and Illegal Profiting Apparatus or followed up by the state (World Bank (Business Anti-Corruption Portal 2014). 2003). Even when decisions are issued in However, they are all closely linked to the favor of citizens, they remain difficult to president, prime minister, or minister of justice. enforce. Thus external audits emerge as inef- Furthermore, their annual reports are not fective and incapable of detecting corruption made public; they are presented only to the practices, further contributing to weak pol- president, minister of justice, or minister of icy implementation. interior (Business Anti-Corruption Portal An ombudsman office, waseet eljumhuria, 2014). In Jordan, the law stipulates that the is normally set up by the state to investigate Anti- Corruption Commission must be individual citizens’ complaints of maladmin- financially and administratively independent istration, especially that of public authorities. of executive interference. Nevertheless, Global The offices established in Jordan, Lebanon, Integrity (2011) has reported that the Morocco, and Tunisia illustrate the lack of commission is not entirely free from political sturdiness of such offices in the context of interference and that appointments can be weak, politically captured institutions. In based on loyalties, nepotism, and favoritism. Jordan, the ombudsman was established to The poor, who are perhaps the most examine complaints from individuals relating dependent on state-run education and health to any decree, procedure, practice, or act of services, face considerable obstacles in refusal by the public administration and to accessing justice institutions because of a help citizens access information withheld by combination of lack of fi nancial resources government agencies (Almadhoun 2012). But and poorly targeted services intended to the ombudsman’s bureau has been annulled benefit them. Not all MENA governments and merged with the Anti- Corruption provide legal aid services to the poor for civil Commission (ACC), and its staff has been cases, which include education and health transferred to the ACC after little more than care. For those that do, services exist pri- four years in existence, according to an arti- marily on paper but are rarely implemented cle in the June 2013 Jordan Times (Abu in practice (Prettitore 2012). In countries Nimah 2013). In Lebanon, evidence on the such as Jordan, where court fees are rela- ground points to the limited effectiveness of tively high, the process of waiving fees based the ombudsman, bound by vague legal on poverty involves vague criteria on estab- frameworks. For example, Law 664 (2005), lishing poverty status and cumbersome which established the office, is still lacking procedures that involve gathering multiple the implementing ordinances (Almadhoun documents from various government enti- 2012). In Morocco, the Office of the ties. The Statistical Survey on the Volume of Ombudsman is relatively effective and Demand for Legal Aid Services conducted releases its information publicly. However, it by the Department of Statistics and the still lacks the authority to initiate investiga- Justice Center for Legal Aid in Jordan in tions and impose penalties (Global Integrity 2012 found that over 90 percent of respon- 2010). In Tunisia, the National Ombudsman dents were unaware of available legal aid Service was loyal to the ruling party under services (Prettitore 2013). Administrative the regime of President Zine El Abidine Ben courts are often present only in the capi- Ali. The new ombudsman is emphasizing the tal city, requiring those in other areas to principle of establishing a dialogue between travel long distances with the associated the public administration and society. costs for travel. They also often have The agencies designed to address corrup- higher fees and require specialized legal rep- tion are rarely independent. Egypt has four resentation, thereby prompting citizens to anticorruption agencies: the Transparency and attempt instead to resolve problems through Integrity Committee, Administrative Control wasta, a form of clientelism described later 134 TRUST, VOICE, AND INCENTIVES in this chapter. Other vulnerable groups support systems. For example, although some such as refugees often face additional legal parliaments in the MENA region have the and practical obstacles that are not fully authority to approve the state budget, they add re s s e d by t he ex i s t i ng s er v ic e s . must review it over a short period, and it is Comprehensive data on the extent of legal often drafted in vague language and requires aid and the court deferment services pro- intensive work to ascertain its details. vided through the courts are lacking. Parliamentarians therefore have only limited Anecdotal evidence suggests that the provi- influence over the outcome (World Bank sion of services is not widespread, at least in 2003). The external audit process, carried relation to demand (Prettitore 2012). out to provide a parliament with reassur- Finally, legislatures in the MENA region ances on the adherence to fi nancial laws, is are heavily constrained and thus limited in often weak and ineffective as well, in part their abilities to initiate reform (Fish and reflecting the restricted role that parliaments Kroenig 2009). At the national level, parlia- play in providing oversight. Because of the ments have little legislative power—many are ineffective checks and controls at every step unable to draft legislation and are easily dis- of the budget cycle,9 parliaments are unable missed by more powerful executives (see to detect corrupt practices associated with figure 5.10). Parliaments also experience fre- government expenditure, further contribut- quent turnover and have weak legislative ing to weak policy implementation. FIGURE 5.10 Legislative influence, autonomy, and power: MENA region, 2009 9 8 7 6 5 Score 4 3 2 1 0 ain Ba s ia ya an ria t, A isia p. an o r ic ait . n q te ep ta cc no Ira bl ab Re ira hr Lib Om rd ge w Qa ,R n o pu ba Ar Ku Tu Jo or Em b Al en Re ra Le M i ud m b ab a Ye Sa Ar yp Ar d Eg n ite ria Un Sy Legislature’s influence over the executive Legislature’s institutional autonomy Legislature’s specified powers Source: Fish and Kroenig 2009. Note: Legislative influence, autonomy, and power are measured as the number of powers of the legislature over the executive, the extent of institutional autonomy of the legislature vis-à-vis the executive, and the number of specified powers held by the legislature. Score: 0 (least powerful) to 9 (most powerful). INSTITUTIONS INFLUENCING THE CYCLE OF PERFORMANCE 135 The role that parliaments, or more pre- In short, the M ENA region suffers cisely parliamentarians, play in providing ser- from weak, captured political institutions that vices is one that reflects the weakness of undermine voice and accountability. Indeed, political institutions, not the importance of the World Bank’s Voice and Accountability parliament. Because of the state’s opacity and measures in its 2012 Worldwide Governance lack of accountability, citizens seek interme- Indicators (figure 5.11) continue to find diaries who can help them access services. the MENA region the most restrictive in the Parliamentarians use their positions to pres- world. On a scale ranging from –2.5 (most sure ministers and bureaucrats to dispense restrictive) to 2.5 (most permissive), no jobs, licenses, and other state resources to MENA country received a positive rating, their constituents, becoming service providers and the region had the lowest average rating. to certain groups rather than providing legis- There is some variation across the region, as lative or executive oversight. They may not be demonstrated in figure 5.10, and certainly able to legislate more effective public services Tunisia (after 2011) stands as an exception. In administration, but parliamentarians can use general, however, citizens in the MENA find their position to help provide such services.10 it difficult to hold policy makers and service In Jordan, for example, many citizens call providers accountable. members of parliament na’ib khadmat, or service parliamentarians, “signaling the per- ception that their primary role is to help their Administrative institutions constituents obtain employment, access to Weak, politically captured regimes are health care, education, and other services, or coupled with—and compound—ineffective necessary permits and licenses from the gov- administrative systems and accountability ernment bureaucracy” (Lust, Hourani, and mechanisms. Accountability mechanisms Al-Momani 2011). within bureaucracies, the command and con- Candidates typically compete in terms of trol elements of public administration, encom- their ability to offer patronage and public pass Weberian notions of hierarchy, rules, and employment to their supporters or to inter- regulations, in addition to market-driven con- vene in the bureaucracy to solve individual cepts and practices such as personnel manage- constituency problems or deliver club ment, performance evaluation, auditing, and goods rather than in their ability to deliver monitoring (Blind 2011). In the MENA public goods and services more efficiently region, the highly centralized system just and effectively to the citizenry at large (Sakai described leaves both administrators and pro- 2001; Lust 2009; Cammett 2009; Kao 2012). viders with little autonomy to enforce rules, As Corstange (2011) has argued about manage human and financial resources, or Lebanon and the Republic of Yemen, “politi- develop solutions. Moreover, political capture cians spend most of their time jockeying on has promoted a system in which there is a behalf of constituencies based on sect, tribe, general lack of clear information on perfor- extended family, and region over who gets mance that can be used for monitoring and hired into the civil service, where the roads evaluation, few consequences when violations get paved, and who keeps their electricity occur, and a general absence of incentives for longest.” In doing so, they not only under- providers and administrators. mine the role that parliaments could play in developing policies that ensure better trans- Lack of coordination and insufficient parency, oversight, and enforcement of capacity for performance management service provision, but also reinforce social institutions that, at least in the MENA con- Within the public sector, policy and cabinet text of clientelism, undermine development coordinating mechanisms are generally weak. of strong administrative and political institu- In many countries, Ministries at the central tions and equal access to services. level are responsible for formulating and 136 TRUST, VOICE, AND INCENTIVES FIGURE 5.11 Index of extent to which citizens can participate in government selection and freedoms of expression, association, and media: MENA and other regions and selected MENA economies, 2012 a. Voice and accountability, MENA and other regions 2.5 2.0 1.5 1.0 Score 0.5 0 –0.5 –1.0 –1.5 –2.0 –2.5 MENA SSA SEA SA EA EE World LA AU WE-NA b. Voice and accountability, selected MENA economies 2.5 2.0 1.5 1.0 0.5 Score 0 –0.5 –1.0 –1.5 –2.0 –2.5 p. Ba . za ic ep Re a ria s n ain ya an sia an q r co ait te i Re no ta bl ra rab Ga Ira ,R ge Lib rd ira Om oc ni hr Qa w pu ba b en Jo Tu nA iA Al nd Ku or Em ra Le m t, A M ud ka b ab Ye Sa an yp Ar tB Eg d es ria ite W Sy Un Source: Worldwide Governance Indicators, 2012. Note: Score: –2.5 (most restrictive) to 2.5 (most permissive). AU = Australia and Pacific; EA = East Asia; EE = Eastern Europe; LA = Latin America; MENA = Middle East and North Africa; SA = South Asia; SEA = Southeast Asia; SSA = Sub-Saharan Africa; WE-NA = Western Europe and North America. implementing national policies and monitor- departments often struggle. For example, ing and evaluation. However, this distribu- El-Jardali et al. (2012) found, using a purpo- tion of operational activities often comes at sive sample across 11 eastern Mediterranean the expense of coordinated strategic actions countries, that the majority of policy makers and vision. Initiatives that require developing cited lack of coordination among government common policies and shared vision and coor- entities and lack of coordination between dination across Ministries, agencies, and government and service providers as a INSTITUTIONS INFLUENCING THE CYCLE OF PERFORMANCE 137 primary hindrance to the use of evidence in A World Bank review of performance health policy making. management policies along the entire employ- The lack of coordination often results in ment cycle of a civil servant at Tunisian’s the overlap of responsibilities for policy imple- Ministry of Industry reveals that the coun- mentation across the Ministries of Health, try’s legal framework for public sector per- Education, Finance, and Planning (Jabbour formance management is in principle well and Yamout 2012). For example, poorly func- designed and set up for recruiting the most tioning communication lines between the qualified candidates and promoting high- Ministries of Education and Health might performing employees. However, in reality result in the latter building medical and nurs- the system suffers from structural weak- ing schools without any needs assessment. nesses on four main fronts: (1) recruitment— The lines of communication between the although competitive recruitment remains Ministry of Health and the Ministry of the main recruiting method, the significant Finance are not straightforward and may not recent increase in other methods such as connect the budget allocation process with direct recruitments and regularizations health care priorities. In many countries, the indicates a deterioration in the quality of final decision-making authority on health lies competencies recruited and in the supervi- not within the Ministry of Health but within sion rate; (2) evaluation—the evaluation the Ministries of Finance and Planning, method, which consists of an annual profes- adding another layer to the decision-making sional rating and a quarterly performance process. In addition, many Ministries struggle rating, suffers from lack of transparency with the legal aspects of hiring, such as the and objectivity and from grade inflation; scope of practice, job description, perfor- (3) compensation—the system remains com- mance evaluation, and licensing, without a plex, and the link between performance rat- clear understanding of the criteria used. ing and compensation remains weak, Services also suffer from the lack of coor- rendering professional evaluation criteria dination within Ministries. For example, irrelevant; and (4) promotions— application- in Morocco the lack of integration and com- based promotions are subjective and empha- munication within the Ministry of Health size seniority over performance (Brockmeyer, and across different levels of care has led to a Khatrouch, and Raballand 2014). structural resource misallocation in the health Although such challenges have been iden- system, contributing to the overuse of hospi- tified in past reforms, attempts to address talization and emergency services. According them have been largely unsuccessful, requir- to a 2009 survey in Morocco (World Bank ing a better understanding of the broader 2013), 54 percent of patients were referred to political economy environment governing a hospital for the same care they could have reforms. received at an outpatient clinic at a lower cost, resulting in a loss of efficiency. Centralization of service delivery Moreover, the system for policy imple- mentation and service delivery has an insuf- According to Tosun and Yilmaz (2008, 8), ficient capacity for performance management. “In all [MENA] countries, the deconcen- With some exceptions (such as the United trated units of the central government Arab Emirates), government agencies do not provide a big chunk of public services, regularly produce clear performance criteria including education and health, under strict against which their efforts could be indepen- guidance of the central government.” As dently monitored and benchmarked. Nor do shown in tables 5.1 and 5.2, with a few government offices at the cabinet level typi- exceptions such as Lebanon, the Ministry of cally monitor performance data on a regular Education assumes all the key responsibili- basis. The same is true of Ministries of Public ties: policy making, fi nancing, and service Administration and Administrative Reform. delivery. The health system in the MENA Table 5.1 Division of responsibility among central ministries, provincial and regional administrations, and schools on policy setting, planning, finance, human resource management, and pedagogy: Selected MENA countries, 2014 Egypt, Iran, Islamic Algeriaa Djibouti Arab Rep. Rep.b Jordan Lebanon Morocco Tunisia Yemen, Rep. Policy National strategy CMs CMs CMs CMs CMs CMs CMs CMs CMs Action plan CMs CMs CMs CMs CMs CMs CMs CMs CMs Planning Creation and closure of primary CMs, PRAs — CMs, PRAs CMs — CMs CMs, PRAs CMs, PRAs CMs, PRAs schools Establishment of input and CMs CMs CMs CMs — CMs CMs CMs CMs infrastructure norms Finance Resource allocation CMs CMs CMs, PRAsc CMs CMs CMs, PRAsd CMs CMs Human resource management Selection of primary and PRAse CMs CMs Schools CMs CMs CMs CMs CMs, PRAs secondary school directors Recruitment of teachers CMs — CMs PRAs CMs CMs CMs CMs CMs, PRAs, schoolsf Management of in- and CMs, PRAs CMs CMs PRAs — — CMs CMs, PRAsg CMs, PRAs preservice training Establishment of teacher CMs, PRAs — CMs PRAs CMs, schools CMs CMs — CMs responsibility Supervision of teachers PRAs, schools PRAs PRAs Schools CMs, PRAs CMs, schools PRAs PRAs CMs, PRAs Pedagogy Definition of curriculum and CMs CMs CMs CMs h CMs CMs CMs CMs CMs textbook context Setting standards and exam CMs PRAs PRAs CMs CMs CMs CMs, PRAs CMs CMs, PRAs management Source: Framework adapted from World Bank (2008). Note: Information was updated to 2014 when possible. CMs = central Ministries; MENA = Middle East and North Africa; PRAs = provincial and regional administrations; — = not available. a. Information dates back to 2005. b. Information dates back to 2005. c. In the Islamic Republic of Iran, resources are determined by the central Ministry, but the allocation to schools is determined at the regional level. d. In Morocco, regional levels (academies) have a say in some resource allocations, including school improvement plans. e. In Algeria, primary school directors are appointed at the regional level, but secondary school directors are appointed by the Ministry of Education. f. In the Republic of Yemen, according to the SABER Teacher Policies (http://saber.worldbank.org/index.cfm?indx=8&tb=1), statutory law suggests that schools are entitled to hire and fire teachers, and yet this is rarely practiced. g. In Tunisia, preservice training for teachers is administered by the Ministry of Education, but in-service training is administered regionally. 138 h. In the Islamic Republic of Iran, the curriculum is determined by the Ministry of Education, but the curriculum for preprimary education is determined regionally. INSTITUTIONS INFLUENCING THE CYCLE OF PERFORMANCE 139 Table 5.2 Division of responsibility among the central ministries, provincial and regional administrations, and health facilities on policy setting, planning, finance, and human resource management: Selected MENA countries, 2014 Egypt, Arab Djibouti Rep. Jordan Lebanona Morocco Tunisia Yemen, Rep. Policy National strategy CMs CMs CMs CMs CMs CMs CMs Action plan CMs CMs CMs CMs CMs CMs CMs Planning Creation and closure of health CMs CMs CMs CMs CMs CMs CMs facilities Establishment of input and CMs CMs CMs CMs CMs CMs, PRAs CMs infrastructure norms Finance Resource allocation CMs CMs, PRAsb CMs HFs CMs, PRAs CMs CMs Human resource management Selection of health facilities’ CMs PRAs CMs, PRAs HFsc CMs CMs PRAs directors Recruitment of health personnel CMs CMs CMs HFs CMs CMs, PRAs PRAs Firing of staff CMs CMs CMs HFs CMs CMs, PRAs PRAs Management of staff training and CMs CMs, PRAs CMs, PRAs HFs CMs CMs CMs, PRAs professional development Standard setting and performance CMs CMs CMs HFs CMs — CMs criteria Supervision of staff CMs CMs, PRAsd CMs, PRAs HFs PRAs CMs, PRAs CMs, PRAs Source: Framework adapted from World Bank (2008). Note: Information was updated to 2014 when possible. CMs = central Ministries; MENA = Middle East and North Africa; PRAs = provincial and regional administrations; HFs = health facilities; — = not available. a. In Lebanon, the 1996 Law No. 544/96, Public Hospital Autonomy, was aimed at enhancing the performance of public hospitals by granting them financial and managerial autonomy and the flexibility for better procurement and recruitment of qualified personnel, thereby also giving them an opportunity to be competitive with the private sector. Currently, appointments of administration board members for public hospitals are issued by government decrees. The appointed board is autonomous, meaning the financial risk is shifted from the central administration down to the level of hospital management, with financial autonomy (control over budget) and administrative autonomy (control over staff management and hiring and firing decisions) at the hospital level, with the exception of one hospital (Ammar 2009). b. In Egypt, most resources are allocated at the regional level, and yet a considerable amount is allocated centrally through the Ministry of Health such as for capital expenditures. c. Although the 1996 decree states that a hospital manager is to be appointed by the hospital board, “the practice has ranged from the board making recommendations that the MOH may or may not accept, to a local political leader submitting one name to the Minister of Health, who then recommends the appointment without consulting with Ministry cadres nor with the hospital board, nor with the MOH division in charge of public hospitals” (Eid 2001). d. In Egypt, medical staff members are monitored technically by means of the central Ministry and the relevant professional syndicate (union) through its local branches. Administrative and financial reporting are carried out by the regional administration. region is organized in a similar fashion. The reported they had their own budget, and Ministry of Health is considered the princi- 33 percent of facilities were receiving their pal governing body of the health system and budget from the Ministry of Health. Among has the mandate for health policy making, the few (n = 11) facilities that had their own planning, regulation, monitoring, and evalu- budget, just over half were able to freely allo- ation and for ensuring access to essential cate between budget items. The budgets of health services (Jabbour and Yamout 2012). these facilities were mostly allocated for non- Administrators and providers have little medical supplies, and few contained line control over the management of financial items for medications, personnel, payroll, resources. According to data gathered in the contract service, or general expenditures. 2010 Egypt Health and Governance Study School financing is also highly centralized in (EHGS), few health facilities have the flexibil- the Republic of Yemen, where budget deci- ity of their own budget and line item alloca- sions are made through a collaborative pro- tions: only 3 percent of surveyed facilities cess involving central, governorate, and 140 TRUST, VOICE, AND INCENTIVES district-level education, administration, and setting, planning, fi nance, human resource fi nance entities, and individual schools have management, and pedagogy) among the no role in their own budgeting process. central Ministries, provincial and regional Local administrators also have little administrations, and health facilities/schools. decision-making authority over the hiring, The tables also illustrate the degree of finan- firing, and training of staff. In Egypt, for cial and managerial autonomy for each. example, the distribution of teachers across As explained earlier and shown in both governorates is controlled centrally, based on tables, the public education and health sys- the number of registered pupils, and the gov- tems remain heavily centralized in the ernorate distributes teachers to schools under MENA region. its jurisdiction (UNESCO 2010/11). Teachers hired on an open-ended contract can be fired Weak budgeting practices and only by the Educational Directorates, but information management systems teachers who are hired on a fi xed-term con- tract can be fired by the local educational The lack of an explicit performance orienta- authorities as well (Systems Approach for tion in the internal budget processes of all Better Education Results, SABER). No facil- MENA countries further weakens internal ity surveyed in the EHGS sample could dis- accountability. The traditional input-based miss a person for bad performance. In Egypt, budget common in many countries through- the district also decides if and what training out the region does not provide policy makers staff receives, with mandated Ministry and with information on what goods, services, governorate approval for certain levels of or policies are being financed by govern- training and grades of staff. Similarly, in the ment expenditures. In the absence of this Republic of Yemen, school principals have information, attempting to ensure the effi- little or no authority over hiring and fi ring ciency and effectiveness of expenditures is a decisions or teacher assessments. challenging task. Consistent with the broader The development of curricula and peda- global trends in the member countries of the gogy is also highly centralized. In Egypt, the Organisation for Economic Co-operation Ministry of Education is responsible not only and Development (OECD) and elsewhere, for educational policy and its implementa- many MENA countries have sought to tion, but also for determining curricula and restructure incentives for performance textbooks and approving teacher qualifica- through their budgeting process in the hope tions. Curriculum development is carried out of aggregating the inputs used to deliver a with the participation of teachers and other particular group of activities.11 To date, how- school-level stakeholders, but they do not ever, limited progress has been achieved exercise fi nal decision-making power. In the across MENA countries. Republic of Yemen, according to SABER, no The slow progress in introducing a curriculum setting occurs at the subnational, performance-based approach stems in large governorate, or local levels. Teachers cannot part from the fact that many of the basic ele- design the curriculum, and their autonomy in ments of the public financial management choosing teaching methods is limited, but (PFM) system must be put in place before it they do have full autonomy in grading stu- can work effectively. Often these weaknesses dents and deciding whether a student should are fully recognized only when work begins repeat or fail a grade. on a pilot. In Syria in 2008, for example, work Tables 5.1 and 5.2 expand more thor- with pilot Ministries revealed that, in addition oughly on the level of decentralization in the to the standard dual budgeting problem, a public education and health systems in large portion of the sectoral expenditure in selected MENA countries. They do so by education and in agriculture was fragmented providing a clear picture of the division of between the relevant sector Ministry and the responsibilities on five indicators (policy Ministry of Local Administration. In the West INSTITUTIONS INFLUENCING THE CYCLE OF PERFORMANCE 141 Bank and Gaza, it is recognized that the monitor the performance of administrators accounting systems need to be modified to and providers. For example, Morocco’s allow programs to be monitored during health information system suffers from unco- execution of the budget. Although charts of ordinated and duplicated parallel systems, account changes12 are manageable, the lack of standardized terminology, lack of accounting system would not be able to skills for managing databases and data pro- manage complex cost allocations. cessing, underutilization of the available Another challenge is to avoid a situation in information, inadequacy of data production which a reform becomes an information col- for the needs of decision makers, and lack of lation exercise, thereby losing sight of the regular updates and rigorous data (World objectives. When fully developed, a program Bank 2013). This weak system hinders policy structure can involve a range of subprograms makers’ routine access to reliable information and activities. Combined with the associated on the health status of the population, risk performance indicators, the information factors, utilization of services, resource flows, demands can be extensive. This approach and provider performance. might make sense in a well-developed budget system in which skilled staff can use the Administrators—Limited in their ability information effectively. However, in an envi- to hold service providers accountable ronment in which the capacity to prepare and use the information is constrained, the data Evaluation of education and health services collation process can become an end in itself. at the point of delivery largely fails to lead to This was a criticism of the initial work on the improvements in service. In many places, results-oriented budgeting in Jordan carried evaluation simply does not take place or is out from 1998 to 2004. In Morocco, the ini- not captured in the data available. In others, tial budget documents included hundreds of where evaluation systems reportedly exist, input or output indicators, but little attention there are generally few consequences, either was paid to them. positive or negative, in response to such As a result of these challenges, in spite of reviews. the considerable interest in MENA countries In education, despite clearly stipulated in introducing a performance-focused minimum scores for both internal and exter- approach to budgeting, success has been lim- nal evaluations and well-articulated conse- ited to date, and the time frame for effective quences for poor assessments, teachers implementation has been consistently appear to be minimally assessed. For exam- extended. Performance budgeting reforms ple, according to SABER, at the national level are complicated and difficult, in part because in Egypt, the last round of external teacher they rely on having many other elements of evaluations was conducted in 2010. However, the PFM system functioning at the level of a Egypt Education Community Scorecard reasonable standard and in part because a (ECSC) data from schools in the Ismailia broader demand for the data being produced governorate of Egypt suggest that teachers in is needed. It is possible that such reforms will that governorate are evaluated even less fre- produce higher returns in the future, particu- quently (Bold and Svensson 2010). Almost larly in countries such as Jordan or Morocco three-quarters (71 percent) of surveyed that have been working to implement them teachers reported that a quality assurance for some time. However, to date these officer (QAO) had never assessed them. reforms have still not proven to be a simple or When assessments occurred, they were spo- direct route to pressuring line departments to radic and infrequent: 19 percent of teachers improve their performance. reported they were visited once a year, In addition, weak information systems 6 percent once a term, 7 percent once or more hinder the ability of policy makers to make a term, and 12 percent less than once a year. evidence-based decisions and constantly A mong those teachers who had been 142 TRUST, VOICE, AND INCENTIVES assessed, only 21 percent reported that the facilities, établissements de soins de santé de QAO taught them any useful practices. base, or ESSBs (figure 5.12). These facilities The Republic of Yemen is another example receive two types of external supervisory of weak oversight and incentives for educa- visits that cover similar topics, yield similar tion providers. Civil service regulations results, and have similar geographic coverage, stipulated that a public school teacher’s com- indicating significant duplication of efforts. pensation package must be reduced in cases ESSBs are visited by health program manag- of absenteeism without valid cause or without ers, allocated in every province, to supervise previous notification. Absenteeism is to be program drugs in terms of program imple- reported to the Governorate Office of mentation and material supplies to ESSBs. Education, an implementing agency within Health program drugs tend to be part of spe- the Ministry of Education. However, the cific national health programs such as those 2006 Public Expenditure Tracking Survey for diarrhea, family planning, maternity, and (PETS) found that of the schools surveyed acute respiratory infection. These programs in which an incident was reported, only are funded, allocated, and supervised by a 37 percent took any action, and most teachers division in the central administration of the were given only verbal or written warnings. Ministry of Health, as shown in figure 5.13. In health, the 2010 Egypt Health and The visit is generally led by the head of the Governance Study provides some clues about Provincial Service of Infrastructure and how health services delivery is evaluated. Ambulatory Care (SIAAP), who supervises About half (57 percent) of facilities reported program managers in the health delegation using a supervisory checklist for health sys- and is in charge of supervising ESSBs for tem components and the provision of health medical matters. ESSBs also receive visits services. However, the interviewer was able from the provincial health delegation, which to see actual documentation in only about supervises and manages administering other two-thirds of these cases. About half vital and essential drugs to ESSBs and other (46 percent) of facilities also reported con- ad m inist rative mat ters (figu re 5.13). ducting a facility-wide review of mortality, Unsatisfactory SIAAP visits are more likely to but documentation was available for only be followed by a written report than are the 70 percent of these facilities. Seventy percent provincial health delegation visits. However, of surveyed facilities reported a periodic audit because these visits are not typically surprise of medical records or service registers, but ones, the ESSBs have time to present their documentation was unavailable for one-third best (but not necessarily representative) face. of these audits. About three-quarters Procurement rules and processes are often (76 percent) of facilities reported they had a not standardized, and procedures remain quality assurance committee or team or a opaque to those managing procurement. For quality improvement program (75 percent) in example, in Egypt the EHGS found that only place, but about one-fifth of each group about one-third (35 percent) of facilities (17 percent and 23 percent, respectively) were reported standardizing their purchases. unavailable for observation by the inter- There is also little experience in using ten- viewer. Only half (57 percent) of facilities ders or calls for bids on procurement, and could show the interviewer any financial data only a minority of facility respondents were on the facility’s expenses. These gaps could able to correctly describe the official pro- indicate lapses in the reporting, organization, curement processes their facilities followed or existence of mechanisms of quality for all types of supplies, whether nonmedical management. supplies, medical supplies, or heavy equip- The 2011 Morocco PETS and 2011 ment, most of which are managed on a Quantitative Service Delivery Survey (QSDS) monthly basis. Similarly, in Morocco the provide details on supervisory visits con- 2011 PETS indicated that drug management ducted in one cadre of primary health lacked standardization and clarity. There are INSTITUTIONS INFLUENCING THE CYCLE OF PERFORMANCE 143 FIGURE 5.12 Percentage of duplication in supervisory visits to primary health facilities (ESSBs): Morocco, 2011 Consequence: written report Consequence: oral advice Consequence: no measure Unsatisfactory visits Satisfactory SIAAP visit Discussed needs of ESSB Checked the presence of personnel Held informal meeting with personnel Checked temperature of drug refrigerators Discussed administrative issues Discussed drug protocols Checked drug “best before” date Checked drug stocks registry Checked patient registry Received at least one visit 0 20 40 60 80 100 Delegation visit SIAAP visit Sources: PETS (health), Morocco, 2011; QSDS (health), Morocco, 2011. Note: ESSB = primary health facility (établissement de soins de santé de base); SIAAP = Provincial Service of Infrastructure and Ambulatory Care (Service d’Infrastructures d’Action Ambulatoires Provincial). two management systems for drugs: one sys- FIGURE 5.13 In-kind transfer flows to primary health facilities tem addresses essential drugs and the second (ESSBs) in the Moroccan public health system covers program drugs. The two systems Funds have different budgets and allocation sys- Vital and essential drugs Ministry of tems, and they have a different manager in Program drugs Health Materials each province who is in charge of supervis- ing program implementation, including the Procurement Health drug and material supplies for the ESSBs. Center division programs These varying systems of drug management complicate the supply chains and likely sow confusion and disorganization about a facil- Health regions Pharmaciesa ity’s stock. The monitoring mechanisms for drug reception are also weak. For example, the Provincial Provincial 2011 PETS found that in Morocco, only Provincial delegations program pharmacies half of the surveyed ESSBs were using a reg- coordinators istry book for essential drugs compared with 83 percent for program drugs. The Service providers ESSBs remainder kept drug reception slips. The process of monitoring materials received by the ESSBs was even less rigorous than that Source: Data from the Government of Morocco. Note: ESSB = primary health facility (établissement de soins de santé de base). of drugs. For example, for the fi rst delivery a. Only eight regional pharmacies had been created at the end of 2011. If no regional pharmacy in the year 2010, 50 percent of ESSBs that exists, in-kind transfers flow directly to provincial pharmacies. 144 TRUST, VOICE, AND INCENTIVES received materials did not have any paper mechanisms and service delivery perfor- record documenting and monitoring the mance. The reasons for this are not readily process of receiving and distributing mate- apparent, but possibly may be traced to the rials but rather responded “by memory” country’s high-income, relatively homoge- when asked. neous population and the vision of the politi- Such lapses lead to leakages of supplies. In cal leadership. Morocco, the 2011 PETS and the 2011 QSDS Taken together, implementation problems found that almost two-thirds of drugs result in a system largely lacking controls and (61 percent) shipped from provincial delega- accountability mechanisms. Monitoring, tions to the ESSBs were not reaching the des- enforcement, and support mechanisms from ignated ESSBs. The leakage of drugs from the the top are not available to encourage provid- Ministry of Health to the delegations was ers across the service delivery chain to engage much smaller, though still non-negligible at effectively. Nor are there incentives and pro- 24 percent. Program drugs were more likely fessional opportunities to foster internal to go missing than essential drugs. Leakage accountability. The result, as described in was even greater with other supplies: chapters 6 and 7 and illustrated in box 5.1, is 50 –80 percent of six other medical sup- inadequate supplies, shabby infrastructure, plies were not reaching the ESSBs from the providers who are absent or show up for lim- delegations. From nurse overalls (57 percent ited periods, and citizens with unmet needs. leakage) to blankets (82 percent), the 2011 PETS and QSDS reported substantial capture between what delegations sent out and what Social institutions ESSBs received (figure 5.14). As Weir (2003) has argued, often “in the In short, the MENA region suffers from a Arab World, political boundaries and relatively wide gap between legal and policy government policies are surface phenomena frameworks and their actual implementation compared to the deeper infrastructures of (figure 5.15). In most cases, as chapters 6 belief, family, kin, and obligation.” Social and 7 elaborate, regulations and protocols institutions, most broadly conceived to are not always followed in schools and health include norms and regulations within society, facilities. Certainly, there are high-performing affect both soft and formal accountability. countries, most notably the United Arab These institutions vary both within and Emirates, which has devoted an extraordi- across countries. They include social capital nary effort to improving accountability and professional norms that, at least ideally, are nurtured by civil society, as well as norms for communal obligations that are often FIGURE 5.14 Percentage of leakage in medical supplies in formed along the lines of blood ties, ethno- shipments from provincial delegations to ESSBs: Morocco, 2011 linguistic divisions, religion, and other social identity markers. % leakage Nurse overalls Social capital, professional norms, and Bed sheets soft accountability Midwife overalls Social trust and norms of reciprocity can Soap pieces increase individuals’ willingness and ability to exchange resources and engage in cooperative Liquid soap activities across segments of society, as well as Blankets strengthen professional norms that ultimately 0 20 40 60 80 100 can boost service provision. Proponents of civil society argue that civil society organiza- Source: PETS (health), Morocco, 2011. tions can facilitate not only democracy but INSTITUTIONS INFLUENCING THE CYCLE OF PERFORMANCE 145 FIGURE 5.15 Global integrity scores, MENA region 80 70 60 50 Score 40 30 20 10 0 Iraq Egypt, Arab West Bank Algeria Yemen, Syrian Arab Qatar United Rep. and Gaza Rep. Republic Arab Emirates Actual implementation score Legal framework score Source: World Bank 2014a, 23. Note: The rule of law indicator comprises four subcategories: (1) anticorruption law; (2) anticorruption agency; (3) rule of law; and (4) law enforcement. The difference between actual implementation scores and legal framework scores represent “implementation gaps.” Actual implementation scores are based on indicators that start with “in practice,” whereas legal framework scores are based on indicators that start with “in law.” Score: very strong (90–100), strong (80–89), moderate (70–79), weak (60–69), very weak (below 60). also service provision. Putnam argued in making than its counterpart in the Republic Making Democracy Work (1993), for exam- of Yemen. Moreover, as Rothstein (2011) has ple, that the formation of civil society organi- noted, political institutions that fail to con- zations could bring together citizens on a trol corruption reduce social trust, further noncompulsory basis for any number of undermining the development of rule of law, reasons, from forming a bowling league to as discussed in chapter 8. launching a charitable association. Such orga- Even where professional associations are nizations can sometimes act as interest groups, given greater leeway, political constraints on putting pressure on the state or providers for voice often undermine the space for policy- better services, or they can be an integral part making influence and professional develop- of nonstate service provision. ment (Gregg 2013). For example, the And yet political institutions have affected Tunisian General Labor Union and the the development of social trust. Authoritarian Jordanian Medical Association have both regimes have constrained the development of long been allowed to play a more assertive civil society in much of the MENA region. role in engaging in politics than Egyptian In countries such as Egypt, for example, pro- trade unions, and they have challenged fessional associations of health workers incumbent policies. But even in these cases, and teachers were organized in a corporatist the Tunisian and Jordanian governments at system that channeled, and constrained, times took steps to repress them (see also associational activities (Bianchi 1989). Toensing 2011). Elsewhere, a study by El-Jardali et al. (2012) At times, professional associations become found that physicians, nurses, and other more politicized and end up playing a greater medical associations in the Republic of role in political competition than in profes- Yemen were having little influence on the sionalization. For example, under President policy-making process. The same study con- Ben Ali, being a syndicalist in Tunisia was cluded that the Jordanian nurses’ association associated with being outside the central was having a greater impact on policy leadership, and even in opposition to Ben Ali. 146 TRUST, VOICE, AND INCENTIVES Box 5.1 Examples of underperforming health centers in the MENA region Consistently low performance characterizes two The center lacked a sense of dynamism and exhibited provincial health clinics in one MENA country low standards of governance and health care. today. As for its low standards of governance, when asked about staff meetings, the head doctor and act- Health center 1 ing head nurse (the actual head nurse was out that The fi rst health center, which appeared rundown, day) replied that meetings are held periodically as had two full-time doctors, one pediatrician who needed rather than regularly. Generally, the head attends regularly, and seven nurses. But very few doctor calls a meeting only when problems arise or staff members, including the head doctor, were after she meets with the chief medical offi cer. The present when the research team arrived at about clinic also had a decided lack of established pro- 2 p.m., even though the facility is officially open cedures for handling administrative records and from 8:30 a.m. to 4:30 p.m. Staff members said the managing the operations of the facility. Apparently, center receives an average of 80 patients a day, most the head nurse handles these tasks, but she had not of whom visit the facility before 1 p.m. communicated her procedures to other staff mem- The administrative procedures and systems for bers. Furthermore, the head doctor was unable to maintaining equipment appeared to be lacking. answer basic questions about the facility’s system for Even though the center rarely experiences stock-outs managing drug stocks and had no knowledge of the of medications or other problems related to its sup- facility’s budget for medications. ply of drugs, the pharmacy did not appear to be well Completing the picture, the health center had not stocked, at least one medication on the shelves had adopted measures to improve the accountability of already expired, and the refrigerator used to store providers to patients and their families. No sugges- vaccines in the pharmacy did not contain a thermo- tion box was visible in the facility, and few notices meter, threatening the quality of the existing supply. regarding procedures were posted on the walls, apart There were also noticeable shortcomings in the from two signs indicating the location of the offi ce system for ensuring compliance with basic profes- for treating chronic diseases and instructing patients sional standards of hygiene. For example, the sink not to request medications for which they lack pre- in the examination room had no soap (a problem scriptions. No effort had been made to organize out- routinely observed elsewhere), and the individual reach activities in the community. The head doctor examination rooms did not have separate medical repeatedly referred to the center’s relationship with waste containers. Instead, a centrally located garbage the municipality, which periodically sends workers can was devoted to medical waste. Furthermore, the to clean the facility’s garden and coordinates the nurse staffi ng the tuberculosis laboratory and treat- maintenance of the local ambulance, but she seemed ment did not herself follow good hygiene practices; entirely unmotivated to participate in a national she was wearing a large number of gold bracelets on program for improving service delivery, stating that her arms, which is not recommended for staff mem- she and the other staff members were not up to the bers treating infectious patients. task of taking part. Indeed, the Ministry official accompanying the research team emphasized that Health center 2 the major difference between this center and higher- At this rural clinic, services were provided by seven performing facilities boiled down to the quality of staff members—a doctor, an ambulance driver, a the management. In fact, the governance concerns hygienic technician, and four nurses, only three of were particularly stark in this facility because the whom lived locally. When the research team visited the center had recently been renovated and had relatively center during the afternoon, no patients were present. good equipment. INSTITUTIONS INFLUENCING THE CYCLE OF PERFORMANCE 147 The same situation affected professional asso- extent to which these social groups are pres- ciations in Jordan under King Hussein. Such ent and powerful, as well as in the extent to organizations acted more as a bellwether of which individual group members identify political sentiment than as groups that could with their communal group, share its inter- act as an interest group to promote workers’ ests, or abide by its norms. Considerable rights or inculcate professional norms. variation also emerges when citizens are Not only has the role of the professional asked whether local clans or families influ- association in developing a collective sense of ence local politics. For example, as shown in identity and promoting professional norms figure 5.16, the citizens of Tunisia are gener- and pride in a job well done been stifled, ally much less likely than those of Jordan to but so, too, have civil society and the devel- tell survey researchers that there are influen- opment of social trust more generally. As tial families or clans present in their location scholars have noted, civil society organiza- and they are much more likely to view these tions can have perverse consequences for actors as influential in some locations than social trust, particularly when they are politi- others. This fi nding likely reflects very real cized or constrained by closed political differences in the constellation and strength regimes (see Berman 1997; Jamal 2007; of social groups present in various locations. Makuwira 2011). Thus in much of the As a World Bank (2003) report noted, MENA region, there is little reason to believe tribes historically behaved much like state that the dramatic increase in civil society actors: they regulated power, ensured the organizations witnessed in the 1990s raised rights of individuals, and defi ned relation- social capital or advanced professional norms ships in society. They provided, “in their that would serve as the foundation for soft own way, values now called participation, accountability. accountability, predictability, justice, the rule of law, and transparency.” Today, how- ever, there is a great deal of variation in the Social diversity and communal nature of tribes, large families, and other obligations social groups in the region. Some have rela- Although civil society has remained rela- tively formal, regulated mechanisms for tively weak throughout the MENA region, choosing electoral candidates, resolving dis- identity-based groups have tended to thrive. putes, and maintaining community coher- The form varies—from predominantly tribe- ence, while others are internally divided and based identities in Jordan, Libya, and the weak. Nevertheless, in much of the region, Republic of Yemen, to sectarian groups in tribes and other social groups continue to Iraq and Lebanon, to prominent families and play some of these roles, at times to the cha- clans in Morocco, Syria, and Tunisia—but grin of citizens, who see them (much like the role of communal identities has remained state elites) as providing unjust decisions or strong. Indeed, states have developed and exploiting their position (Corstange 2008; maintained political institutions by taking Aslam 2014). into account communal and collective identi- The extent to which tribes or other social ties: a state’s local administrative boundaries groups provide a mechanism to hold service are sometimes drawn along tribal or sectar- providers and their superiors accountable ian lines; local sheikhs serve as the local depends on a number of factors. First, it administrative officials; legal regulations are depends on the extent to which social identi- designed to incorporate tribal or religious ties are linked to strong norms of obligation. law; and projects are implemented in close As described in chapter 4 and as shown in cooperation with the local social elites (Weir evidence from other regions as well, the 2006; Fattah 2011; Aslam 2014). strength of social identities and importance There is important variation within as of social obligations can vary, depending in well as across the MENA economies in the part on state policies and incentives (Jabar 148 TRUST, VOICE, AND INCENTIVES 2001; Sakai 2001; Morris MacLean 2002; diversity and the provision of public goods in Baylouni 2010). Moreover, the role that Jordan, Gao (2012) found that heterogeneous social groups play can depend in part on their areas where either multiple tribes existed or relationship with society. For example, Tsai’s tribes were internally divided were able to 2007 study of service provision in China sug- obtain more resources from the state and gests that social organizations are better at enjoyed higher levels of service provision. fostering good services when they are embed- Similarly, a study conducted in the Republic ded within the community, giving local of Yemen concluded that more resource authorities a reason to cooperate with them. investment in education was correlated with More formally organized social groups are more tribally heterogeneous areas (Egel 2011). also likely to have a greater capacity for gath- Unfortunately, it appears that in both cases ering resources, monitoring services, and more heterogeneous areas are associated with enforcing compliance. greater levels of migration, which makes it The diversity of social groups is also likely difficult to determine whether greater levels to affect service provision. Some observers of development stem from competition or believe more heterogeneous areas provide from the demands of a more mobile, and incentives for political elites to compete for potentially more resourceful, population. support and thus yield better services. In her Indeed, the relationship between social study of the relationship between tribal diversity and service provision remains FIGURE 5.16 Influence of families and clans: Tunisia and Jordan, 2014 a. Do you agree or disagree with the following statement: “Families and clans [in Tunisia] are influential”? 100 90 80 70 60 Percent 50 40 30 20 10 0 ok e Fe am M ana Ka e R n M A Ke h en Mg ra Sid e S ira lB h El ika ra um a Gh i Bo ma Ja una i B S ala zid ba He ine Bi Sa ha B th Le ardo Sa en s kie zah iro b S lh n S ika H uth sra uz Bo llah El mra la o e o t Sid El F rth er r H r Sid Je it us zo s ou ir i A 'Sa ur Bi elh en Ou eg ir Bi led ine er r n A uj Th l as d E li Ha rte id So Mi llal eb l So f Dj Kas ateu M Eas M i ch ke M sin e Bo gu az h s u laa iad i E out M n Ch he zz El Zarz zz nast let M ech rte ua ar u A O aj ed l bi Am i H eri Ka Ba l Me u m Be uro Sid M ho Na affo in h ro M Fa us do nc ou bi ze am b k ua ou ss No Kr n l r r M As o ri uh ba e tE ze o m i E Sid ba se Sa us So Strongly disagree Disagree Agree Strongly agree Source: Transitional Governance Project, 2014. (continued next page) INSTITUTIONS INFLUENCING THE CYCLE OF PERFORMANCE 149 FIGURE 5.16 Influence of families and clans: Tunisia and Jordan, 2014 (continued) b. Do you agree or disagree with the following statement: “Families and clans [in Jordan] are influential”? 100 90 80 70 60 Percent 50 40 30 20 10 0 fa ran ia W di MAl ’ 2 Ka 3 W Al T ayb ia ad a a 3 Al mm n ou h 1 sh Shmsa 4 a A Rakh ow Kaf nubib Al B am ni M as ali a A As hm 1 ad ss hn 2 a A hn 1 Al ssu a 2 M kun Al U nara yn Ni 62 Al Ka l Ba ara ir nje a Ha 2 As a 1 Ja A Hamna l A ab a nn iria Al l Af ha An ararah Al nah ma Qa Ab 4 isi r Ghous 1 Ka ran a 2 yb Mu anjeal 1 M As 5 Rahala q Sa hd h Sh a 2 Al B ’Ou 1 M ayt da A A mia s As l Ma l Ba 1 ka Al ns lad l M dis ra as ia 1 Al Ba Raba M si ’i Al aqa ra 2 a in ro 2 a m ik im AAl J rin i A l T orn ir yb b Ka ou a 2 3 i M us s ha i M S f Al riba usa f A S za a a K d yim Su o u h a ad o Ee Am fra yd ba l Z z Al mi wd a g t ut su De l Ta aki a T ar Al je Sa ub as ira l J A sak ia id ak a Al Al nje a fr d Al ma Ar lith fra sa l S ha Th bil ju aj r n A Qa ou as R m s’a A az f i M yb i M lit A w h a d ad ha h K W Al T a 'A m a w W kh un a a W Kh sh As Ha un Al Ta sh As Al Strongly disagree Disagree Agree Strongly agree Source: Governance and Local Development survey, 2014. opaque (vom Hau and Singh 2014). Some care, electricity, and education in middle scholars have argued that members of one schools (Banerjee and Somanathan 2007). social group are less likely to cooperate with Other scholars have turned from the members of other social groups, and so impact of social groups on intragroup rela- diverse societies are less likely to invest in tions and toward their ability to sanction public goods (Easterly and Levine 1997; their own members. Behavioral games have Alesina et al. 2003; Costa and Kahn 2003). found that co-ethnics are more likely to Others have argued that social diversity cooperate for fear of being sanctioned and in depresses investment in public goods only recognition that by being more closely linked inasmuch as it serves as a proxy for compet- through social networks they face a higher ing interests. They argue that it is the level of possibility of sanction (Habyarimana between-group inequality or polarization et al. 2007). Similarly, an experiment in (Baldwin and Huber 2010) or a history of community-based monitoring in Uganda state building (Miguel 2004; Singh 2010) and found that more homogeneous areas were not social fragmentation per se that depresses able to push for better health service delivery service provision. A longitudinal study in and overcome the free-rider problems inher- India found that the relationship varies across ent in community monitoring after an infor- services, with a negative relationship between mation intervention, apparently because the fragmentation and the provision of health community in homogeneous areas could 150 TRUST, VOICE, AND INCENTIVES impose more sanctions on health care provid- importance on maintaining the social status ers areas where social norms and institutions of their own network. apply to everyone (Bjorkman and Svensson In general, these norms underpin a prac- 2010). In addition, a study in Kenya found tice called wasta, which is Arabic for both that school committees in ethnically diverse “intermediary” (referring to the person) and areas were associated with lower local public “intercession” (the act). Wasta is based on an goods provision. The study showed that implicit social contract in which typically school committees in ethnically diverse areas relatives or members of social or tribal groups were more likely to suffer from local collec- who are in positions of power are obligated tive action failures because of imposing fewer to provide assistance (favorable treatment) to community social sanctions and using less others within the same group (Barnett, verbal pressure against parents who do not Yandle, Naufal 2013). A form of clientelism, contribute to public fund-raising or pay “wasta involves social networks of interper- school fees, resulting in lower funding of sonal connections rooted in family and kin- local schools and worse facilities in 84 pri- ship [as well as other personal] ties and mary schools (Miguel and Gugerty 2005). implicating the exercise of power, influence, Questions remain about the impact of and information sharing through social and social diversity on service provision, but con- politico-business networks. It is intrinsic to siderable evidence suggests that this factor the operation of many valuable social pro- has a significant impact on the quality of ser- cesses, central to the transmission of knowl- vices in the MENA region. Indeed, not only edge and the creation of opportunity” are savvy administrators constrained by (Hutchins and Weir 2006, 143–44). social ties, but they also can activate them to Wasta provides people with solutions to a mobilize community support and incentivize set of social problems and resource allocation their staff. These dynamics are discussed in issues. It allows individuals to obtain public the case studies presented in chapter 3, and services and receive preferential treatment are nicely illustrated in box 5.2 by the Sakhra when dealing with administrative proce- Comprehensive Health Center, where exploit- dures. Those who have wasta can gain uni- ing close ties in the community and good versity admission, obtain a business license, governance practices resulted in exceptional and manage a wide array of other daily tasks. service. Those seeking to change their children’s schools or to fi nd a hospital bed for a sick parent, for example, often require someone The role of social norms, wasta, and in the right place willing to pave the way.13 informal payments Wasta is also often instrumental in gaining The impact of social institutions on public privileged access to economic opportunities, organizations, service provision outcomes, including employment. As Meles (2007) and citizen engagement extends far beyond notes, “Wasta has now become a right and the presence of tribes, or even ethnic, sectar- an expectation” in Arab societies. ian, or other social groups. Underlying Yet wasta also undermines equality. norms that extend across society also affect Many view wasta as a source of nepotism, the provision of services, and they are par- cronyism, and corruption, with historical ticularly influential in much of the MENA roots in a tribal system of social organiza- region because of the weak state and admin- tion (Al-Ramahi 2008).14 It is a form of istrative institutions. Two social norms are social informal networks in a job search, but particularly widespread and influential: the poor and other disadvantaged residents first, individuals are more obligated to remain unlikely to be able to penetrate such respond to friends, family, and others in networks. In the economic sphere, wasta is their social networks than they are to considered a form of “crony capitalism” in strangers, and, second, individuals place which success in business depends on a close INSTITUTIONS INFLUENCING THE CYCLE OF PERFORMANCE 151 Box 5.2 Improving quality of care against all odds: Mobilizing social ties in Jordan The Sakhra Comprehensive Health Center is small program, the Sakhra center also has a local health and slightly disheveled, with evident resource con- committee, which serves as a channel for aggregat- straints. Nevertheless, it is teeming with activ- ing and addressing community demands and ideas. ity and resourcefulness. A sheet on the wall is the Based on its formalized structure and diverse mem- “screen” from the previous night’s presentation on bership, the committee supports the clinic by iden- the safe use of certain medications. A blue curtain tifying and mobilizing resources to address leading cordons off a corner in the maternal and child unit, issues of concern to the community. Closeness to providing privacy for nursing mothers. Staff mem- the community is also visible in the patient-centered bers promptly respond to calls placed over a pub- approach, with a billboard posting responses to lic address system that they proudly explained they patient comments and signs emphasizing patients’ purchased using staff donations. Nestled in one of rights and safety, as well as statements about the the poorest regions of Jordan (Ajloun governorate) goals and principles of the center. and with a catchment area of a little over 15,000 The professional goals of the center are transpar- inhabitants (and recently an additional 1,500 – ent as well. Each year, the center produces an opera- 2,000 Syrian refugees), the Sakhra Comprehensive tional plan that identifies the activities it will carry Health Center is a bustling hub. A dedicated chief out, with clear targets such as reducing hyperten- medical offi cer at the Sakhra center has been able sion by a given percentage and launching a minimal to nurture and draw on close social ties—with the number of outreach programs. directorate, among the staff, and with the close-knit Is good management of the center responsible for community—in order to raise the standards of care its apparent good governance in delivering quality at this rural Jordanian clinic. services? In part it seems to be, but good manage- Strong facility management, complemented by ment can only do so much. The active engagement innovative and decisive leadership, has created a of community leaders and their capacity to motivate clinic environment that fosters staff motivation, local residents to communicate their demands and to encourages feedback, and is responsive to community act as their own health advocates are also vital, espe- needs. Staff members are visibly energized, patient- cially because these factors strengthen accountabil- centeredness is omnipresent, and patients, commu- ity. In this regard, the local health committee plays nity members, and staff have a voice—a combination a crucial role in facilitating communication between of governance drivers that helps explain the clinic’s the center and the community, so that expectations success. Staff members are encouraged to express and responsibilities become clear, rendering provid- their grievances and suggestions and to participate ers accountable to their patients. Thus the Sakhra in determining and implementing solutions to prob- center benefits from good governance both within lems encountered at the clinic. During our visit, the the facility itself and in the community where it is staff repeatedly noted, “We are one hand,” and they located. showed pride in the center’s accomplishments. The clinic also emphasizes engagement with Note: The research team applied systematic measures of the relevant aspects of quality to all the facilities visited. The team was unable to be as systematic the community, which is socially homogeneous in its observations as it could be using close-ended survey questionnaires. and dominated by a relatively influential tribe. Moreover, the sampling was not random but based on the Ministry of Health accreditation process. Indicators considered included the presence and condi- Engagement occurs via multiple means such as out- tion of basic infrastructure; medical records and how they are kept; the system reach programs, informational lectures held at the of maintaining drugs; the credentials of health personnel; the continuous center and the local mosque, home visits for medi- training of health personnel; the presence of protocols for treating specific health conditions; and staff absenteeism. Based on the team’s observations, cal testing and consultations, and health fairs. Like the Sakhra center was providing quality services relative to its constraining other clinics participating in Jordan’s accreditation environment. 152 TRUST, VOICE, AND INCENTIVES relationship between business owners and have grave consequences for effort and government officials. It may take the form of performance. favoritism, tax breaks, legal permits, or Social norms, combined with weak other forms of state intervention (World administrative accountability mechanisms, Bank 2014b). Meanwhile, services may be are also a basis for informal payments. These delivered adequately to members of groups under-the-table fees are charges for services that are politically connected, but poorly to and supplies that are supposed to be free of the rest of the citizenry as a whole. In the charge in the public system (Balázs 1996; absence of weak internal and external Gaal et al. 2006). In addition to monetary checks, such affi liations often come at the payments, informal payments can take the expense of meritocracy and fairness and form of tangible gifts or other favors. become the basis for awarding employment, Informal payments are thought to be sub- obtaining services, and solving disputes, cir- stantial within bureaucratic units and public cumventing states’ formal rules and further sector jobs where there is an excess supply of undermining them. capital and human resources, weak incentive Wasta perpetuates the problems associ- schemes, lack of accountability and govern- ated with weak administrative systems as ment oversight, and an overall lack of trans- well. It contributes to the lack of accountabil- parency. Figure 5.17 shows the percentage of ity, oversight, and internal checks and bal- fi rms in selected countries that make infor- ances within public sector administration. mal payments to public officials to “get Staffing of bureaucracies and Ministries is things done” in the areas of customs, taxes, often based on wasta rather than on merit or licenses, regulations, services, etc. The esti- performance. Indeed, both the 2013 Gallup mates of informal payments tend to be high World Poll and the Arab Barometer 2010–11 in developing and transitioning countries, (Wave II) poll found that the vast majority of which include among them some MENA citizens believe that knowing people in high countries for which data are available such as places is important in obtaining a govern- Iraq, the Republic of Yemen, and Syria. ment job.15 In 6 out of 10 countries surveyed Attempts to understand the scope of informal by the Arab Barometer, the majority of citi- payments and the motivations underlying zens felt that political affiliations were more their existence are often empirically inconclu- or as important as qualifications and experi- sive and contradictory (Gaal et al. 2006). ence in obtaining government employment,16 However, there is evidence that informal and more than half of respondents in all payments are not always viewed as morally countries believed that family and tribal iden- reprehensible (Allin, Davaki, and Mossialos tities were as or more important than qualifi- 2006). And yet even if they are viewed as cations in obtaining a government job. The socially acceptable or expected, informal results did not differ significantly across fees are detrimental to citizens’ welfare in income, employment, or education. terms of equal access to services and the Hiring based on networks and personal quality received. In the health sector, it has relationships not only helps those in adminis- been argued that informal fees can encour- trations simplify their search (as it often does age unprofessional behavior among physi- outside the MENA region as well), but also cians, motivating them to provide good care allows those in positions of power to fulfi ll for only those who make extra payments and social obligations and to demand greater loy- rendering their services unaffordable for the alt y (Fernandez and Weinberg 1997; poor and most disadvantaged. In addition, Al-Ramahi 2008; Pankani 2014). This situa- informal payments could undermine the effi- tion can create perverse incentives for provid- ciency of the health system as a whole; gov- ing loyalty at the expense of effort, and it ernments may turn a blind eye to problems shuts out well-qualified job seekers. The of poor resources and underpayment, and in resultant lack of meritocracy in hiring could turn physicians may resist attempts at reform Estonia (2009) Israel (2013) St. Lucia (2010) St. Vincent and the Grenadines (2010) Chile (2009) Georgia (2008) Colombia (2010) Belize (2010) Uzbekistan (2013) Costa Rica (2010) Ethiopia (2011) Vanuatu (2009) Antigua and Barbuda (2010) Spain (2004) Bosnia and Herzegovina (2013) Armenia (2013) Turkey (2013) Mauritius (2009) Cabo Verde (2009) St. Kitts and Nevis (2010) Honduras (2010) Rwanda (2011) Guatemala (2010) Latvia (2013) Azerbaijan (2013) Grenada (2010) Zimbabwe (2011) West Bank and Gaza (2013) Botswana (2010) Jordan (2013) Suriname (2010) Macedonia, FYR (2013) Uruguay (2010) Ireland (2004) Nicaragua (2010) Burkina Faso (2009) Lithuania (2013) Zambia (2013) Dominican Republic (2010) Bhutan (2009) Fiji (2009) FIGURE 5.17 Percentage of firms making informal payments to public officials: Selected economies, 2005–13 China (2012) Malawi (2009) Namibia (2006) Mexico (2010) Ecuador (2010) Djibouti (2013) Brazil (2009) El Salvador (2010) Czech Republic (2009) Tonga (2009) Belarus (2013) Serbia (2013) Trinidad and Tobago (2010) Morocco (2007) Croatia (2013) Sri Lanka (2011) Slovenia (2013) Korea, Rep. (2004) Nepal (2013) Portugal (2004) Barbados (2010) Mozambique (2007) Indonesia (2009) South Africa (2007) Egypt, Arab Rep. (2008) Moldova (2013) Togo (2009) Montenegro (2013) Paraguay (2010) Bolivia (2010) Jamaica (2010) Argentina (2010) Bulgaria (2013) Senegal (2007) Albania (2013) Guyana (2010) Philippines (2009) Romania (2013) Slovak Republic (2013) Bahamas, The (2010) Timor-Leste (2009) Sierra Leone (2009) Mali (2010) Russian Federation (2012) Kazakhstan (2013) Lebanon (2013) Kosovo (2013) Peru (2010) Greece (2005) Venezuela, RB (2010) Ghana (2007) Uganda (2013) Lesotho (2009) Kenya (2013) Samoa (2009) Mongolia (2013) Panama (2010) Iraq (2011) Niger (2009) Hungary (2013) Lao PDR (2012) Tajikistan (2013) Côte d’Ivoire (2009) Swaziland (2005) Nigeria (2007) Afghanistan (2008) Central African Republic (2011) Chad (2009) Source: World Bank Enterprise Survey, 2005–13. Gabon (2009) Madagascar (2013) India (2005) Pakistan (2007) Bangladesh (2013) Angola (2010) Congo, Dem. Rep. (2013) Tanzania (2013) Cameroon (2009) Kyrgyz Republic (2013) Benin (2009) Gambia, The (2006) Vietnam (2009) Liberia (2009) Burundi (2006) Cambodia (2007) Guinea-Bissau (2006) Algeria (2007) Yemen, Rep. (2010) Ukraine (2013) Poland (2013) Congo, Rep. (2009) Syrian Arab Republic (2009) Mauritania (2005) Guinea (2006) 0 10 20 30 40 50 60 70 80 90 Percent 153 154 TRUST, VOICE, AND INCENTIVES and formalizing informal payments if their some of the subnational variations in service profits from their practice are substantial delivery found in chapter 7. The weak and are not subject to state taxation administrative and political institutions (Mossialos and Karokis 1992). leave much space for social ties and institu- Moreover, evidence from Morocco dem- tions to play an important role in shaping onstrates that side payments can be a serious service delivery. obstacle to poor patients who are trying to access medical care. A survey of 1,000 households conducted by Transparency Potential for reform? International in 2002 in Morocco found that This chapter has described how in much of 40 percent of respondents admitted making the MENA region both formal and informal an illicit payment for a service that was sup- institutions undermine accountability. posed to be free, and 81 percent of those who Political institutions are characterized by made the payment thought that the bribe was centralized power in highly authoritarian, effectual and led to the desired result (Allin, politically captured institutions, limited Davaki, and Mossialos 2006). As the health voice, scarce information, and largely absent minister summed up the problem in 2002: enforcement. Each of these factors breaks “56 percent of those that have the means to down the chain of accountability, thereby pay are benefiting from public hospitals, leading to a gap between how service provi- while 15 percent of the country’s poorest are sion is intended to work and the poor quality paying out of their pockets” ( L a Vie that users encounter at the point of delivery. Economique 2005). Finally, even when infor- Administrative institutions are hampered mal payments are seen as gifts, it is hard to by lack of management capacity in the public determine how much regulation and over- sector, which may take the form of a limited sight are exercised over such payments or flow and monitoring of inputs and outputs, whether children whose parents donated limited performance-related information, receive any preferential treatment by the and a limited authority to impose sanctions school administration, for example. or reward good effort. The result is the The form and relative influence of social inability of one level along the chain to hold and state institutions vary across states, sub- another level accountable. This limited capac- national regions, and individuals. I n ity appears to be a symptom rather than a Lebanon, Libya, and the Republic of Yemen, root cause of the problem in administrative for example, the near absence of effective institutions. The cause appears to be embed- political and administrative institutions ded in political and social institutions, which gives social forces free reign, while in Egypt orient the incentives of public servants away and Tunisia state institutions are more effec- from performance and toward clientelism tive (Lust 2013). Social norms differ as well. and patronage. The by-products are lax rules Although we fi nd high correlations between and regulations, a limited emphasis on moni- the extent to which wasta is seen as neces- toring and results such as the quality of sary for obtaining various goods and ser- services provided to the poor and nonprivi- vices (such as government jobs, health care, leged populations, and weak management and education), the expectation that wasta is practices—all of which further shape the essential in such situations varies consider- incentives of government employees within ably across localities (see figure 5.18 for a the bureaucracy. demonstration of the importance of wasta Finally, social institutions shape the incen- and bribes in Tunisia; similar results have tives of service providers across the chain. been reported for Jordan as well). The same Strong social norms that emphasize obliga- variation is found for informal payments. tions to the members of one’s tribe, ethnic Such variation in the strength of social insti- group, or other social network (often at the tutions at the local level is likely to explain expense of the greater good) produce an Percent Percent So 100 10 20 30 40 50 60 70 80 90 us 0 10 20 30 40 50 60 70 80 90 100 0 El se E El A Am He ra Sid l M mr i A ou a Al nch Sa b r So El Mgh a ba S elh ouj us M ir let id am se o a Ou i Bo id Sid Be uro led u A i A n A uj As li be rou Ja ker Sa lham s m Sa kie i m ba tE d Sa Mat el let z kie eu Ou Jam zit led m t r M Ezzi As el ’S t Al ake M ker a M n Na teu g sra r H hira lla El ench h M a E Je e Bi l M lma Bo nza r M en uh h ch zah M a e Gh eg jla M rgu ar rin on a El e M M ast ok ir e S lh Agree completely M nin Na bib eg e s r a B ri B a Ka ou ne Bi en llah r M Ar laa ha o K jla ch us er B eb g Gh oum ira So J ua ar h us elm e se a M El M l R So azz elh M iad us ou M okn h Agree somewhat se na az in zo e Ka Ria s d u Sid r He h Sa na i B llal Bo ou ou um af A he Sb li Ka Fa l ib Ka sr H hs Fa a laa ell Sa hs o Ba Ke al b bir M uaf So a M 'Sa Disagree somewhat ed k ui en B en Sid B ka a. Use of wasta: Medical treatment, Tunisia b. Payment of bribes for medical treatment, Tunisia in ard i H ard e as o Ba So o s b u B M in Sid So th u M ize ona e ed rt st Bi i Ha ika ze ss en N ir e rte in in or e S th No e o Sid Ha rth Ha uth i E ou ffo lB z Ha Le uz Disagree completely e Dj ssi Kra Bi Le chir er El m z e Dj rt Kram ba Fe er e S Sid Mi rid ba o i E dou M u lB n Ez id th e Ha zou oun Bi Om chir ss ho Ka zert ran iE u iro e S e lF r e ua ou Fe rid n Ka O rian Ez So th iro m a zo ut uh h FIGURE 5.18 Importance of wasta and bribes in obtaining medical treatment and education: Tunisia, 2014 ua ran o nS e Fe ur ou r Ch iana Za th Sid eb rz i B Z ika INSTITUTIONS INFLUENCING THE CYCLE OF PERFORMANCE Sid Th is ou ar i B Che ala zid zis ou bi Ea zid ka s Ea Th t st ala 155 (continued next page) Percent Percent 156 100 100 E 10 20 30 40 50 60 70 80 90 10 20 30 40 50 60 70 80 90 Sa 0 0 El So bal A El l Am us et H mr Sa M ra se Ou en a ba ou Sid le ch let ro iA dA a So Ou Hen uj be ske us led ch El lham r se S A a M Sid idi ske Sid ou id i A Bo r i B rou be u A ou j Sa lham li A kie i Sa Mat li d kie eu Al t Ez t r M zit gh Al Ezz M i Ja ira Be gh t m n A ira M me Be ate l Source: Transitional Governance Project, 2014. M rou n ur ’Sa s Bi Jam ken N a Arou rM sr s m M alla Gh che el El ’Sak h ar rgu B i M en El a r M en M Mel ch zah on h e Na ast Bo rgu ir Agree completely uh a El srall a M ah Gh Meg jla en a r TRUST, VOICE, AND INCENTIVES M za So r El ine ok h n us M So Meg ine se elh us ri Ri se ne Sa adh R ou Bo iad a uh h Je f ajl l Ka a Ka S ma laa Fa laa bib h a Agree somewhat M Keb s Ka Keb az ira sr ira zo M He un az lla a zo l Ka Sb sr iba M una He ok M nin Sa llal on e M ou as ed tir en Je af in lm Bo Fah Ba e So a um s b ut he So h M Sid Bar l B ui Disagree somewhat ed i H do Sid oum ka en as d. Payment of bribes to obtain education, Tunisia in sin i H he as l e s Ba So e c. Use of wasta to obtain education, Tunisia Bi B ine b uth ze a So rte rd u Bi Haf ika No o ze fo Sid Ha rth r t Sid e N uz f i o Ha i El B fouz Ha El B rth ss ec ss ech iE h i E ir l F ir lF Le erid Le erid K Disagree completely Dj O Kram Ka er m iro Om ram Ka ba M ran ua ra iro id e n ne u o Ez So Bi an un Bi zou uth ze So ze h rt u rte ou Ez e So th So r zo ut Dj uh h er Z uth ba a Fe o u r M rzis ria id n o Za a Fe un Sid Sid ria i B Che rzis i B Che na ou bi ou b zid ka zid ika FIGURE 5.18 Importance of wasta and bribes in obtaining medical treatment and education: Tunisia, 2014 (continued) Ea Ea s s Th t Th t ala ala INSTITUTIONS INFLUENCING THE CYCLE OF PERFORMANCE 157 environment conducive to corruption and formal and informal and, similarly, that infor- one that legitimizes the usage of connections mal political and administrative mechanisms to solve individual problems or obtain ser- can operate as well. vices without creating pressures for improv- 3. Expert-coded measures of judicial independence, compiled by the Bertelsmann Transformation ing service delivery systems. Index (Bertelsmann Foundation 2014) and by Overall, this chapter paints a bleak the CIRI Human Rights Dataset (Cingranelli picture. Most MENA countries have a web of et al. 2014) find the MENA region to be among weak and captured political institutions, inef- the regions with the least independent judicia- fective administrative systems, and social ries. It is worth noting that surveys of business norms that place personal connections above leaders conducted by the World Economic the public interest. The result undermines Forum (2014) find that judiciaries in the MENA accountability and service delivery, which is region are similar to those throughout much discussed at greater length in chapters 6 and 7. of the world, although considerably less inde- It also makes effective reform difficult pendent than those in Western Europe, North because weak institutions and poor perfor- America, Australia, and the Pacific. 4. For more on this subject, see Huntington mance undermine citizens’ trust and shape (1968); Drèze and Sen (1989); Wade (1990); engagement, which will be the subject of Maravall (1994); Diamond and Plattner chapters 8 and 9. (1995); Ross (2006); Hasnain (2008); Herb (2009); Fukuyama (2012); Gerring, Thacker, and Alfaro (2012); and McGuire (2013). Notes 5. The public knowledge percentages were taken 1. The World Bank (2003, 39) has defined inclu- from a 2010 survey implemented by the siveness as follows: “the rights and interests Al-Urdun Al-Jadid Research Center, cited in of all groups—particularly women, ethnic Almadhoun (2012). Jordan’s 2007 law scored and religious minorities, and any vulnerable 52 out of 150 points on the ATI rating devel- group—are guaranteed and their concerns are oped by the Center for Law and Democracy addressed by the government.” and Access Info. 2. According to North (1990, 3), institutions 6. These insights are derived from a regional “are the rules of the game in a society, or more online survey administered by the Mohammed formally, are the humanly devised constraints Bin Rashid School of Government (MBRSG) that shape human interaction.” Institutions Governance and Innovation Program in include “formal” rules such as constitutions collaboration with Bayt.com (http://www and laws enforced by the state and “informal” .mbrsg.ae/getattachment/ff70c2c5-0fce constraints such as “codes of conduct, norms -405d-b23f-93c198d4ca44/The-Arab-World or conventions,” both of which are generally -Online-2014-Trends-in-Internet-and.aspx). enforced by society’s members. Similar under- The survey targeted almost 3,000 respondents standings of institutions underpin the litera- in 22 Arab economies (Algeria, Bahrain, ture on domestic politics and international Comoros, Djibouti, Egypt, Iraq, Jordan, relations. See, for example, North and Kuwait, Lebanon, Libya, Mauritania, Weingast (1989); Khalil (1995); Greif (1998); Morocco, Oman, Qatar, the Republic of Simmons and Martin (2001); Hodgson Yemen, Saudi Arabia, Somalia, Sudan, the (2006); and Greif and Kingston (2011). Our Syrian Arab Republic, Tunisia, the United Arab view of institutions has much in common with Emirates, and the West Bank and Gaza). the trifold classification adopted by the World Seventy-five percent of the sample was male; Bank (2014a). Whereas the classification set 14 percent of the sample was between the ages forth by the Bank distinguishes among politi- of 15 and 24; and 44 percent was between the cal, economic, and social and cultural institu- ages of 25 and 34. The countries most repre- tions, equating social and cultural institutions sented in the sample size were (from the largest with informal institutions, we distinguish to the smallest) Egypt, Saudi Arabia, the United between political and administrative institu- Arab Emirates, Algeria, Jordan, and Morocco. tions and, more important, recognize that 7. Accenture Report January 2014, which social and cultural institutions can be both defines digital government as “the optimal 158 TRUST, VOICE, AND INCENTIVES use of electronic channels of communication question “In general, do you agree that know- and engagement to improve citizen satisfac- ing people in high positions is critical to getting tion in service delivery, enhance economic a job?” (wasta) was as follows: 10 percent, competitiveness, forge new levels of engage- Lebanon (n = 99); 12 percent, Jordan (n = 97); ment and trust, and increase productivity of 19 percent, Kuwait (n = 98); 20 percent, public services. A digital government encom- Morocco (n = 98); 20 percent, Republic of passes the full range of digitalization—from Yemen (n = 98); 22 percent, Tunisia (n = 97); the core digitalization of public services to 23 percent, Libya (n = 80); 24 percent, Egypt the digital infrastructure, governance and (n = 99); 25 percent, Saudi Arabia (n = 88); processes, including both front- and back- 27 percent, Algeria (n = 99); 29 percent, Qatar office transformation needed to deliver the (n = 82); 31 percent, West Bank and Gaza new service paradigm.” (n = 96); and 32 percent, Iraq (n = 94). 8. Interview with Ali Ibrahim Al-Khawaldeh, direc- 16. In Lebanon, 2 percent of respondents said tor, Civil Society Organizations Department, employment is obtained without con- Ministry of Political Development, Jordan, by nections; 24 percent, sometimes through Aida A. Essaid, October 18, 2010, Amman, connections; and 75 percent, always through Jordan, and cited in Saif (2012). connections; Republic of Yemen: 5 percent, 9. For a thorough review of the six elements of without connections; 25 percent, sometimes the budget cycle and the reforms that the through connections, and 70 percent, always MENA countries have pursued to date to fur- through connections; West Bank and ther strengthen these elements, see World Gaza; 3 percent, without connections; Bank (2012). 30 percent, sometimes through connec- 10. According to the Arab Barometer, 2010–11 tions; 67 percent, always through connec- (Wave II), the following believe that the par- tions; Iraq: 4 percent, without connections; liament has little or no role in the formation 30 percent, sometimes through connections; of policies: 63 percent, Republic of Yemen; 66 percent, always through connec- 61 percent, Algeria; 61 percent, Lebanon; tions; Jordan: 4 percent, without connections; 50 percent, Iraq; 46 percent, the West Bank 31 percent, sometimes through connections; and Gaza; and 31 percent, Jordan. and 65 percent, always through connec- 11. This discussion draws on Beschel and Ahern tions; Tunisia: 6 percent, without connections; (2012). More updated information on the 33 percent, connections are sometimes impor- experiences of various MENA countries’ with tant; 61 percent, connections are always public sector reforms can be found in Beschel important; Algeria: 10 percent, connections and Yousef (2014). are not important; 35 percent, connec- 12. A chart of accounts (COA) is a financial tool tions are sometimes important; 55 percent, for classifying, recording, and reporting infor- connections are always important; Egypt: mation on financial plans, transactions, and 10 percent, connections are not important; events in a systematic and consistent way. In 35 percent, connections are somewhat impor- particular, it specifies how financial transac- tant; 55 percent, connections are very impor- tions are recorded in a series of accounts by tant; Saudi Arabia: 13 percent, connections defining the scope and content to capture the are not important; 45 percent, connections relevant financial information, and provides are somewhat important; 42 percent, connec- a coding structure for the classification and tions are always important. recording of relevant financial information (both flows and stocks) within the financial management and reporting system (Cooper References and Pattanayak 2011). Abu Ni ma h , H . 2013. “W hy A n nu l t he 13. For a discussion of wasta, see Kilani and Ombudsman Bureau.” Jordan Times, June 11. Sakija (2002). ht tp: //jordantimes.com /why-annul 14. For a full review of the historical roots and -the-ombudsman-bureau. evolution of wasta, see Barnett, Yandle, and Accenture. 2014. Digital Government. Pathways Naufal (2013). to Delivering Public Services for the Future: 15. In the Gallup World Poll (2013), the percent- A Comparative Study of Digital Government age of respondents who disagreed with the Performance across 10 Countries. http:// INSTITUTIONS INFLUENCING THE CYCLE OF PERFORMANCE 159 nstore.accenture.com/acn_com/Accenture Jordan and Lebanon. Bloomington: Indiana -Digital-Government-Pathways-to-Delivering University Press. -Public-Services-for-the-Future.pdf. Berman, S. 1997. “Civil Society and the Collapse Alesina, A., A. Devleeschauwer, W. Easterly, of the Weimar Republic.” World Politics 49 S . K u r l a t , a n d R . Wa c z i a r g . 2 0 0 3 . (April): 401–29. “Fractionalization.” NBER Working Paper Beschel, R. P., and M. Ahern. 2012. Public Series 9411, National Bureau of Economic Sector Financial Management Reform in the Research, Cambridge, MA. Middle East and North Africa: An Overview Allin, S., K. Davaki, and E. Mossialos. 2006. of Regional Experience. Washington, DC: “Pay i ng for ‘ Fre e H e a lt h C a re’: T he World Bank. Conundrum of Informal Payments in Post- Beschel, R. P., and T. Yousef. 2014. Public Sector Communist Europe.” In Global Corruption Reform in the Middle East and North Africa: Re por t 20 0 6 , ed ited by Transpa renc y Lessons of Experience for a Region in International, 62–71. Berlin: Transparency Transition. Washington, DC: World Bank. International. B i a nch i , R . 1989. Un r u ly C o r po ra t i s m : Almadhoun, S. 2012. “Access to Information in Associational Life in Twentieth- Century the Middle East and North Africa Region: An Egypt. New York: Oxford University Press. Overview of Recent Developments in Jordan, Bjorkman, M., and J. Svensson. 2010. “When Is Lebanon, Morocco and Tunisia.” World Bank Community-Based Monitoring Effective? Institute and Middle East and North Africa Evidence from a Randomized Experiment in Social Development Department, World Primary Health in Uganda.” Journal of Bank, Washington, DC. http://wbi.worldbank European Economic Association 8 (2–3): .org /wbi / Dat a /wbi /wbicms /f i les /d r upa l 571–81. - acqu ia / wbi /A l m ad hou n-AT I _ i n _ M NA Blind, P. K. 2011. “Accountability in Public _ Region_ENGLISH.pdf. Service Delivery: A Multidisciplinary Review Al-Ramahi, A. 2008. “Wasta in Jordan: A of the Concept.” Paper prepared for Expert Distinct Feature of (and Benefits for) Middle Group meeting, “Engaging Citizens to Enhance Eastern Society.” Arab Law Quarterly 22: Public Sector Accountability and Prevent 35–62. Corruption in the Delivery of Public Services,” Ammar, W. 2009. “Health beyond Politics. United Nations, New York. http://unpan1.un Beirut, Lebanon.” Country Office, World .org /i nt rado c /g roups /publ ic /do cu ment s Health Organization. /un-dpadm/unpan046363.pdf. Aslam, G. 2014. “Notes from Fieldwork in Yemen Bold, T., and J. Svensson. 2010. “Baseline Data Conducted for SA Flagship.” Unpublished raw Report from the Community Scorecard Pilot data, World Bank, Washington, DC. in Ismailia Governorate.” I nstitute for Balázs, P. 1996. “Our Underlying Interest in International Economic Studies, Stockholm. Dealing with Gratitude Payment.” Lege Artis Brinks, D., and V. Gauri. 2014. “The Law’s Medicinae 6: 660–63. Majestic Equality? The Distributive Impact of Baldwin, K., and J. Huber. 2010. “Economic Judicializing Social and Economic Rights.” versus Cultural Differences: Forms of Ethnic Perspectives on Politics 12 (2): 375–93. Diversity and Public Goods Provision.” Brockmeyer, A., M. Khatrouch, and G. Raballand. American Political Science Review 104 (4): 2014. “Public Sector Size and Performance 644–62. Management: A Case Study of Post Revolution Banerjee, A., and R. Somanathan. 2007. “The Tunisia.” Policy Research Working Paper 7159 Political Economy of Public Goods: Some World Bank, Washington, DC. Evidence from India.” Journal of Development Business Anti-Corruption Portal. 2014. “Qatar Economics 82: 287–314. Country Profile: Private Anti-corruption Barnett, A., B. Yandle, and G. Naufal. 2013. I n it i at ive s .” ht t p: // w w w.bu si ne s s - a nt i “Regulation, Trust, and Cronyism in Middle -corruption.com /country-profiles/middle Eastern Societies: The Simple Economics -east -north-africa/qatar/initiatives/private of ‘Wasta.’” Journal of Socio-Economics -anti-corruption-initiatives.aspx. 44 (1): 41–46. Cammett, M. 2009. “Democracy, Lebanese- Baylouni, A. M. 2010. Privatizing Welfare in the Style.” Middle East Research and Information Middle East: Kin Mutual Aid Associations in Project, August 18, 2009. 160 TRUST, VOICE, AND INCENTIVES Cingranelli, D. L., D. L. Richards, and K. C. Clay. El-Jardali, F., J. N. Lavis, N. Ataya, D. Jamal, 2014. “The CIRI Human Rights Dataset.” W. Ammar, and S. Raouf. 2012. “Use of Ve r s i o n 2 014 . 0 4 .14 . h t t p : / / w w w Health Systems Evidence by Policymakers in .humanrightsdata.com. Eastern Mediterranean Countries: Views, Cooper, J., and S. Pattanayak. 2011. Chart of Practices, and Contextual Influences.” BMC A c c o u n t s : A C r i t i c a l E l e m e n t of t he Health Services Research 12 (1): 200–19. Public Financial Management Framework. Fattah, K. 2011. “Tribes and Tribalism in the Washington, DC: International Monetary Arab Spring.” Center for Middle Eastern Fund. Studies at the University of California-Berkeley Corstange, D. 2008. “Tribes and the Rule of Law C om ment a r y. ht t p: //cme s.b erkeley.edu in Yemen.” Paper prepared for 2008 Annual /com mentar y/tribes-and-tribalism Conference of the Middle East Studies -arab-spring. Association. http://www.bsos.umd.edu/gvpt Fernandez, R. M., and N. Weinberg. 1997. /corstange/doc/corstange-tribes.pdf. “Sifting and Sorting: Personal Contacts and ———. 2011. “Ethnic Clientelism in the Middle H i r i ng i n a R e t a i l B a n k .” A m e r i c a n East.” Unpublished manuscript, University of Sociological Review 62 (6): 883–902. Maryland. Fish, M . S ., and M . K roenig. 20 09. The C ost a , D., a nd M . K a h n. 20 03. “C iv ic Handbook of Nation al L egislatures: A Engagement and Community Heterogeneity: Global S ur ve y. New York: Cambridge An Economist’s Perspective.” Perspectives on University Press. Politics 1 (1): 103–11. Fukuyama, F. 2012. “The Future of History: Davies, T. 2013. “Open Data Barometer: 2013 Can Liberal Democracy Survive the Decline G l o b a l R e p o r t .” Wo r l d W i d e We b of the Middle Class?” Journal of Democracy Foun d ation an d O pe n Dat a In stitute . 23 (January): 14–26. h t t p : // w w w . o p e n d a t a r e s e a r c h . o r g /d l / Gaal, P., et al. 2006. “Informal Payments for o db2 013/ O p e n - D at a- B a rom e t er -2 013 - Health Care: Definitions, Distinctions, and Global-Report.pdf. Dilemmas.” Journal of Health Politics, Policy Desai, R. M., A. Olofsgard, and T. M. Yousef. and Law 31 (2): 251–93. 2009. “The Logic of Authoritarian Bargains.” Gao, E. 2012. “Diverse but Not Diverse: Tribal Economics and Politics 21 (March): 93–125. Diversity and Public Goods Provision in Diamond, L., and M. F. Plattner. 1995. Economic Jordan.” PhD diss., University of Michigan. Reform and Democracy. Baltimore: Johns Gerring, J., S. C. Thacker, and R. Alfaro. 2012. Hopkins University Press. “Democracy and Human Development.” Drèze, J., and A. Sen. 1989. “China and India.” In Journal of Politics 74 (1): 1–17. Hunger and Public Action , edited by Jean Global Integrity. 2009. Global Integrity Report Drèze and Amartya Sen, 204–25. Oxford, 2 0 0 9. ht t ps: // w w w.g loba l i nteg r it y.org U.K.: Oxford University Press. /global_year/2009/. Easterly, W., and R. Levine. 1997. “Africa’s ———. 2010. Global Integrity Report 2010. G row t h Tragedy: Pol icie s a nd E t h n ic h t t p s : / / w w w. g l o b a l i n t e g r i t y. o r g Divisions.” Quarterly Journal of Economics /global_year/2010/. 112 (4): 1203–50. ———. 2011. Global Integrity Report 2011. Egel, D. 2011. “Tribal Heterogeneity and the h t t p s : / / w w w. g l o b a l i n t e g r i t y. o r g A llocation of Development Resou rces: /global_year/2011/. Ev id e n c e f rom Ye m e n .” U npubl i s h e d G re g g , H . S . 2 013. “ T he P ro sp e c t s for manuscript. Demo crat i zat ion i n t he M idd le E ast.” Eid, F. 20 01. “Hospital G over nance and In Govern ance in the Middle E ast and Incentive Design: The Case of Corporatized Nor th Afric a: A Han dbook , edited by P ubl ic Ho spit a l s i n L eba non .” Pol ic y A b b a s K a d h i m , 11 2 – 32 . N e w Yo r k : Research Working Paper WPS 2727, World Routledge. Bank, Washington, DC. http://documents Greif, A. 1998. “Historical and Comparative .w o r l d b a n k . o r g /c u r a t e d /e n / 2 0 01 Institutional Analysis.” American Economic /11/1643372 /hospital -governance -incentive Review 88 (2): 80–84. -design-case -corporatized -public Greif, A., and C. Kingston. 2011. “Institutions: -hospitals-lebanon. Rules or Equilibria?” In Political Economy of INSTITUTIONS INFLUENCING THE CYCLE OF PERFORMANCE 161 Institutions, Democracy and Voting, edited by Newsletter of the American Political Science Norman Schofield and Gonzalo Caballero, Association Organized Section for Qualitative 13–43. Berlin: Springer-Verlag. and Multi-Method Research 7 (1): 29–35. Habyarimana, J., M. Humphreys, D. Posner, ———. n.d. “Institutionalized Ethnicity and Civil and J. Weinstein. 2007. “Why Does Ethnic War.” Unpublished manuscript. Diversity Undermine Public Goods Provision? Lust , E . 20 09. “L eg islat ive E lec t ions i n An Experimental Approach.” American Authoritarian Regimes: Competitive Political Science Review 101 (4): 709–25. Clientelism and Regime Stability.” Journal of Hasnain, Z. 2008. “The Politics of Service Democracy 20 (July): 122–35. Delivery in Pakistan: Political Parties and the ——— . 2 013 . “ I n s t i t u t i o n s a n d S o c i a l Incentives for Patronage, 1988–1999.” Pakistan Governance.” In The Middle East, 13th ed., Development Review 47 (2): 129–51. edited by Ellen Lust. Washington, DC: CQ Herb, M. 2009. “A Nation of Bureaucrats: Press. Pol it ic a l Pa r t ic ipat ion a nd E conom ic Lust, E., S. Hourani, and M. Al-Momani. 2011. Diversification in Kuwait and the United Arab “Jordan Votes: Election or Selection?” Journal Emirates.” International Journal of Middle of Democracy 22 (2): 119–29. East Studies 41: 375–95. Maktabi, R. 2013. “Female Citizenship in the Hodgson, G. M. 2006. “What Are Institutions?” Middle East: Comparing Family Law Reform in Journal of Economic Issues 40 (1): 1–25. Morocco, Egypt, Syria and Lebanon.” Middle Huntington, S. 1968. Political Order in Changing East Law and Governance 5 (3): 280–307. Societies. New Haven, CT: Yale University Makuwira, J. 2011. “Civil Society Organizations Press. (CSOs) and the Changing Nature of African Hutchins, K., and D. Weir. 2006. “Guanxi and Politics: The Case of CSO - Government Wa s t a : A C ompa r i s on .” T h u n d e rb i rd Relationship in Malawi.” Journal of Asian Inte r n ation al Bu sin ess Re vie w 48 (1): and African Studies 46 (6): 615–28. 143–44. Maravall, J. M. 1994. “The Myth of the Jabar, F. A. 2001. “Sheikhs and Ideologues: Aut horit a ria n Adva nt age.” Jour n al of Deconstruction and Reconstruction of Tribes Democracy 5 (October): 17–31. under Patrimonial Totalitarianism in Iraq, McGuire, J. W. 2013. “Political Regime and 1 9 6 8 –1 9 9 8 .” I n Tr i b e s a n d Po w e r : Social Performance.” Contemporary Politics Nationalism and Ethnicity in the Middle East, 19 (1): 55–75. edited by Faleh Abdul-Jabar and Hosham Meles, M. 2007. “Understanding People’s Dawod, 69–109. London: Saqi. Attitude towards the Use of Wasta.” Master’s Jabbour, S., and R. Yamout, eds. 2012. Public thesis, Cranfield School of Management, Health in the Arab World. Cambridge, U.K.: Bedford, U.K. Cambridge University Press. Miguel, E. 2004. “Tribe or Nation? Nation- Jamal, A. 2007. Barriers to Democracy: The building and Public Goods in Kenya versus Other Side of Social Capital in Palestine and Tanzania.” World Politics 56 (3): 327–62. the Arab World. Princeton, NJ: Princeton Miguel, E., and M. Gugerty. 2005. “Ethnic University Press. Diversity, Social Sanctions, and Public Goods Kao, K. 2012. “Jordan’s Ongoing Election Law in Kenya.” Journal of Public Economics B at tle.” Sada (on li ne), Ju ne 5. ht t p: // 89: 2325–68. carnegieendowment.org/2012/07/05/jordan Morris MacLean L. 2002. “Constructing a Social -s-ongoing-election-law-battle/ck59. Safety Net in Africa: An Institutionalist Khalil, E . L . 1995. “Organizations Versus Analysis of Colonial Rule and State Social Institutions.” Journal of International and Policies in Ghana and Côte d’Ivoire.” Studies Theoretical Economics 151 (3): 445–66. in Comparative International Development Kilani, S., and B. Sakija. 2002. Wasta: The 37 (3): 64–90. Declared Secret . A mman: Jordan Press Mossialos, E., and A. Karokis. 1992. “Greece: Foundation. Health-care Reforms.” The Lancet 340 (8810): La Vie Economique (Morocco). 2005. February 4. 41–42. Lieberman, E. S., and P. Singh. 2009. “Measuring North, D. C. 1990. Institutions, Institutional the State’s Institutionalization of Ethnic C ha nge a n d E c o n o m i c Pe r fo r m a n c e . C at e g o r i e s a c r o s s T i m e a n d S p a c e .” Cambridge, U.K.: Cambridge University Press. 162 TRUST, VOICE, AND INCENTIVES North, D. C ., and B . R. Weingast. 1989. Relations, edited by Walter Carlsnaes, Thomas “C onst it ut ions a nd C om m it ment: T he Risse, and Beth A. Simmons, 192–211. New Evolution of Institutional Governing Public York: Sage. Choice in Seventeenth-Century England.” Singh, P. 2010. “We-ness and Welfare: A Jo u r n al of E c o n o m i c H i s t o r y 49 (4): Longitudinal Analysis of Social Development 803–32. in Kerala, India.” World Development 39 (2): Olivier de Sardan, J. P. 1999. “A Moral Economy 282–93. of Corruption in Africa.” Journal of Modern Toensing, C. 2011. “Tunisian Labor Leaders African Studies 37: 25–52. Reflect upon Revolt.” Middle East Report 41 (1). Pankani, W. 2014. “System Overload: A Reversed http://www.merip.org/mer/mer258 /tunisian Collective Action Dilemma in Botswana and -labor-leaders-reflect-upon-revolt-0. Ghana.” PhD diss., University of Florida. Tosun, M., and S. Yilmaz. 2008. “Centralization, Prettitore, P. 2012. “Who Needs Legal Aid Decentralization and Confl ict in the Middle Services? Addressing Demand in Jordan.” East and North Africa.” Policy Research MENA Knowledge and Learning Quick Notes Wo r k i n g P a p e r 47 74 , Wo r l d B a n k , Series (Number 62), World Bank, Washington, Washington, DC. DC. https://openknowledge.worldbank.org Ts a i , L . 2 0 07. “Ac cou nt abi l it y w it hout /handle/10986/10845. Democracy: Solidary Groups and Public ———. 2013. “Justice Sector Services and the G oods Prov ision i n Ru ra l C h i na.” I n Poor in Jordan: Determining Needs and Cambridge Studies on Comparative Politics. Priorities.” MENA Knowledge and Learning Cambridge, U.K: Cambridge University Press. Quick Notes Series (Number 96), World Bank, UNESCO (United Nations Educational, Scientific Wa sh i ng ton , D C . ht t p: //sit ere s ou rc e s and Cultural Organization). 2010/11. World .worldba n k .or g / I N T M E N A / R e sou rc e s Data on Education. Egypt. 7th ed. http:// /QN96.pdf. www.ibe.unesco.org/fileadmin/user_upload Putnam, R. 1993. Making Democracy Work. /Publications/WDE/2010/pdf-versions/Egypt Princeton, NJ: Princeton University Press. .pdf. Ross, M. 2006. “Is Democracy Good for the Vom Hau, M., and P. Singh. 2014. “Ethnicity, Poor?” American Journal of Political Science S t a t e C a p a c i t y, a n d D e v e l o p m e n t : 50 (4): 860–74. Reconsidering Causal Connections.” In The Rothstein, B. 2011. The Quality of Government. Politics of Inclusive Development, edited by Chicago and London: University of Chicago Samuel H ickey, Kunal Sen, and Badru Press. Bukenya. Oxford: Oxford University Press. Saif, I. 2012. Political Economy Analysis for Wade, R. 1990. Governing the Market: Theory Policy Dialogue and Programming (PEA): The and the Role of Government in East Asian Case of Jordan. New York: United Nations Industrialization. Princeton, NJ: Princeton Development Programme. University Press. Sakai, K. 2001. “Tribalization as a Tool of State Weir, D. T. H . 20 03. “Hu man Resou rce Control in Iraq: Observations on the Army, the Development in the Middle East: A Fourth Cabinets and the National Assembly.” In Paradigm.” In HRD in a Complex World , Tribes and Power: Nationalism and Ethnicity edited by M. Lee, 69–82. London: Routledge. in the Middle East, edited by Faleh Abdul- Weir, S. 2006. A Tribal Order: Politics and Law Jabar and Hosham Dawod, 136–64. London: in the Mountains of Yemen. Austin: University Saqi. of Texas Press. Shryock, A., and S. Howell. 2001. “‘Ever a Guest World Bank. 2003. Better Governance for in Our House’: The Emir Abdullah Shaykh Development in the Middle East and North Majid al-Adwan, and the Practice of Jordanian A f r i c a : E n ha n c i ng In c l u si ve n e s s a n d House Politics, as Remembered by Umm Accountability. Washington, DC: World Bank. Sultan, the Widow of Majid.” International ———. 2004. World Development Report 2004: Journal of Middle East Studies 33 (2): Making Services Work for Poor People. 247–69. Washington, DC: World Bank. Simmons, B. A., and L . L . Martin. 2001. ———. 20 07. Decentralization and Loc al “International Organizations and Governance in MENA: A Survey of Policies, Institutions.” Handbook of International Institutions, and Practices. A Review of INSTITUTIONS INFLUENCING THE CYCLE OF PERFORMANCE 163 Decentralization Experiences in Eight Middle Egypt Health and Governance Study, World Bank, E a st an d Nor th Afric a C ountries. http://documents.worldbank.org /curated /en Washington, DC: World Bank. http://www / 2010 / 0 6/16332 545/eg y pt -ma nagement - wd s .w o r l d b a n k . o r g /e x t e r n a l /d e f a u l t -service-quality-primary-health-care -facilities /WDSContentServer/WDSP/IB/2010/07/21/0 -alexandria-menoufia-governorates 0 0333 037_ 2 010 07 210 03109/ R endere d Gallup World Poll, http://www.gallup.com / PDF/365160ESW0whit1Box /services/170945/world-poll.aspx 0349464B01PUBLIC1.pdf. Global Integrity Index (no longer published), Global ———. 2008. The Road Not Traveled: Education Integrity, https://www.globalintegrity.org/ Reform in the Middle East and North Africa. Governance and Local Development survey, Washington, DC : World Bank. Prog ra m on G over na nce a nd L o c a l ——— . 2 01 2 . P u b l i c S e c t o r F i n a n c i a l Development, Yale University, http://gld Management Reform in the Middle East and .commons.yale.edu/research/ North Africa: An Overview of Regional Open Budget Index, International Budget Experience. Washington, DC: World Bank. Partnership, http://survey.internationalbudget ———. 2013. “Public Expenditure Review of the .org/#rankings Health Sector in Morocco.” World Bank, Open Data Barometer, Open Data Research Washington, DC. Network, http://www.opendataresearch.org ———. 2014a. “Between Discretion and Fairness: /project/2013/odb The Rule of Law in Everyday Life in the PETS (Public Expenditure Tracking Survey), Middle East and North Africa.” Report No. World Bank, ht tp: //web.worldbank.org 88962 - M N A , Pover t y R e du c t ion a nd / W B SI T E / E X T ER NA L / TOPIC S/ E X T Economic Management (MNSPR), World SOC I A L DE V E LOPM E N T/ E X T PC E NG Bank, Washington, DC, June 24. /0,,content M DK:2050770 0 ~pagePK :14 8956 ~p i P K : 216 618 ~ t h e S i t e P K ———. 2014b. The Unfinished Revolution: :410306,00.html Bringing Opportunity, Good Jobs and Greater Polity IV Annual Time-Series 1800–2013, Center Wealth to All Tunisians. Washington, DC: for Systemic Peace, http://www.systemicpeace World Bank. http://www-wds.worldbank.org .org/ /external/default/WDSContentServer/WDSP QSDS (Quantitative Service Delivery Survey), /IB/2014/09/16/000456286_201409161447 World Bank, ht tp: //web.worldbank.org 12/Rendered/PDF/861790DPR0P12800Box38 / W B SI T E / E X T ER NA L / TOPIC S/ E X T 5314B00PUBLIC0.pdf. P OV E RT Y/ E X T P S I A / 0 , , c o n t e n t M D K World Economic Forum. 2014. “The Global :20467190 ~isCU RL:Y~menuPK:1108016 Competitiveness Report 2014–2015.” http:// ~pagePK:148956~piPK:216618~theSitePK reports.weforum.org/global -competitiveness :490130,00.html -report-2014-2015/. SABER (Systems Approach for Better Education Results), World Bank, http://saber.worldbank .org/index.cfm Data sources Tra nsit iona l G over na nce Projec t , ht t p: // Arab Barometer, 2010–11 (Wave II), http://www transitionalgovernanceproject.org/ .arabbarometer.org/instruments-and-data World Bank Enterprise Survey, http://www -fi les .enterprisesurveys.org/ Bertelsmann Transformation Index, Bertelsmann Worldwide Governance Indicators, World Bank, Foundation, ht tp: //w w w.bti-project.org ht t p: //data.worldba n k.org /data- catalog /downloads/bti-2014/ /worldwide-governance-indicators Service Delivery Performance PART III I n part II, we introduced the cycle of performance framework, explaining that the weaknesses in education and health services delivery are rooted in institutions. We also described how state institutions in the Middle East and North Africa often lack both internal and external accountability, which undermines policy implementation. In this part, we turn our attention to performance at the point of service delivery. We explore the efforts and abilities of teachers and health professionals and the avail- ability of key inputs such as instructional materials in schools and medicines in health facilities. We then look at how such performance is affected by institutions. Drawing on surveys, we focus first on the national level (chapter 6) and then explore the nature and extent of subnational variation in service delivery performance (chapter 7). 165 MAP III.1 Student scores on TIMSS mathematics in MENA, 2007–11 IBRD 41535 | MARCH 2015 Black Sea Ca s pi an Sea Mediterranean Sea ATLANTIC Me SYRI AN TUNISIA MALTA dite OCEAN rranean Sea LEBANON A.R. IRAQ ISLAMIC REP. M OR OCCO WEST BANK AND GAZA OF IRAN JORDAN KUWAIT A L G E R I A L I B Y A ARAB REP. BAHRAIN OF EGYPT SAUDI QATAR UNITED ARABIA ARAB EMIRATES Student scores on TIMSS OM AN Re mathematics, 2007–11 d Se a More than 450 401 – 450 REP. OF YEMEN 350 – 400 en Arabian Ad MENA average 407 f of Sea Gul Less than 350 Non-MENA/non-OECD average 445 DJIBOUTI No data available OECD average 504 Data source: TIMSS (Trends in International Mathematics and Science Study), Boston College. Source: World Bank (IBRD 41535, March 2015). Service Delivery: Gaps in Providers’ Efforts and Abilities 6 and in Inputs • Performance indicators at the point of service delivery in the Middle East and North Africa (MENA) reveal weaknesses in providers’ efforts and abilities and in the availability of key inputs such as instructional materials and medicines. • Absenteeism seems widespread, particularly in public schools and health clinics. • Adherence to curriculum and health care protocols is limited, in part driven by limited abilities. • Schools in the MENA region are short on instructional materials, and health facilities are short on medications. C aptured political institutions and weak administrative structures, as described in the preceding chapter, often undercut incentives for providers, administrators, and policy makers to ensure the provision of quality services for the poor and other disadvantaged populations. There is variation across the Middle East and North Africa (MENA) region and within countries, but close study of the service delivery process in education and health reveals overall weak- Performance: effort nesses in providers’ efforts and abilities, as and ability to meet well as in the distribution of key inputs into citizens’ needs service delivery by administrators and policy makers. In this chapter, we draw broadly on the World Bank Group’s Service Delivery Indicators (SDI) approach1 in examining the efforts and abilities of staff and the 167 168 TRUST, VOICE, AND INCENTIVES Table 6.1 Key service delivery indicators International Mathematics and Science Study (TIMSS), on average, 22 percent of students Education Health in MENA countries attended schools whose What providers do (providers’ efforts) principals reported that teacher absenteeism Absence from school Caseload was a serious problem (figure 6.1), and Absence from classroom Absence from facility 25 percent of students attended schools where Time spent teaching late arrival of teachers was a serious problem What providers know (providers’ abilities) (TIMSS, 2011). Furthermore, 14 percent of Minimum knowledge Diagnostic accuracy school principals in Jordan and Tunisia and Test scores on English, Adherence to clinical guidelines 9 percent in the United Arab Emirates mathematics, pedagogy Management of maternal/neonatal complications reported in the Programme for International What providers have to work with (availability of resources) Student Assessment (PISA) that teacher absen- Students per textbook Drug availability teeism seriously hindered student learning Equipment availability Equipment availability (PISA, 2012). Absenteeism in both schools Infrastructure availability Infrastructure and health facilities has also surfaced as a Source: http://www.sdindicators.org/. problem in the available country surveys (figure 6.2), which allow analysis at the sub- availability of the key inputs and resources national level and in relation to other service that contribute to a functioning school or delivery indicators (see chapter 7). In the health facility. The full set of these indicators Republic of Yemen, for example, one-third of is shown in table 6.1. Because focused Service surveyed public health facilities had at least Delivery Indicators surveys have not been two employees absent at a time, and in all conducted in MENA countries, this chapter hospitals at least one staff member was and the next, which analyzes subnational absent, with an average of about nine employ- variation in service delivery performance, ees absent. In Morocco, the World Bank’s cover a subset of indicators determined by 2011 Quantitative Service Delivery Survey the available data. (QSDS) and 2011 Public Expenditure Tracking Survey (PETS) found that 27 percent of staff members were absent across basic Providers’ efforts health care facilities (établissements de soins The efforts of teachers and health workers, as de santé de base, ESSBs). well as of administrators and policy makers, According to TIMSS and survey data, broadly represent a manifestation of the exist- absenteeism is somewhat lower in private ing institutions and accountability mecha- schools than in public schools. Analysis of nisms. Weak political and administrative the 2010 Egypt Health and Governance accountability mechanisms leave space open Study (EHGS) data has suggested that for social institutions seeking to motivate pro- 40 percent of female staff members were viders. These institutions fill the void effectively absent at least once a month as opposed to by motivating teachers and health workers in 52 percent of male staff members. Similarly, some local contexts, as our examples of local staff on term contracts had lower absentee successes in chapter 3 illustrate. But, more rates than staff on open-ended contracts, and commonly, the weaknesses in formal account- staff in facilities where the local media were ability hurt or distort providers’ efforts. reportedly critical had lower absentee rates than those in localities where the media were positive or did not play a role. Absenteeism and lack of motivation While at school, teachers may not be The available data suggest weaknesses in pro- teaching in the classroom. As the World viders’ efforts in both education and health, B a n k’s Systems Approach for B et ter particularly in terms of relatively high rates of Education Results (SABER) and the available absenteeism. According to the Trends in surveys document, in some MENA countries, SERVICE DELIVERY: GAPS IN PROVIDERS’ EFFORTS AND ABILITIES AND IN INPUTS 169 FIGURE 6.1 Percentage of students whose principals report that teacher absenteeism is a serious problem in their schools: OECD, non-OECD, and MENA economies, 2011 60 50 40 Percent 30 20 10 0 pu rab co sia ia an an A za s r ain n D CD p. ta te no EN EC ab Re Ga oc rd Om ni Qa OE ira hr ic Re n A ba O M Ar Tu Jo or bl ic Ba nd Em n- Le am ria M i ud No ka Sy ab Isl Sa an Ar n, tB Ira d ite es W Un Source: TIMSS, 2011. Note: MENA = Middle East and North Africa; OECD = Organisation for Economic Co-operation and Development. teachers’ working time is defined only as FIGURE 6.2 Percentage absent of total employed the time spent at school, and their hours in health facility: Arab Republic of Egypt, Morocco, of instruction are not given (Arab Republic of and Republic of Yemen, 2010 and 2011 Egypt) or are not recorded (Republic of 40 Yemen). During the PETS visits in the Republic of Yemen in 2006, only 42 percent 35 of teachers were in the classroom teaching. % absent of all employed 30 Nineteen percent were absent; 8 percent were reported as “idle”; and the rest were carrying 25 out authorized nonteaching tasks. 20 The efforts of providers are only partly 15 associated with their workload and satisfac- tion. By international standards, provider 10 workloads appear to be lower in schools and 5 higher in health facilities. Provider satisfac- 0 tion is relatively low in the MENA region, Egypt, Arab Rep. Morocco Yemen, Rep. especially in public schools (TIMSS). In Egypt, Morocco, and the Republic of Yemen Sources: Egypt: EHGS, 2010; Morocco: PETS (health), 2011; QSDS (health), 2011; Republic of Yemen, QSDS (health), 2010. (for which the relevant surveys are available), satisfaction among health professionals is on 170 TRUST, VOICE, AND INCENTIVES average very low, while their workload, par- not available on the costs of such dual ticularly in Morocco, is relatively high by employment—for example, higher absentee- international standards. Among the surveyed ism and deterioration of the quality of teach- staff at Morocco’s ESSBs, only 20 percent of ing or health care—it is clear that, in doctors reported job satisfaction, whereas principle, it presents a conflict of interest and 58 percent expressed their desire to leave the potentially perverts incentives (that is, teach- ESSB they served (PETS, Morocco, 2011; ers might not teach well during school hours QSDS, Morocco, 2011). in order to increase the demand for their pri- Staff motivation is reportedly a problem. vate lessons afterward). The EHGS (2010) found in Egypt that nearly Private tutors are also attractive to parents half of health facility managers reported lack because their children do better in school, of staff motivation as the main constraint to especially if tutored by their own teachers. service delivery, followed by lack of supplies For this reason, the surveyed parents spoke (39 percent), shortage of qualified staff against any ban on tutoring by public school (38 percent), lack of equipment (33 percent), teachers. and lack of medications (29 percent). Similarly, patients in public health facili- ties often find their physicians—as public sec- tor employees who may simultaneously hold Tutoring and dual practice jobs in the private sector3 —available only in Weaknesses in providers’ efforts in public their private practice. As some users of public schools generate a demand for (and are exac- facilities noted, “There isn’t anything in the erbated by) the supplementary services avail- evening except the private examination of the able in the private sector. Schoolchildren in unit doctor.” In the private sector, by con- many MENA countries rely heavily on tutor- trast, qualitative analysis from the EHGS ing to gain basic skills. For example, the shows that opening times are significantly Survey of Young People in Egypt (SYPE) con- more reflective of patient needs. Although the ducted by the Population Council found that exact magnitude of this phenomenon in the about 58 percent of students in primary edu- MENA region is not fully understood, anec- cation and 64 percent in secondary education dotally it appears that as much as 100 percent were being tutored privately, often by their of doctors may be engaged in dual practice in own public school teachers outside the class- some MENA countries. room (SYPE, 2009). The theoretical and empirical literature For teachers, tutoring (even if banned for indicate that dual practice may shift physi- public school teachers such as in Egypt) often cians’ attention toward generating income in becomes an important source of income, pos- their private practice, thereby raising absen- sibly compromising their efforts in the class- teeism in public hospitals and further compli- room (Hartmann 2008; Sobhy 2012). cating staffing in rural facilities. Ferrinho Teachers in the five Egyptian governorates et al. (2004) found that in República surveyed by PETS tended to hold multiple Bolivariana de Venezuela, doctors missed jobs simultaneously; 90 percent in Fayoum, about 37 percent of their contracted service 85 percent in Dhakalia, 79 percent in Cairo, hours, while in Costa Rica, a majority of 67 percent in Luxor, and 65 percent in Minea doctors and nurses felt physicians were unjus- reported multiple jobs. Poorer governorates tifiably absent from work or, even when pres- reported a lower proportion of such arrange- ent, often saw private patients in public ments, but this may stem from fewer job hospitals during public sector hours. And opportunities. 2 Focus group discussions in because urban centers offer more opportuni- Egypt and Tunisia indicated that teachers ties for dual or private practice, the recruit- received significantly higher income from ment and retention of physicians in rural tutoring than from their salary. Although areas become even more challenging, widen- system at ic , qu a nt it at ive ev idenc e i s ing health inequity. SERVICE DELIVERY: GAPS IN PROVIDERS’ EFFORTS AND ABILITIES AND IN INPUTS 171 Where dual practice occurs, particularly private practice in public facilities (France). in low-income countries, physicians have Many MENA countries have attempted to been known to give their best performance regulate dual practice to minimize its at their private practice while exerting a negative impacts on social welfare. At one minimal effort at public hospitals. According extreme, economies such as Egypt permit to a study by Das et al. (2013), public sector dual practice without restrictions, while at doctors in Madhya Pradesh, India, are more the other extreme West Bank and Gaza and likely to give the correct treatment for Tunisia have tried to establish a complete angina in their private practice than in their ban on dual practice. Other countries such public practice. Physicians also reportedly as Saudi Arabia have imposed a complete refer public facility patients to their private ban on all public sector physicians with the practice or engage in “cream skimming” of exception of those working in university profitable patients by intentionally altering hospitals. And yet most MENA countries the quality of treatment, increasing waiting continue to grapple with dual practice, times, and reducing communication with whatever the instituted policy, to mitigate its patients in the public hospitals to divert potentially adverse effects on health service patients to their private practice (Jan et al. quality, efficiency, and equity. 2005). Meanwhile, patients with poorer To succeed, MENA policy makers need to education and health may be more vulnera- recognize that dual practice in developing ble to such inducements (Eggleston and Bir countries is a symptom of a deeper problem 2006). The result is the capture of clients of wea k st a nd a rds , mot ivat ion , a nd and often efforts to push patients from the accountability. In MENA countries, as in low-cost public services to the more expen- many other countries, service standards sive private ones, further increasing the typically do not exist or are not monitored fi nancial burden on them. and enforced; professional bodies play a lim- If appropriately regulated, dual practice ited role in regulating and monitoring per- can have positive effects. These include the formance; and civil society is not empowered ability of the public sector to hire qualified to exert peer pressure or consumer pressure doctors at a reduced cost; a reduced reliance on physicians to adhere to ethical and pro- on informal payments; mobilization of pri- fessional norms. The poor organizational vate participation in health care and transfer and management practices in public hospi- of skills and knowledge between private and tals, the poor monitoring systems, and the public practices; and reductions in waiting high levels of impunity in the health sector times for patients (Araújo, Mahat, and result in the underperformance of health Lemiere 2014). Whether these possible posi- workers in many developing countries. As tive effects outweigh the potentially negative illustrated by the recent experience in effects mainly depends on the strength of the Greece,4 dual practice policies have little existing accountability mechanisms, includ- effect unless issues of standards, motivation, ing the norms of professional conduct and and accountability are addressed (Araújo, public service. Mahat, and Lemiere 2014). By contrast, The regulatory approaches to dual prac- strong monitoring and enforcement capacity, tice vary widely, ranging from allowing dual transparency, and well-established health practice without restrictions to a complete financing systems facilitated a successful reg- ban (such as in Canada and Turkey). Other ulation of dual practice in Canada, France, countries fall within that range—for exam- and the United Kingdom (Araújo, Mahat, ple, offering exclusive contracts and supple- and Lemiere 2014). mentary payments for full-time public A systematic review of international expe- sector employees (Spain and Portugal), rience identified the presence of the follow- restricting private sector income (France ing factors as key to success in addressing and the United Kingdom), or allowing dual practice: (1) a well-organized health 172 TRUST, VOICE, AND INCENTIVES fi nancing system; (2) strong regulatory and influence of institutions on performance, monitoring systems covering prices, services, provider compliance has been found related and quality in both the public and private i n a st at ist ic a l ly sig n i f ic a nt way to sectors, including strong professional boards accountability. to monitor and regulate providers; (3) well- As a proxy for providers’ abilities, adher- established civil society groups to provide ence to policies, curricula, and clinical care feedback and to curtail loss of quality in pri- protocols appears relatively low in the vate and public services; and (4) a political MENA countries where observations have commitment to action as well as a profes- been made. In Egypt, for example, as mea- sional commitment to ethics (Kiwanuka sured by the 2010 Eg ypt Health and et al. 2010). Governance Study, the majority of doctors did not follow the standard protocols of clinical care. Only about half of patients Providers’ abilities with diabetes or coronary heart disease What providers know and what approaches (figure 6.3) and two-thirds of children they apply to the existing policies and stan- patients (figure 6.4) were weighed as part of dards are to a significant degree a function of an outpatient visit. In schools, according to institutions. In this context, our observations the limited evidence available, curricula of the abilities of teachers and health workers and teaching policies are not adequately largely relate to the gap between the rules implemented. The 2010 Egypt Education and policies, on one hand, and the reality, on Community Scorecard project found that the other—that is, the challenge of policy what teachers actually do in the implementation and enforcement docu- classroom—in terms of practice, content, mented in the preceding chapter. and pedagogy—may differ widely from the In the empirical research, provider com- existing policies and regulations (Bold and pliance with policies and standards has been Svensson 2010). Although many MENA predictive of health care quality (Das and countries have developed robust strategic Ham mer 2014). A nd considering the frameworks and policies for education, FIGURE 6.3 Percentage of chronic care observations conducted: Arab Republic of Egypt, 2010 Take blood pressure in the arm Examine the chest Examine the abdomen while the patient is in supine position Weigh the patient Examine feet or any wound or pressure points Examine the back of the thorax with a stethoscope Examine the patient for sensations and reflexes Measure the blood sugar with stick Measure the circumference of the waist Conduct an eye exam using an ophthalmoscope Measure the circumference of the hips 0 20 40 60 80 100 120 % of chronic care observations Diabetes Coronary heart disease Source: EHGS, 2010. SERVICE DELIVERY: GAPS IN PROVIDERS’ EFFORTS AND ABILITIES AND IN INPUTS 173 as demonstrated by the World Bank’s FIGURE 6.4 Percentage of child outpatient visit observations SABER project (2010–11), which targeted conducted: Arab Republic of Egypt, 2010 the technical and vocational education sys- tems, student assessment systems and Take temperature using thermometer teacher policies in those countries perform Use stethoscope on chest and back less well on system oversight and service Weigh the child delivery. For example, although curricula Check throat with tongue compressor reforms across the MENA countries seek to without light Plot weight on a grow chart (% of those reduce rote learning, 70 percent of students who did weigh the child) in those countries reported memorizing for- Count respiration (breath) mulas or procedures in at least half of the Check skin turgor for dehydration math classes they attended. These weak- Check for pallor by looking at lower lip nesses are particularly pertinent as MENA Look in ear and feel behind ear countries seek further reforms to match the Feel the child for fever or hotness supply of skills with demand. The reliance Check throat with tongue on memorizing lessons in math seems on compressor and light average higher in MENA countries than Press both feet (check for edema) elsewhere (figure 6.5). Check arms and shoulders Among teachers, skills in math and peda- 0 50 100 gogy may be more limited, particularly in % of child outpatient visit observations rural areas. As reported by principals in the Source: EHGS, 2010. FIGURE 6.5 Percentage of students (grade 8) who report reliance on memorizing lessons in math: OECD, non-OECD, and MENA economies, 2011 100 90 80 70 60 Percent 50 40 30 20 10 0 an za CD ain on a p. ic co D an r s A sia ta te bi EN EC bl Re Ga oc rd Om na ni Qa OE hr ira ra pu O M Jo Tu or ic iA ba Ba nd Em n- Re am M ud Le No ka b ab Isl Sa ra an Ar n, nA tB Ira d ria ite es W Sy Un Never Some lessons About half the lessons Every or almost every lesson Source: TIMSS, 2011. Note: MENA = Middle East and North Africa; OECD = Organisation for Economic Co-operation and Development. 174 TRUST, VOICE, AND INCENTIVES TIMSS, over half of MENA students attend Surveys, 2012). In the Republic of Yemen, schools that have a shortage of qualified almost 40 percent of the pool of teachers math teachers (figure 6.6). Although most lacked a postsecondary teaching diploma teachers in the MENA region have relatively from a Teacher Training Institute, which is high levels of education and report feeling the minimum educational qualification prepared to teach (TIMSS, 2011), the Early required by the Ministry of Education to Grade Reading Assessment (EGRA) points to become a teacher. Most of these unqualified a lack of knowledge of the specific pedagogi- teachers were found in rural schools cal techniques that lead to better reading (76 percent) and were teaching basic educa- skills and the Early Grade Mathematics tion (91 percent). Only 35 percent of the Assessment (EGMA) to a focus on teaching teachers teaching grades 1–6 held postsec- procedural skills rather than a deep under- ondary diplomas or higher qualifications standing of mathematical concepts. In Iraq, (PETS, Republic of Yemen, 2006). Other evi- half of teachers reported that they had dence suggests that teachers often cannot received no preservice training in how to identify weaknesses in foundational math teach reading and math (Iraq Education and reading skills, and when they can they do FIGURE 6.6 Percentage of students (grade 8) who attend schools with severe shortages of specialized math teachers: OECD, non-OECD, and MENA countries, 2011 70 60 50 40 Percent 30 20 10 0 CD CD n r ain s ia an A za co ic sia ta te no EN bl b Ga oc rd ni Qa OE OE ira hr ra u ba M Tu Jo or ep iA Ba nd m n- Le M R bE ud No ka ab Sa ra an Ar dA tB n ite ria es W Un Sy Source: TIMSS, 2011. Note: MENA = Middle East and North Africa; OECD = Organisation for Economic Co-operation and Development. SERVICE DELIVERY: GAPS IN PROVIDERS’ EFFORTS AND ABILITIES AND IN INPUTS 175 not address them adequately so they can countries—attend schools that have severe cover the necessary material as prescribed by shor t age s of i nst r uc t iona l m ater ia ls the national curricula (EGRA, EGMA). (figure 6.7). The reported shortages of mate- Recognizing the challenges associated rials were especially high in Saudi Arabia, with acquiring the necessary skills, especially the Syrian Arab Republic (preconfl ict), and in the context of confl ict, aid organizations West Bank and Gaza. and ministries of education sometimes work Serious shortages have been reported by together to launch teacher training programs. both public and private schools and are par- Kurdish Iraq, which has a concentration of ticularly high in rural areas. In Morocco, for Syrian and Iraqi refugees, saw the launch of example, 63 percent of surveyed rural schools an advanced training program for all teach- compared with 19 percent of surveyed urban ers working in primary schools across the schools reported serious shortages of instruc- camps within the Erbil governorate (UNHCR tional materials in 2011 (TIMSS). Some 2014). The training covers psychosocial sup- shortages may arise because of leakage. The port, education methodologies, project man- PETS in Egypt revealed that schools in the agement , as well as mon itori ng a nd Luxor and Ismailia governorates received evaluation. And yet many teachers in class- only 0.3 and 1.7 textbooks per student, rooms fi lled with refugees or displaced chil- respectively, whereas the records of the dren do not have the skills or experience Ministry of Education reported textbook needed to teach large class sizes and students procurement for these governorates of over with differentiated needs and from different 22 textbooks per student (PETS, Egypt, backgrounds. Students with special needs are 2008). Some computers were available in at a particular disadvantage in this instance. most schools covered by TIMSS in the In the health sector, across the MENA MENA region, but nearly half of schools in region, rural areas are especially prone to the Islamic Republic of Iran, one-third in experiencing shortages of qualified profes- Saudi Arabia, and 12 percent in Morocco sionals, and openings are partly filled by reported having no computers at all (TIMSS, expatriate staff in the Gulf countries. 2011). On a more basic level, overcrowding Surveys in Egypt, Morocco, and the Republic and lack of electricity, water, and toilets (par- of Yemen have documented serious staff ticularly separate toilets for female teachers shortages in most rural localities. In the and girls) appear to be an acute problem in Republic of Yemen, for example, during the the Republic of Yemen and Djibouti. 2010 QSDS survey, only 27 percent of public Overcrowding and reliance on multiple shifts health units had physicians on the roster, have become a challenge more recently in the and only 40 percent had a nurse. In health communities in Lebanon and Jordan hosting centers, 59 percent had a physician on staff Syrian refugees. and 62 percent had a nurse. Among hospi- In health, shortages of medications are tals, 13 percent had no doc tor, and widespread. The surveyed facilities typically 25 percent had no nurse. lacked essential medications such as oral pen- icillin (Egypt) and metformin (Morocco)— see figure 6.8. In Morocco, in three-quarters Availability of key inputs of ESSBs, at least one drug was out of stock; Revealing the weaknesses in internal controls almost a third of the sampled drugs were out and other basic administrative institutions, of stock; and replenishment of out-of-stock MENA countries experience shortages of key drugs took two months on average (QSDS, inputs in services such as instructional mate- Morocco, 2011). Furthermore, rural ESSBs rials in schools and medicines in health facili- with maternity wards lacked the basic equip- ties. According to the 2011 TIMSS data, on ment for pregnancy monitoring and baby average about 45 percent of students in the delivery. In the Republic of Yemen, facilities MENA region—in both richer and poorer of all types across governorates were found 176 TRUST, VOICE, AND INCENTIVES FIGURE 6.7 Percentage of students (grade 8) who attend schools with severe shortages of instructional materials: OECD, non-OECD, and MENA economies, 2011 70 60 50 40 Percent 30 20 10 0 CD n co CD sia s an r ain A an ia za ic ta te no EN bl ab Ga oc Om rd ni Qa OE OE ira hr pu ba M Ar Tu Jo or Ba nd Em n- Re Le M i ud No ka ab b Sa ra an Ar nA tB d ite ria es W Un Sy Source: TIMSS, 2011. Note: MENA = Middle East and North Africa; OECD = Organisation for Economic Co-operation and Development. poorly resourced and short of standard utili- well as their relatively high cost are the main ties as well as medications. Similarly in Egypt source of citizen dissatisfaction. and Tunisia, facilities, especially in rural The prevalence of gaps in the availability areas, have been found underequipped (Saleh of key inputs such as instructional materials et al. 2014). and essential medicines can be traced back Shortages in key inputs appear to nega- to systemic weaknesses in the public pro- tively affect service quality and citizen satis- curement systems in the MENA countries. faction. With the exception of Oman and These weaknesses range from a lack of Saudi Arabia, the reported lack of instruc- transparency and procurement planning to tional materials is associated with lower stu- excessive centralization, poor quality of bid- dent test scores on the TIMSS. Surveys have ding documents and technical specifications, reported that shortages of medications as repetitive rejection of bids, and the limited SERVICE DELIVERY: GAPS IN PROVIDERS’ EFFORTS AND ABILITIES AND IN INPUTS 177 FIGURE 6.8 Medication availability: Morocco (2011) and the Arab Republic of Egypt (2010) 100 80 60 Percent 40 20 0 % not available % not available Morocco Egypt, Arab Rep. Aspirin Metformin 850 g Vitamin A Oral Penicillin Sources: Morocco: PETS (health), 2011; QSDS (health), 2011; Egypt: EGHS, 2010. participation of suppliers in both the educa- can often be traced to the wider political, tion and health sectors. These bottlenecks, administrative, and social institutions and coupled with political interference and cor- accountability mechanisms. However, more ruption, also add cost. The prices of essential research is needed to pinpoint which institu- medicines, for example, are high in the tions and types of formal and informal MENA countries compared with the inter- incentives influence provider behavior in nationally referenced prices. which way. In this respect, the case studies presented in chapter 3 offer a useful illustra- tion of the possible positive institutional Interpreting service delivery influences at the local level. performance It is important as well to recognize the This chapter has documented the selected demand-side constraints that affect service performance weaknesses at the point of ser- delivery performance and results. In the vice delivery. In doing so, it has highlighted a MENA region, these constraints stem mainly mix of gaps in providers’ efforts and abilities from the labor market distortions flowing and the availability of key inputs. Although from the existing social contract, as noted in simple internal control mechanisms can chapter 2 and further discussed in chapter 5. ensure an adequate distribution of textbooks These distortions mainly affect the perfor- to schools and medicines to health facilities, mance of the education and training systems. much more complex institutions and account- They are especially pronounced in the Gulf ability mechanisms—informal and well as Cooperation Council (GCC) countries, and formal—are at play in driving the behavior of they are connected to the established welfare- teachers and health workers. sharing mechanism at the core of the social In chapter 5, we outlined the key politi- contract (box 6.1). cal, administrative, and social institutions Chapter 7 describes the extent and nature that shape the incentives of providers. We of performance variation at the point of ser- argued that the roots of service delivery per- vice delivery within countries and illustrates formance problems, such as provider absen- the importance of looking beyond averages in teeism, poor quality of teaching or care, and examining service delivery challenges and shortages of textbooks and medicines, their institutional drivers. 178 TRUST, VOICE, AND INCENTIVES BOX 6.1 The demand-side story in the quality of service provision D evelopment sp ecia l ist s now recog n i z e t he Rents therefore affect education and health supply-side “good governance” mechanisms that in different ways. They do not have a statistically drive the quality of service provision: transparency, significant impact on health, but they negatively high-quality public sector management, monitor- affect the quality of education. Analysis of the 2011 ing and regulatory mechanisms, and independent TIMSS math scores worldwide for fourth- and accountability institutions. The demand-side story, eighth-grade students as a measure of education however, has received less attention. The general quality reveals that the relation between rents and assumption is that citizens demand and value high- education is statistically significant and substan- quality services across all sectors. And yet citizens’ tial. Controlling for institutions and other covari- demands for services differ across countries and sec- ates, an increase in rents by about US$22,000 per tors, and the quality of services is associated with capita decreases, other things being equal, the the level of demand citizens exhibit. TIMSS scores by 70–90 points. By contrast, when This finding is clearly demonstrated by exam- one controls for institutions, in 2012 rents had no ining the impact of natural resource rents on the statistically discernible effect on health outcomes, quality of education and health services. Rents may as measured by standardized mortality rates and impede the development of institutions that give per capita disability-adjusted life years (DALY) for providers incentives to use resources efficiently. But noncommunicable diseases only (Global Health they also shape citizens’ demands for different ser- Observatory, 2012). vices. Although the demand for high-quality health Moreover, analysis of the 2011 TIMSS stu- services is high across both rentier and nonrentier dent surveys in eight MENA GCC and non-GCC countries, citizens in rentier economies are less likely countries finds that the differences in education to demand high-quality education, leading to a provision are driven by a lower concern about lower supply of high-quality education even if insti- education. With few exceptions, students in non- tutional quality is taken into account. rentier countries report that their parents talk more The explanation for this is simple: citizens in about school with them, they feel more like they rentier economies are less concerned about attain- belong in school, they believe it is important to do ing a high-quality education because students and well at school, and they discern a closer connection their parents do not see education as critical to a between success at school and their future career career. In the rentier system, they are able to obtain than students in rentier countries. good jobs and a high standard of living regardless These findings offer new insights that are of the quality of education they attain. For exam- worth considering. For one thing, they highlight ple, when college graduates are guaranteed public the importance of taking citizens’ demand for ser- sector employment—as in Qatar—their incentives to vice delivery more seriously. Moreover, because attain a high-quality education decline. By contrast, the demand for different services can vary signifi - although unearned foreign income may free citizens cantly between rentier and nonrentier countries from the need to attain a high-quality education, it and across different sectors such as education and does not free them from concerns about their health. health, they suggest that policy makers and prac- Thus rents have no effect on health status, and so titioners must consider much more carefully both the demand for high-quality health care is high in supply- and demand-side forces. both rentier and nonrentier countries. Source: Alaref, Lueders, and Lust (2014). SERVICE DELIVERY: GAPS IN PROVIDERS’ EFFORTS AND ABILITIES AND IN INPUTS 179 Notes Eggleston, K., and A. Bir. 2006. “Physician Dual Practice.” Health Policy 78 (2–3): 157–66. 1. http://www.sdindicators.org/. F e r r i n h o , P. , W. Va n L e r b e r g h e , I . 2. By law in Egypt, teachers employed in a pub- Fronteira, F. Hipólito, and A. Biscaia. 2004. lic school are allowed to simultaneously hold “Dual Practice in the Health Sector: Review of the other teaching positions in other schools. Evidence.” Human Resources for Health (1): 14. However, teachers on open contracts are not Hartmann, S. 2008. “The Informal Market of legally allowed to hold another job (SABER). Education in Egypt: Private Tutoring and Its 3. This phenomenon in the health care setting Implications.” Working Paper 88, Institut für is known as dual practice or “moonlight- Ethnologie und A frikastudien, Main z , ing.” It has been documented as a common Germany. practice in both developed and developing Ja n , S . , Y. B i a n , M . Ju mp a , Q. M e n g , countries. N. Nyazema, P. Prakongsai, and A. Mills. 4. Greece’s attempt to ban dual practice 2005. “Dual Job Holding by Public Sector (1983–2002) is an example of an attempt to Health Professionals in Highly Resource- regulate the phenomenon without address- Constrained Settings: Problem or Solution?” ing the deep underlying issues that led to it Bulletin of the World Health Organization in the first place; it backfired. Because of the 83 (10): 771–76. weak accountability and monitoring system K iwa nu ka, S . N., A . A . K i neng yere, in the Greek health sector, the ban did not E. Rutebemberwa, C. Nalwadda, F. Ssengooba, result in a reduction of informal payments, Olico-Okui, and G. W. Pariyo. 2010. Dual nor did it eliminate dual practice, which con- Practice Regulatory Mechanisms in the Health tinued outside the regulatory jurisdiction of Sector: A Systematic Review of Approaches government. and Implementation. London: EPPI-Centre, Social Science Research Unit, Institute of Education, University of London. References Saleh, S. S., et al. 2014. “The Path Towards Alaref, J., H. Lueders, and E. Lust. 2014. “Rents Universal Health Coverage in the Arab and Returns: The Political Economy of Uprising Countries Tunisia, Egypt, Libya, and Resource-Endowments on Social Service Yemen.” The Lancet 383 (9914): 368–81. Provision.” Paper presented at the Workshop Sobhy, H. 2012. “The De-Facto Privatization of on Comparative Political Development in Post- Secondary Education in Egypt: A Study of Soviet Russia/Eurasia and the Middle East, Private Tutoring in Technical and General Abu Dhabi, December 8–10. Schools.” Compare: A Journal of Comparative Araújo, E., A. Mahat, and C. Lemiere. 2014. and International Education 42: 47–67. “Managing Dual Job Holding among Health UNHCR (United Nations High Commissioner for Workers: A Guidance Note.” World Bank, Refugees). 2014. “Interagency Operational Washington, DC. Update: Syria.” UNHCR, Geneva. Bold, T., and J. Svensson. 2010. “Baseline Data Report from the Community Score Card Pilot in Ismailia Governorate.” I nstitute for Data sources International Economic Studies, Stockholm EGMA (Early Grade Math Assessment), U.S. University. Agency for International Development, https:// Das, J., and J. Hammer. 2014. “Quality of www.eddataglobal.org/math/ Primary Care in Low-Income Countries: Facts EGRA (Early Grade Reading Assessment), U.S. a n d E c o n o m i c s .” A n n u a l R e v i e w o f Agency for International Development, https:// Economics 67: 525–53. www.eddataglobal.org/reading/ Das , J., A . Hol la , M . K remer, a nd EHGS (Egypt Health and Governance Study), K. Mu ralidharan. 2013. “Qualit y and World Bank, http://documents.worldbank.org Accountability in Health: Audit Evidence from /c u r a t e d /e n / 2 010 / 0 6 / 16 332 5 45 /e g y p t Primary Care Clinics in India.” http:// -management-service-quality-primary-health federation.ens.fr/ydepot/semin /texte1213 - care-facilities-alexandria-menoufia /DAS2013QUA.pdf. -governorates 180 TRUST, VOICE, AND INCENTIVES Global Health Observatory, World Health QSDS (Quantitative Service Delivery Survey) Organization, http://apps.who.int/gho/data World Bank, http://web.worldbank.org /WB /node.main.DALYCTRY?lang=en SITE/EXTERNAL/TOPICS/EXTPOVERTY I raq E ducation Sur veys, U. S. Agency for /EXTPSIA/0,,contentMDK:20467190~isCUR International Development, http://pdf.usaid L:Y~menuPK:1108016~pagePK:148956~piPK: .gov/pdf_docs/PA00JBVT.pdf 216618~theSitePK:490130,00.html PETS (Public Expenditure Tracking Survey), SABER (Systems Approach for Better Education World Bank, ht tp: //web.worldbank.org Results), World Bank, http://saber.worldbank / W BSI T E / E X T ER NA L / TOPIC S / E X T .org/index.cfm SOC I A L DEV E LOPM E N T / E X T PC E NG SYPE (Survey of Young People in Egypt), Population /0,,content MDK:20507700~pagePK:148956 Council, http://www.popcouncil.org/research ~piPK:216618~theSitePK:410306,00.html /survey-of-young-people-in-egypt-final-report PISA (Programme for International Student -and-data-sets Assessment), Organisation for Economic TIMSS (Trends in International Mathematics and Co-operation and Development, http://www Science Study), Boston College, ht tp: // .oecd.org/pisa/ timssandpirls.bc.edu/ Subnational Variation in Service Delivery Performance 7 • Service delivery performance varies widely within countries, even as service delivery systems are centralized in countries in the Middle East and North Africa. • Even where the political and administrative accountability mechanisms are weak, some communities are able to motivate providers to adhere to standards and deliver quality services. • Apart from the familiar local factors such as wealth, it matters how local leaders and local institutions fill the institutional gaps penetrating service delivery systems from the national level. • The existing forms and sources of variation allow identification of where an effective response to the existing institutional constraints already exists. A s earlier chapters have described, they can, and a few that are extraordinary (as many challenges are associated with we saw in the case studies in chapter 3), the delivery of social services in the somehow managing to overcome the same Middle East and North Africa (MENA). The difficulties confronting everyone else and array of complex challenges range from a consistently deliver outstanding services. lack of basic resources and limited account- This chapter seeks to identify the nature ability to outdated practices and perverse and extent of this variation in performance, incentives. Levels of performance are on aver- to explain where and why it occurs, and to age below where they should be, but there is draw on these explanations as a basis for great variation within them, all the way down promoting systemic improvement. The analy- to the individual villages and even within sis draws on a combination of secondary and specific schools and health clinics. No matter (some) primary data, and so does not always what unit of analysis is selected, one can iden- reach the level of precision we might desire, tify some areas in which the quality of perfor- not least because most of the data were not mance is truly awful, a majority in which designed to answer the specific questions most providers are struggling to do the best we are asking. For present purposes, 181 182 TRUST, VOICE, AND INCENTIVES however, the goal is aspirational, and our data, the analysis was conducted at both the evidence is indicative: to bring about the governorate and district levels. The priority reforms required, MENA countries must indicators of interest in analyzing both data acquire a more comprehensive and accurate sets were those best reflecting the process of sense of how they are performing and, on service delivery. They included the availabil- this basis, gain insights into what improve- ity of specific forms of infrastructure and ments might be sought and what could be medical equipment, the rates of absenteeism done to bring them about. Lessons can be among medical and other staff, and the learned from outside sources, to be sure, but availability of essential medications. Annex they mostly are found first at home, from A briefly describes the QSDS data set for the those who have already figured out how to Republic of Yemen. make things work. In this chapter, we document the nature Variation in infrastructure and and extent of the subnational variation in equipment availability service delivery quality primarily in the area of health services and (to a lesser extent) In the Republic of Yemen, there is consider- education. Although these services are typi- able subnational variation in the provision of cally administered in a highly centralized infrastructure such as electricity, beds, and manner, as described in chapter 5, this chap- water (figures 7.1 and 7.2). At the governorate ter aims to detect better-performing entities level, Ibb had the highest percentage of health at the facility, district, or regional level in care facilities with electricity (86 percent) and order to identify more proximate and tangi- also the highest number of beds (an average ble factors associated with their relative suc- of 8.4 per facility). Sana’a had the highest cess. In the opening sections on health percentage of health care facilities with water services in the Republic of Yemen and (91 percent), while at the other end of the Morocco, we draw on both quantitative sur- spectrum Raimah had the lowest access to vey data and qualitative material from inter- both electricity (only 6.3 percent of facilities) views with country experts about how they and water (12.5 percent). Very few facilities perceive and understand variation in service had access to telephones, wireless, heating, delivery. Where possible and appropriate, we and vehicles or ambulances, preventing integrate the results of the quantitative and subnational analysis. qualitative analyses into a single account. Even wider variation is evident at lower Subsequently, we present summary evidence units of analysis. At the district level, Alshaar of a more indicative nature, based on sur- had the highest mean number of beds across veys in a range of countries: education and facilities, with an average of 8.4. The five dis- health services in the Arab Republic of tricts with the highest number of beds (10 or Egypt and education in the Republic of more) and their respective governorates were Yemen, Morocco, Saudi Arabia, Oman, in ascending order: Alshaar (Ibb), Alsawadiah Qatar, the United Arab Emirates, Bahrain, (Al-Baidha), Mothikerah (Ibb), Mokairas and Tunisia. (Al-Baidha), and Arhab (Sana’a). Subnational variation in health Variation in absenteeism services delivery in the Republic At the governorate level, Raimah had the low- of Yemen est level of absent health facility employees, In the Republic of Yemen, subnational varia- 27 percent, compared with 29 percent in tion in health services delivery was examined Al-Baidha, 31 percent in Ibb, and 40 percent using the 2010 Quantitative Service Delivery in Sana’a. Raimah, however, was by far the Survey (QSDS) for that country.1 Because of weakest performer in terms of access to water the nat u re and coverage of the and electricity, suggesting that problems in one SUBNATIONAL VARIATION IN SER VICE DELIVERY PERFORMANCE 183 FIGURE 7.1 Availability of electricity and water in health care facilities, by governorate: Republic of Yemen, 2010 100 90 80 70 % of facilities with access 60 50 40 30 20 10 0 Ibb Sana’a Al-Baidhai Raimah Electricity Water Source: QSDS (health), Republic of Yemen, 2010. domain (logistics, access) are not necessarily different than the mechanisms for maintain- associated with problems in another (relational ing equipment and improving amenities—in challenges such as attendance). There also was other words, strategies for successful reforms considerable variation within Raimah itself; at will have to be tailored to the specific charac- the district level, the level of absent health teristics of the problems and contexts facility employees varied between 0 and 90 involved. percent (figure 7.3). From an analysis of the relationship Explaining variation in health services between measures at the district and gover- in the Republic of Yemen norate levels emerge strong positive correla- tions (0.85 or above) between the availability What is the explanation for why and where of amenities (electricity, water, heat, tele- such variation occurs in the Republic of phone), but just moderate correlations for Yemen? There is no straightforward answer. the availability of beds (table 7.1). Staff The variables one might ordinarily expect to absenteeism, however, correlates only very account for these differences, such as wealth weakly with both equipment and amenities or location (for example, remote rural), seem (less than 0.10) at the district level; more to be far from consistent predictors. Of the moderate correlation appears between 4 governorates (of 21) covered in the QSDS absenteeism and the availability of beds, survey, Raimah tends to be consistently the electricity, and water at the governorate most poorly resourced and underperforming level. At the facility level, there is less overall governorate, and yet compared with the other correlation in general, with weak correla- governorates it had the fewest (though not tions among all investigated measures. significantly) number and percentage of These important findings suggest that the absentee workers and the lowest percentage mechanisms for managing staff may be very (22 percent) of patients who reported having 184 TRUST, VOICE, AND INCENTIVES FIGURE 7.2 Average number of beds per health care facility, by district: Republic of Yemen, 2010 Alshaar Alsawadiah Mothikerah Mokairas Arhab Hamdan Baadan Thi Sofal Alqafr Alhaymah Aldak Ibb Rural Saafan Yareem Alfaraa Alssowmaah Alttafah Bani Mattar Sabah Alsadah Bilad Alttaam Noaman Radman Alttyal Hobeish Manakhah Bani Hosheish Alaymah Alkhar Almalagem Alnaderah Algaafariah Alodein Bani Dhabian Algobein Albaidhaa Almakhader Alsayani Geblah Khawlan Kosmah Alqoraishiah Mazhar Alsalafiah 0 5 10 15 20 25 30 35 40 Average number of beds Source: QSDS (health), Republic of Yemen, 2010. SUBNATIONAL VARIATION IN SER VICE DELIVERY PERFORMANCE 185 FIGURE 7.3 Absenteeism in health care facilities, by district: Republic of Yemen, 2010 Alttafah Bani Mattar Kosmah Geblah Bani Dhabian Alshaar Hamdan Manakhah Alsalafiah Alaymah Alkharigiah Mazhar Mokairas Sabah Khawlan Alhesn Hobeish Arhab Almakhader Yareem Alttyal Alnaderah Bani Hosheish Alssowmaah Alqoraishiah Alhaymah Aldakhiliah Algaafariah Baadan Noaman Alodein Alsadah Alsawadiah Mothikerah Alfaraa Bilad Alttaam Alqafr Algobein Thi Sofal Saafan Radman Khawlan Ibb Rural Alsayani Almalagem Albaidhaa TOTAL 0 0.2 0.4 0.6 0.8 1.0 % absent of all employed Source: QSDS (health), Republic of Yemen, 2010. 186 TRUST, VOICE, AND INCENTIVES Table 7.1 Correlation of absenteeism in health care facilities and other measures at the governorate and district levels, Republic of Yemen Absenteeism No. of beds Electricity Heat Water Telephone Governorate Absenteeism 1 No. of beds −0.246 1 Electricity −0.192 −0.160 1 Heat −0.015 0.578 0.647 1 Water −0.392 −0.064 0.977 0.636 1 Telephone −0.001 0.552 0.664 0.999 0.649 1 District Absenteeism 1 No. of beds 0.017 1 Electricity −0.013 0.164 1 Heat 0.051 0.388 0.861 1 Water −0.003 0.116 0.930 0.882 1 Telephone 0.055 0.293 0.882 0.942 0.901 1 Source: QSDS (health), Republic of Yemen, 2010. to pay for the care they received. And yet in country’s low population density (it has Raimah only 11 percent of patients reported some 133,000 population sites), the use of receiving a receipt listing all their fees, com- stationary health facilities is not an effective pared with 37– 68 percent among those strategy. The use of mobile outreach teams accessing facilities in other governorates. For has helped address differential utilization mobile health units, Raimah residents rates, primarily through mobile vaccination reported the shortest visits and the highest programs, and this approach should be con- percentage noted that the amount of time tinued to bridge the gap between less remote was not appropriate. and more remote areas. In addition, cul- In this study, interviews with country tural differences, particularly the mobility experts revealed that two key factors were of women, lead to differential patterns of thought to shape the delivery of health ser- seeking birthing facilities. Meanwhile, the vices: geographic accessibility and culture. home care delivery system is addressing These factors interact in important ways in these problems of distance between facili- the Republic of Yemen, particularly in the ties and the communities they are intended behavior of women seeking health care. In to serve. the north, there are fewer facilities, and The absence of any clear-cut explanation although on paper the coverage rate for for the variation in facilities, resources, and fi xed unit facilities is about 67 percent, in staff diligence in the Republic of Yemen sug- reality it is probably closer to 35 percent gests that distinct dynamics are associated because of the greater density of facilities in with each type of problem and that would-be urban areas. In addition, cultural factors in solutions to these problems must be tailored the north limit the movements of women to these idiosyncrasies. Because these without a chaperone, and so women are problems are likely to be highly context- unlikely to travel far for health care when specific, the best source of solutions is likely alone. If facilities were closer to women’s to be those professionals and others who homes, access might not be as limited, but have already managed to work within the cultural issues exacerbate the problem of prevailing constraints to find a better way of health care utilization. Because of the doing things. SUBNATIONAL VARIATION IN SER VICE DELIVERY PERFORMANCE 187 Subnational variation in health beds across facilities (mean = 7.0), whereas services delivery in Morocco two other regions barely had one (figure 7.4). Similar variation is found at the provincial This analysis uses two World Bank surveys of level (figure 7.5), with Skhirate-Témara hav- the health care sector in Morocco: the Public ing the highest average number of beds across Expenditure Tracking Survey (PETS) and the facilities (mean = 8.0), whereas several other QSDS. These surveys covered 180 health cen- provinces had only one bed per facility. At the ters (établissements de soins de santé de base, regional level, Fès-Boulemane had the highest ESSBs) across 12 of Morocco’s 16 regions and mean number of thermometers (mean = 26.7), 31 provinces/prefectures, thereby limiting the stethoscopes (6.4), (units of) blood2 (4.9), and ability to generalize to all of Morocco. In exam tables (4.4) across facilities, suggesting addition, only a specific cadre of health facili- that the capacity to provide basic resources ties was sampled in Morocco, which limits the (medical equipment) extends across the com- ability to extrapolate the findings to all health ponent elements—that is, being able to pro- care facilities. Furthermore, because these vide one resource (or not) correlates strongly facilities were purposively sampled to be in the with being able to provide others. Using these vicinity of a hospital, they likely show better items as measures of equipment availability, performance than those not as closely located at the provincial level facilities in Fès (n = 7) to a higher-tier health care setting. Annex B were the best equipped, followed (in no par- describes the Morocco PETS and data sets. ticular order) by Berkane, Agadir, Al Fida- Mers Sultan, and Skhirate-Témara. In addition to basic physical infrastructure, Variation in medical equipment health facilities must have a reliable and ade- availability quate stock of essential medical supplies such as At the regional level, Taza-Al Hoceima- aspirin. At the regional level, aspirin was avail- Taounate had the highest average number of able at all facilities in Chaouia-Ouardigha and FIGURE 7.4 Availability of medical equipment at health facilities, by region: Morocco, 2011 30 Mean number across facilities 25 20 15 10 5 0 a a ca e âa an e s uz r l l aë nè gh d at an ta ila lan ar ou Ab ao un ien r-Z ek Az lem i -D rd ét lH b - ao M ala a- ou Or Ou sa sa -T ou -A dl T as Ca m k er a- a- Ta ift uk s-B s-M em ng ui im d ns Do an Fè ao Ta ce -Z Te us Gr Ch Ho alé h- So c t-S l ke -A ba ra za ar Ra Ta M Beds Thermometers Stethoscopes Blood Exam tables Sources: PETS (health), Morocco, 2011; QSDS (health), Morocco, 2011. 188 TRUST, VOICE, AND INCENTIVES FIGURE 7.5 Availability of medical equipment at health facilities, by province/prefecture: Morocco, 2011 10 Mean number across facilities 9 8 7 6 5 4 3 2 1 0 ni ick Ta ra a- am ir Ifr s s S di Al ltan Ho uz Be ima c e ich fa El oua El ida ra â ia Je e ém a ar e M ech ar or Ou te a Se t t Ta er t t sa Ber llal Kh sset La fra s ém i -T id ba tta rir ni Fè nè Er ela Kh rad aj d an n an M r ach Fid oh ad a id Ch a-an er a ng za Ou N ad Bé ’s Al ao bl ka te S i Tiz e ou m m d én Ra ek k ch nM ce a Al y M Ag u i M K za Ja ra H ra hi Sk Ca a -H aâ Beds (mean) Thermometers (mean) Stethoscopes (mean) Blood Exam Table eb nS Aï Sources: PETS (health), Morocco, 2011; QSDS (health), Morocco, 2011. Note: The mean number of thermometers exceeds the axis range. FIGURE 7.6 Availability of aspirin, by region: Morocco, 2011 10 60 9 8 50 Number of weeks 7 % stocked out 40 6 5 30 4 3 20 2 10 1 0 0 da e âa te n ca r a t z l l aë ale an ta gh u ua ila na lan r Ab ao -D ien r-Z lem Az to di fil ou lH la- b sa a ar é ou a- Or sa -T Ta -T ou as -A ka Ou dl m Ca s er a- -M nè ift Ta sB uk a- em ng im d ns s ek Do ui Fè an us Ta -Z e Te ao M oc So Gr alé h- Ch lH c t-S ke -A ba ra za ar Ra Ta M % stocked out Average length of stock-out (weeks) Sources: PETS (health), Morocco, 2011; QSDS (health), Morocco, 2011. Taza-Al Hoceima-Taounate. At the provincial The availability of metformin (used to treat level, aspirin was available in all facilities in 12 type 2 diabetes), however, was less uniform. different provinces (see figure 7.6). Among At the regional level, many facilities were provinces where more than five facilities were stocked out. Although the percentage of surveyed, Meknès, Marrakech, Rabat, facilities without stock was lowest in Rabat- Chichaoua, Settat, and Al Hoceima had perfect Salé-Zemmour-Zaër and Taza-Al Hoceima- records of aspirin availability. Taounate (33 percent of facilities had no SUBNATIONAL VARIATION IN SER VICE DELIVERY PERFORMANCE 189 FIGURE 7.7 Availability of aspirin, by province/prefecture: Morocco, 2011 25 120 100 20 80 15 Number Percent 60 10 40 5 20 0 0 Be cei n n M ma c ne ich fa ua ra e M ech Ra ès i B Set t no t Tiz si El nit H â ra z Na ia Ja r Ta da Kh er Ou Ag fra a A ir za l e -T da a bl rk k La ate am ès ni rirt ra Je set ar ella ém di Bé aou d l El do ba er ta Er aou sa Be Sic Al Kela ar ar n ch ta Ho lta jd ad us ’ id Ch a-An ng T ja n oh F ao Kh ma di te ra an a M Ifra i ém z is To én ek k ra ch n Al Su Ou M s er M Sid M a- hi Fid ay Sk Ca -H Al aâ eb nS Aï No. of weeks stocked out (mean) No. of facilities % of facilities stocked out Sources: PETS (health), Morocco, 2011; QSDS (health), Morocco, 2011. stock), the average stock-out time was only percentages of absences among employed six weeks in Rabat-Salé-Zemmour-Zaër, staff (20 percent); absences in Tanger- compared with 28 weeks in Taza-Al Hoceima- Tétouan and Fès-Boulemane were more Taounate. At the provincial level, four than double this level (44 percent). At the provinces—Al Fida-Mers Sultan, Rabat, Sidi provincial level (figure 7.11), Jerada (located Bernoussi, and Skhirate-Témara—reported in Oriental) had the lowest percentage of all no stock-outs. However, few facilities in each staff absenteeism, 0 percent. However, that of these provinces reported. Only in Rabat province had only one facility, employing a were more than five facilities surveyed (n = 7). total of nine staff. Using a cut-off criterion Using the stock-outs in aspirin and of five facilities surveyed, Marrakech (96 metformin as measures of essential drug staff across nine facilities) had only 14 per- availability, facilities in Rabat had the best cent absenteeism, while Rabat (57 staff records of aspirin and metformin availabil- across seven facilities) had only 16 percent ity at both the regional and provincial/ absenteeism. prefectural levels (see fi gures 7.6, 7.7, 7.8, Absenteeism is a particular concern and 7.9). among doctors, the key providers of medical services, who are missing on average 42 percent of the time in Morocco. At the Variation in absenteeism regional level (figure 7.12), absences among Beyond material infrastructure and basic doctors were lowest in Rabat-Salé-Zemmour- medical supplies, health clinics need trained Zaër (21 percent) and Grand Casablanca and diligent staff. One measure of diligence (23 percent) and highest in Tadla Azilal is staff absenteeism, and in Morocco the (73 percent) and Fès-Boulemane (an astound- average rate is 27 percent (figure 7.10). This ing 81 percent). At the provincial level average, however, masks wide variation. At (figure 7.13), Casablanca (four doctors in one the regional level (figure 7.10), both Oriental facility), Jerada (two doctors in one facility), and Grand Casablanca had the lowest and Skhirate-Témara (four doctors in one 190 TRUST, VOICE, AND INCENTIVES FIGURE 7.8 Availability of metformin, by region: Morocco, 2011 30 90 80 25 % of facilities stocked out 70 Number of weeks 20 60 50 15 40 10 30 20 5 10 0 0 da an e et ha âa uz l ca l r te an ila ta aë lal Dr ou ig Ab ao na lan ien Az lem r-Z afi rd ét lH sa ou la- ab a- Or ua s-T ou -T as ou t-A ka Ta dl as aO er s-M m nè Ta s-B uk a- dC sif ng em eim ek ui Do us en Fé Ta an ao M -Z So -T oc Gr Ch alé ch lH t-S ke -A ra ba za ar Ra Ta M % of facilities stocked out Average length of stock-out (weeks) Sources: PETS (health), Morocco, 2011; QSDS (health), Morocco, 2011. FIGURE 7.9 Availability of metformin by province: Morocco, 2011 35 120 30 100 25 80 Number Percent 20 60 15 10 40 5 20 0 0 M m ir s S di Al Ha n Be ceim z a sa Ber lal ca e ich fa El ua El da ra â ia Ifr s Je e ém a Kh set La fra ar e M ch s ar or aA e ad Sid S at hi rn t -T si Ta ra Ta er t t l ni ck ta Sk i Be etta rir ni Fè nè Ho ou Er la Kh ad an n an M ach t Al ulta a- am ad te us id Ch -An er a a b ng Ou zaza Ou Nad Bé M’Si el ke ao di nj bl ka To i Tiz Ke is ou ém én Ra ek ch r ra o Fid oh Ag M Ja ra r n jd Al y M Ca Ha â- a No. of weeks stocked out (mean) No. of facilities % of facilities stocked out eb nS Aï Sources: PETS (health), Morocco, 2011; QSDS (health), Morocco, 2011. facility) each had 0 percent absenteeism nine facilities), and Rabat (30 doctors across among their employed doctors. Al Fida-Mers seven facilities)—20 percent of employed Sultan and Ben M’Sick each had only doctors were absent, performing better than 14 percent of doctors absent (each had seven the national average of 42 percent of doctors doctors across two facilities). In those prov- absent. inces in which a minimum of four facilities Even in some of the “best” regions and were surveyed—Agadir (10 doctors across provinces, however, it is clear that structural four facilities), Marrakech (25 doctors across incentives (including lack of accountability) SUBNATIONAL VARIATION IN SER VICE DELIVERY PERFORMANCE 191 conspire to enable doctors working in public FIGURE 7.10 Absenteeism among health facility staff, by region: facilities to routinely shirk their duties. It Morocco, 2011 would be highly instructive to learn how and why some facilities (and not others) are able Total to overcome these constraints. Without fur- Tanger-Tétouan ther investigation, we can only speculate— Fès-Boulemane based on our case study evidence presented in Doukkala-Abda chapter 3 and the secondary literature—that the combination of strong midlevel leadership Tadla-Azilal (which expects, requires, and encourages Taza-Al Hoceima-Taounate facility staff to do their jobs to the best of Meknès-Tafilalet their ability) and social ties (exerting local Chaouia-Ourdigha pressures and motivation via community Souss-Massa-Drâa norms and reputation mechanisms) explains Marrakech-Tensift-Al Haouz why and where absenteeism is low. Rabat-Salé-Zemmour-Zaër The correlation between absenteeism and Grand Casablanca the measure of equipment and availability was substantial but unusual, with geographic Oriental level affecting the direction of correlation 0 10 20 30 40 50 (table 7.2). For example, at the regional level, % absent of all employed staff and doctor absenteeism generally were correlated moderately negatively with other Sources: PETS (health), Morocco, 2011; QSDS (health), Morocco, 2011. FIGURE 7.11 Absenteeism among health facility staff, by province/prefecture: Morocco, 2011 0.6 16 14 0.5 12 0.4 10 Number Percent 0.3 8 6 0.2 4 0.1 2 0 0 Be a-A a M ’S a La kech e i B Su at Ou ern ltan nj i te Tiz d ém it M ara Na ès Ch rk r Ou icha ne za a én e Kh Set ra is t Al Tao set ce rt Be l Ha a i M ouz oh g l Er m ir ch i I ia Ja e El ida laâ Ta Fès TO er L ra k a A ss ra ad M A ella Be do ém ta TA ar ic c ad n M nf ar ou im ch Kh zat El fran a am ad -T n Ho uri id if Sid ers Rab ng n jd ou a Ke d ek ra lan Jer n A M ra b a- sa hi ay Fid Sk Ca -H Al aâ eb nS Aï % absent of all staff employed No. of facilities surveyed Sources: PETS (health), Morocco, 2011; QSDS (health), Morocco, 2011. 192 TRUST, VOICE, AND INCENTIVES FIGURE 7.12 Absenteeism among doctors, by region: supplies (beds, thermometers, stethoscopes, Morocco, 2011 exam tables, and blood supply). At the pro- vincial level, however, this direction reversed, TOTAL with a moderate positive correlation among Fès-Boulemane the measures. In general, supplies were posi- Tadla-Azilal tively correlated with absenteeism (but only modestly so).3 Box 7.1 describes some notable Doukkala-Abda performers in health services delivery in Chaouia-Ourdigha Morocco. Tanger-Tétouan Meknès-Tafilalet Explaining variation in health services Taza-Al Hoceima-Taounate in Morocco Oriental Marrakech-Tensift-Al Haouz In Morocco, country experts cited the gap between urban and rural areas and the origin Souss-Massa-Drâa of health care facility staff as the major fac- Grand Casablanca tors driving variation in service delivery per- Rabat -Salé-Zemmour-Zaër formance in health. Because health care 0 20 40 60 80 100 facilities in urban areas tend to be closer to % absent of all employed hospitals than the primary health centers in more remote areas, the quality of care in Sources: PETS (health), Morocco, 2011; QSDS (health), Morocco, 2011. urban areas tends to be higher. As for FIGURE 7.13 Absenteeism among doctors, by province/prefecture: Morocco, 2011 0.9 16 0.8 14 0.7 12 0.6 10 Number 0.5 Percent 8 0.4 6 0.3 4 0.2 0.1 2 0 0 Fid rat Je nfa M ém a Be s Su ra ’Si n ra ir Ch Ra ch Sidujda aou t i B An a no d ra i Na he Al T dor Kh oce nit Ou ém ma za t I te r e M kane oh ou s Al ma t Ha di El ouz én â Se fra Ta ttat Be l Jad er Er Me a ch l ia TO ès L M Aga ck La uss ra lla O ich ba ar isse am rir M Ta nè TA a- e-T rad Kh Kela ni id Be fran n M lta er ja ar d id er a E ng za F ke c i H iz i -A ek ca an bl sa Al hi ay Sk Ca -H aâ eb nS Aï % absent of all doctors employed No. of facilities surveyed Sources: PETS (health), Morocco, 2011; QSDS (health), Morocco, 2011. SUBNATIONAL VARIATION IN SER VICE DELIVERY PERFORMANCE 193 Table 7.2 Correlation between absenteeism and other measures at regional and provincial/prefectural levels, Morocco Staff Doctor No. of No. of Blood No. of exam absenteeism absenteeism No. of beds thermometers stethoscopes supply tables Region Staff absenteeism 1 Doctor absenteeism 0.769 1 Number of beds −0.404 −0.440 1 No. of thermometers −0.193 −0.042 0.572 1 No. of stethoscopes −0.353 −0.360 0.707 0.784 1 Blood supply −0.419 −0.327 0.817 0.741 0.918 1 No. of exam tables −0.505 −0.458 0.739 0.761 0.968 0.926 1 Province/prefecture Staff absenteeism 1 Doctor absenteeism 0.780 1 No. of beds 0.180 0.105 1 No. of thermometers 0.268 0.443 0.345 1 No. of stethoscopes 0.346 0.439 0.316 0.560 1 Blood supply 0.385 0.553 0.548 0.466 0.731 1 No. of exam tables 0.382 0.459 0.393 0.531 0.915 0.770 1 Sources: PETS (health), Morocco, 2011; QSDS (health), Morocco, 2011. BOX 7.1 Notable performers in health services delivery in Morocco Taking various summary indicators into consider- The region had about average levels of other ation, the following regions of Morocco were the medical equipment, although facilities in the best performers according to the 2011 PETS and region reported no beds, which may stem from QSDS: their proximity to hospitals for inpatient care. • Oriental. Twenty-two facilities were surveyed • Rabat-Salé-Zemmour-Zaire. Thirteen facili- in this region (population, 2 million). The ties were surveyed in this region (population, region had the lowest absenteeism among all 2.4 million). The region had low levels of staff (20 percent) and was lower than average doctor and all staff absenteeism and had lower in terms of doctor absenteeism (35 percent). than average percentages of facilities out of Metformin was not available in 43 percent stocks of metformin. The region performed of its facilities, and aspirin was stocked out less well in terms of equipment and aspirin in 19 percent of Oriental’s surveyed facili- availability: it was frequently stocked out ties (slightly above average). In terms of other (42 percent of facilities) and for long periods of medical equipment, facilities in the region had time (mean = 8 weeks). about average levels of equipment. • Grand Casablanca . Ten facilities were sur- veyed in this region (population, 3.6 million). At the provincial/prefectural level, the following The region had low levels of absenteeism were the stand-out performers (where at least two among both doctors (23 percent) and all staff facilities were surveyed): (20 percent). Stock-outs of metformin were around the national average of 58 percent of • R abat . Seven facilities were surveyed in facilities, and 22 percent of Grand Casablanca’s Rabat (population, approximately 620,000). fac i l it ie s were sto cked out of aspi r i n . Rabat’s facilities performed well on measures (continued next page) 194 TRUST, VOICE, AND INCENTIVES BOX 7.1 Notable performers in health services delivery in Morocco (continued) of equipment and drug availability, as well as 105,840) is home to smaller cities (popula- doctor (20 percent) and all staff (16 percent) tions ranging from 2,000 to 44,000). Despite absenteeism. Although no facilities reported the fact that only two facilities were surveyed, having beds, facilities were relatively well the levels of absenteeism were extremely supplied with blood, stethoscopes, and exam low— 0 percent among both all staff (n = 9) tables. Furthermore, all facilities reported the and doctors (n = 4). Despite these staffi ng suc- availability of both metformin and aspirin. cesses, high percentages of facilities in Jerada • Berkane . Four facilities were surveyed in were stocked out of both of the essential drugs Berkane (population of the town, approxi- investigated. mately 80,000; population of the province, • Al Fida-Mers Sultan. Two facilities were sur- 270,328). It is unclear from the data and veyed in this prefecture (population, 332,682) documentation whether the sampled facilities of Grand Casablanca. The facilities employed were all in the town or simply the province of 12 staff, 5 of whom were doctors. Al Fida- Berkane. Facilities in the province appeared to Mers Sultan was relatively well equipped, had be well stocked with thermometers and blood. both aspirin and metformin available, and had Employing across its four facilities 42 staff, lower than average percentages of absenteeism 12 of whom were doctors, it had average lev- among all employed staff (17 percent) and doc- els of doctor absenteeism and below average tors (14 percent). (21 percent versus 27 percent) levels of all • Ben M’Sick. Two facilities were surveyed in staff absenteeism. Aspirin was available in all this prefecture (population, 285,879) of Grand facilities, but half were stocked out of met- Casablanca. Absenteeism among both all staff formin, slightly below the national average of (15 employed, 14 percent absenteeism) and 58 percent. doctors (7 employed, 13 percent absenteeism) • Jerada . Only two facilities were surveyed was better than average in this prefecture. in the town of Jerada (population, approxi- Aspirin was available in both facilities, but mately 30,000). Jerada province (population, metformin was not available in either. retaining more qualified and experienced information management also tend to have staff, employing staff from the areas sur- better drug management, both of which are rounding a health facility appears to be a usually facilitated by the human resources of more successful strategy for staff retention that facility. In the Republic of Yemen, both than looking farther afield. No single histori- the quantitative and qualitative data point to cal or institutional factor was cited as a driver a diverse range of contingent factors shaping of variation in health services delivery. superior (and, for that matter, inferior) per- The financial factors, however, are less clear: formance at any given location or for any health fi nancing is not based on volume or unit of analysis: there is no obvious overlap activity or catchment, and so it is difficult to or complementarity between problems in one establish what factors related to financing domain and those in another. Even so, impor- could be contributing to variation at the sub- tant implications are that each problem needs national level. to be addressed on its own terms and that As for the factors shaping the effectiveness investments have to be made in data collec- of organizational governance procedures, tion and management to enable practitioners country experts cited the quality of informa- to identify, track, and investigate outcome tion systems. Those facilities that have better variations at the level at which supportable SUBNATIONAL VARIATION IN SER VICE DELIVERY PERFORMANCE 195 solutions can be implemented. Building a FIGURE 7.14 Adherence to sick child care protocol, counting system that can deliver incrementally higher breaths: Arab Republic of Egypt, 2010 average (and lower variability) performance requires mechanisms that give priority to 100 90 organizational learning over a quest for uni- % of clinical observations 80 versal (“best practice”) solutions. And pre- 70 cisely because there are limited resources for 60 addressing all manner of service delivery 50 problems in different communities, compre- 40 hensive data that are accessible to officials, 30 providers, and citizens alike can aid the pro- 20 10 cess of assigning priority to these challenging 0 problems (and the potentially harsh trade- l n l ed ed re ed ed M ria fia ra ta ba ou Ru d l, n form rm m rm To offs associated with each) on an informed Ur an Ru for en fo fo ex e re e and participatory basis. l, r nr Al on n, no ra ba n, The next sections turn to less detailed— Ur ra ba Ru Ur but equally instructive—examinations of variation in education and health services in Source: EHGS, 2010. Egypt and in education in the Republic of Note: “Reformed” and “nonreformed” are Egyptian classifications. Within the traditional Ministry of Health facilities, reformed facilities are those contracted by the Family Health Fund (FHF) and that Yemen, Morocco, Saudi Arabia, Oman, operate according to distinctive procedures. Nonreformed facilities are those not contracted by the Qatar, the United Arab Emirates, Bahrain, FHF or that do not operate in accordance with its procedures. Alexandria and Menoufia are the only two governorates included in the Egypt Health and Governance Study (EHGS). and Tunisia. FIGURE 7.15 Adherence to diabetes care protocol, examining feet Education and health services and legs for pulsations: Arab Republic of Egypt, 2010 in Egypt 100 As a large and populous nation, Egypt is 90 % of clinical observations bound to display considerable variation in the 80 effectiveness of its service delivery, all of 70 which can be a basis for analysis and learn- 60 ing. In health care, variation can be seen in, 50 40 among other areas, adherence to protocols. 30 For example, as part of a routine examina- 20 tion of sick children, medical staff are 10 required to check the child’s breath count 0 (figure 7.14); when examining diabetes al n l ed ed re ed ed M ria fia ra ba t ou d Ru no orm m l, n form rm To Ur an Ru efor patients, they should check the feet and legs en fo ex f re e nr Ru al, r Al on n, for pulsation (see figure 7.15). But even these ba r n, Ur ra ba supposedly uniform and simple tasks are not Ur always carried out and not everywhere. Some areas perform these standard tasks much Source: EHGS, 2010. Note: On the distinction between reformed and nonreformed facilities, see note, figure 7.14. more effectively than others. In nonreformed clinics in rural areas, for example, breath Explaining variation in health services counts are checked only 11 percent of the in Egypt time, whereas in reformed urban clinics (which are otherwise the best performers) the Key informants cited weak administrative rate is still only marginally more than institutions, including organizational frag- 50 percent. (The difference between reformed mentation (that is, systems with differing and nonreformed clinics is described in the authority and operating procedures) as note to figure 7.14.) the primary source of variation in health 196 TRUST, VOICE, AND INCENTIVES services delivery in Egypt. Even within state- reduced rate and purchasing more with owned facilities, different organizational t hose f u nds , safeg ua rd i ng aga i nst authorities are charged with overseeing dif- stock-outs. ferent regulations, implementing different requirements and guidelines, and operating Explaining variation in education in accordance with different systems of services in Egypt incentives. These entities, such as the Health I n su ra nc e Orga n i z at ion , t he Fa m i ly In Egypt, the differences in education out- Health Fund, and the traditional Ministry of comes at the subnational level follow similar Health structures, have their own manage- lines—for example, socioeconomic status rial processes and reporting relationships, and the urban-rural divide. Aside from the which result in differences in service deliv- composition of the population being served, ery. Although the average Egyptian would key informants cited budgeting processes, likely not know the difference between one training institutions, and cultural norms as facility and another, these differences are the major factors distinguishing between likely evident at the quality level. For poorly performing and well-performing edu- the user, these differences may appear in the cation facilities and subnational entities. form of the availability of qualified staff, Although the budget and allocation process opening hours, drug availability, etc. The is in principle democratic and occurs from 2010 Egypt Health and Governance Study both the bottom up and the top down, in (EHGS) report and analyses did not account practice this process does not lead to an equi- for t h is nua nced but cr ucia l fac tor. table distribution of resources. Rather, Meanwhile, of the traditional Ministry of because schools and districts must negotiate Health facilities, some are reformed and their budgets, many resource allocations are some are nonreformed (see note to figure 7.14 the result of the negotiating power and ability for an explanation of this distinction). Some of administrators, which lead to unequal out- yet not all reformed facilities are contracted comes in the distribution of resources and by the Family Health Fund, leading to educational achievement. The imbalance may another layer of regulatory and administra- be further exacerbated by possible distribu- tive differences. As noted, these various sys- tion problems as table 7.3 illustrates, drawing tems of governance result in differing levels on the 2007/08 PETS (education). of quality and service delivery, and in turn in Institutional factors also lead to variation islands of well-performing facilities because in education services delivery at the subna- there are not enough resources to apply tional level. The Higher Teacher Training improvements across the spectrum. There Institutions both drive and reinforce differ- are no major cultural, historical, ethnic, or ences in teacher quality. Because some train- religious determinants of variation because ing facilities are better than others and they Egypt is largely homogeneous in this regard. recruit teachers based on where the teachers Rather, variation results mostly from the dif- live, those in poorer areas are recruited to ferent systems of bureaucracy and quality of teach in those same disadvantaged areas. actual implementation. Another issue cited as a factor in the varia- Health facilities that are reformed or are tion in education services delivery is the operated under a number of different pilot absence of a culture of accountability. This is programs become the objects of stream- manifested in a number of ways, but a lined governance, with more resources, primary example offered by a key informant clearer reporting relationships, and often was the examination guidelines provided by better supply and drug management. For the National Center on Education and the example, reformed facilities tend to be Economy (NCEE). NCEE creates a template better equipped with medicines because for examinations (essentially an outline) of they use a revolving fund, selling drugs at a what they should cover. The template is then SUBNATIONAL VARIATION IN SER VICE DELIVERY PERFORMANCE 197 Table 7.3 Subnational variation in textbook distribution, by governorate: Arab Republic of Egypt, 2007 and 2008 Cairoa Dakhalya Ismailia Fayoum Menia Luxor Total Total textbooks, 2007/08 (PETS) n.a. 21,344,855 351,485 14,283,899 15,853,034 27,000 Total textbooks, 2007/08 (MoE, textbook sector) 48,527,809 21,079,493 4,914,197 12,490,819 19,944,434 2,136,125 109,092,877 Total students (MoE, EMIS) 1,527,520 990,189 202,340 538,975 963,533 96,439 4,318,996 Average number of textbooks per student (PETS) 21.6 1.7 26.5 16.5 0.3 Average number of textbooks per student (MoE, textbook sector) 31.8 21.3 24.3 23.2 20.7 22.2 25.3 Source: PETS (education), Egypt, 2007/08. Note: PETS = Public Expenditure Tracking Survey; MoE = Ministry of Education; EMIS = Education Management Information System; n.a. = not applicable. a. In Cairo, textbooks are sent directly from the Ministry of Education to the districts of education. adapted at the governorate level, leading to Study (TIMSS) and PETS on subnational different examinations across the country. variation in education in the Republic of Without standardization, it is difficult to Yemen enables such comparisons. For compare the situations in governorates in any example, the data reveal wide variation in meaningful way, and because the tests change overall test scores in grades 4 and 6, largely on a yearly basis there is no way to observe along an u rban-ru ral continuu m any progress within or across governorates. (figure 7.16), with students in mixed (male- This exemplifies the lack of mechanisms for female) ru ral schools scoring almost accountability and enforcement in education 20 percent lower than their urban counter- services delivery. Without even basic com- parts. Some of this variation may be attrib- mon metrics, a policy maker has no ability to utable to the quality of teachers (in urban implement a culture of accountability. settings there are twice as many teachers Managerial abilities and leadership— with the highest qualifications as in rural preventing elite capture at the local level— areas—see figure 7.17). But whatever their were also cited as a major driver of variation. level of training, teachers must be present According to one key informant, there are no and active in the classroom for actual learn- managerial inputs in terms of training or ing to occur, and here again considerable monitoring, and thus management of educa- variation was found: at the governorate level, tion is not aimed at achieving results. Schools over a quarter of teachers were regularly tend to be overstaffed with teachers but absent in Shabwah, whereas in Hadramout undermanaged. Absenteeism is therefore not this rate fell to 12 percent (figure 7.18). a major problem because teachers are present This type of preliminary analysis of sub- in high numbers, but few are actually teach- national variation could be conducted on all ing. This is also related to the heavy reliance kinds of issues, but the following are some on tutoring in Egypt. In schools that perform additional summary fi ndings. At the gover- well, it is often because of opportune leader- norate level, Hadramout is the best performer ship or a strong educational mission. The in terms of having low dropout rates headmaster is engaged in the community and (1.9 percent—it is over 10 percent elsewhere) is able to engage parents in the school envi- and failure rates (14 percent— 44 percent ronment. This engagement tends to be espe- elsewhere), but Hadramout is only midrange cially fruitful in communities that include in terms of grade repeat rates (15 percent— local entrepreneurs and benefactors. from 12.8 percent to 20.1 percent elsewhere). It pays its teachers on time at the most reli- able levels (62 percent), but also, interestingly, Education services in the it has the highest rate of teachers expressing Republic of Yemen dissatisfaction with their work environment The availability of data from the Trends in (24 percent—13 –16 percent elsewhere). International Mathematics and Science The presence of parent-teacher associations 198 TRUST, VOICE, AND INCENTIVES FIGURE 7.16 Student performance on TIMSS mathematics (grades 4 and 6), by sex and location (rural or urban): Republic of Yemen, 2011 290 Estimated average achievement based on plausible values 280 270 260 250 240 230 220 210 Grades 4 and 6, Grades 4 and 6, Grades 4 and 6, Grades 4 and 6, Grade 4 Grades 4 and 6, Grades 4 and 6, Grades 4 and 6, no Foundation no Foundation no Foundation no Foundation no Foundation no Foundation Foundation Grades Initiative, Grades Initiative, Grades Initiative, Grades Initiative, Grades Initiative, Grades Initiative, Grades Initiative mixed, rural male, rural male, urban female, rural mixed, urban female, urban Source: TIMSS, 2011. Note: Categories are based on the Yemen Explicit Stratum definition in the Trends in International Mathematics and Science Study (TIMSS). Foundation Grades Initiative refers to the program developed by the Ministry of Education in the Republic of Yemen to help students perform better in the 2011 TIMSS exercise. The program was implemented in a sample of schools to determine whether schools subject to the program would outperform comparison schools. The program aimed at addressing the unfamiliarity of pupils with test items and testing procedures, any incompetence to read, and inadequate mathematics and science teaching (Duret, Abdulmalik, and Jones 2010). in the Republic of Yemen is also highest in goals of education and therefore greater Hadramout (all but one school in the gover- acceptance of service delivery failures. In norate has one). addition, in many areas in the south of the Republic of Yemen, the lasting influence of the British occupation and its legacy of edu- Explaining variation in education cation remains, resulting in a stronger orien- services in the Republic of Yemen tation toward education when compared In the Republic of Yemen, key informants with areas in the north. Taiz, for example, argued that the existence or nonexistence of has more of a business culture and sense of a business community reinforced the quality social corporate responsibility in terms of of educational performance by shaping the building and supporting the university. culture and motivations of the individuals in Individuals who contribute to the workings that community. Areas with an active busi- of the university feel that when they invest in ness community and stronger ties to the mar- education they are investing in the well-being ket economy have higher historical, cultural, and future of the people in that region. This a nd e conom ic i ncent ive s for b et ter- attitude trickles down to both primary and performing education systems. In areas in secondary education. which this culture of entrepreneurship is not Clearly, further analysis is needed to present, there is less of a commitment to the identify more precisely where and why SUBNATIONAL VARIATION IN SER VICE DELIVERY PERFORMANCE 199 FIGURE 7.17 Level of teacher education by ISCED level, grades 4 and 6 TIMSS math scores: Republic of Yemen, 2011 100 90 80 70 60 Percent 50 40 30 20 10 0 ru n at n e4 ur ion , ru ion , ru ion rb n rb n d, tio iti io , u tio , u tio ad l ive n l l an an In at d, at ale at ale dat ra ra ra ixe nda ale nda ale da ba es nd ixe und m nd Gr , m un m un , m ou ad ou , fe ou , m ou , m Fo ive o , fe Fo ive F Gr , F ive o F at o F ive F at no ive o at no ive no 6 at , n at n iti n nd iti 6, iti 6, In 6, iti , at , iti 6 In d 6 6 4a In d In d es nd In nd es and n es an es an es 4 a ad a es 4 a es ad s 4 Gr es 4 ad 4 ad 4 ad In iti ad s Gr des ad es Gr des Gr rade Gr de Gr ad Gr rad a a a Gr Gr Gr Gr G G Did not complete ISCED level 3 ISCED level 3 ISCED level 4 ISCED level 5A, 1st Source: TIMSS, 2011. Note: TIMSS = Trends in International Mathematics and Science Study; ISCED = International Standard Classification of Education. Categories are based on the Yemen Explicit Stratum definition in the TIMSS. Foundation Grades Initiative refers to the program developed by the Ministry of Education in the Republic of Yemen to help students perform better in the 2011 TIMSS exercise. The program was implemented in a sample of schools to determine whether schools subject to the program would outperform comparison schools. The program aimed at addressing the unfamiliarity of pupils with test items and testing procedures, any incompetence to read, and inadequate mathematics and science teaching (Duret, Abdulmalik, and Jones 2010). variation occurs, but for now the central accentuated (perhaps not surprisingly) in point should be clear: variation is ubiqui- comparisons of public schools with the tous, and the usual factors that might higher-performing private schools. Within account for it only go so far. For more accu- public schools across the country, there is rate, more useful fi ndings, local researchers relatively little variation in mathematical per- a nd p ol ic y m a ker s mu st t hem s elve s formance (figure 7.19), which itself is an undertake such analyses. interesting analytical point of departure. This lack of wide variation is all the more noteworthy when one considers the wide Education services in Morocco variation in the provision of basic inputs Variation in education indicators is also evi- (availability of instructional materials, dent in Morocco, although the variation is figure 7.20) and teacher effort (absenteeism, 200 TRUST, VOICE, AND INCENTIVES figure 7.21). How this diversity in upstream Explaining variation in education resource provision and its corresponding services in Morocco importance in each setting combine to yield In Morocco, variation in service delivery is relatively similar student performance is a primarily attributed to the persistent divide topic for future study. between urban and rural areas. This divide is also correlated with and compounded by dif- FIGURE 7.18 Teacher absenteeism, by governorate: Republic of ferences in socioeconomic status, with urban Yemen, 2006 areas tending toward greater wealth and rural areas toward higher levels of poverty. 30 Culture and tradition produce variation in education services delivery in Morocco as % absent of all employed 25 well because communities that are already 20 more educated (usually urban) have a stron- ger tradition of education than less educated 15 communities (usually rural). Another differ- 10 ence possibly driving variation in education services delivery at the subnational level is 5 language. Children who speak Arabic have 0 greater exposure to and engagement with the h e ah h t ou ag da wa Arabic education system than children from ied am er ’a ab Sa od Av dr Sh Berber backgrounds. -H Ha Al Setting aside the composition of the popu- lation, education services delivery is also Source: PETS (education), Republic of Yemen, 2006. affected by the ability or inability of FIGURE 7.19 Student performance on TIMSS mathematics (grade 8), by region: Morocco, 2011 Private Meknès-Tafilalet Oued Ed-Dahab-Lagouira/Laâyoune-Boujdour-Sakia el-Hamra/Guelmim-Es Smara Marrakech-Tensift-Al Haouz Grand Casablanca Souss-Massa-Drâa Rabat-Salé-Zemmour-Zaër Fès-Boulemane Chaouia-Ouardigha/Doukkala-Abda/Tadla-Azilal Gharb-Chrarda-Beni Hssen/Tanger-Tétouan Oriental/Taza-Al Hoceima-Taounate 0 100 200 300 400 500 600 Score Source: TIMSS, 2011. Note: TIMSS = Trends in International Mathematics and Science Study. SUBNATIONAL VARIATION IN SER VICE DELIVERY PERFORMANCE 201 FIGURE 7.20 Availability of instructional materials, by region: Morocco, 2011 Private Souss-Massa-Drâa Oriental/Taza-Al Hoceima-Taounate Marrakech-Tensift-Al Haouz Rabat-Salé-Zemmour-Zaër Chaouia-Ouardigha/Doukkala-Abda/Tadla-Azilal Fès-Boulemane Oued Ed-Dha-Lagouira/Laâyoune-Boujdour Gharb-Chrarda-Beni Hssen/Tanger-Tétouan Grand Casablanca Meknès-Tafilalet 0 10 20 30 40 50 60 70 80 90 100 Percent A lot Some A little Not at all Source: TIMSS, 2011. providers to adapt their service provision to experience and have taught in a rural area the specific needs of a community and spe- for a number of years, they are entitled to cial groups. Variations in performance are request a transfer to a place of their choos- also affected by differences in the types of ing. This system therefore perpetuates the people who are attracted to education and variation in teacher quality between rural the types of people who are hired as teachers and urban areas. Quality of infrastructure and administrators. Differences in who is also leads to variation in both students and attracted to the education sector and who is teachers because insufficient or lacking hired are affected as well by location. transportation and facilities acts as a barrier Because urban areas such as Casablanca, to accessing and providing education ser- Rabat, and Fès are considered more desir- vices. These structural barriers are com- able places to work and live, and places in pounded when teachers do not live near which individuals have more educational where they work. Finally, schools located in options, there is more competitiveness rural areas generally have weaker systems among providers. This competitiveness of accountability than those in urban areas, results in a greater likelihood of competency which allows greater variation in the ser- among providers and greater experience vices provided and poor quality. A key among both teachers and administrators. informant cited a better educated, more Ultimately, it is hoped that the services they affluent, more empowered citizenry as a provide are of higher quality. factor in greater accountability in the school The civil service framework reinforces system because this factor translates into variation between urban and rural areas more i nvolve d pa rent s w it h g re at er because younger, less experienced teachers expectations. are usually assigned to more rural and Across all three countries featured remote areas. Once they have gained more here—Egypt, Morocco, and the Republic of 202 TRUST, VOICE, AND INCENTIVES FIGURE 7.21 Teacher absenteeism, by region: Morocco, 2011 Rabat-Salé-Zemmour-Zaër Grand Casablanca Meknès-Tafilalet Chaouia-Ouardigha/Doukkala-Abda/Tadla-Azilal Fès-Boulemane Marrakech-Tensift-Al Haouz Gharb-Chrarda-Beni Hssen/Tanger-Tétouan Oriental/Taza-Al Hoceima-Taounate Private Oued Ed-Dahab-Lagouira/Laâyoune- Boujdour-Sakia el Hamra/Guelmim-Es Smara Souss-Massa-Drâa 0 10 20 30 40 50 60 70 80 90 100 Percent Serious problem Moderate problem Minor problem Not a problem Source: TIMSS, 2011. Yemen—key informants argued that elite community by building a school or health capture at the local level coupled with the facility, but sometimes forcing the state to diligence of local entrepreneurs and bene- cover the operational costs once built. In factors were driving sources of variation in other instances, interviewees cited exam- the quality of education and health services ples of one headmaster, manager, or high- delivery. They also cited the urban-rural ranking Ministry of Health or Ministry of divide, regional historical and cultural dif- Education administrator at the directorate ferences (some of which stem from colonial- level who created and sustained his or her ism), and the presence and strength of a own microsystem of accountability and business community as key components of superior service delivery performance. subnational variation in service delivery These examples were often discussed in related to education. In a number of exam- terms of chance or the personalities of ples, key informants pointed to one indi- the key people involved. Less was known vidual or family who had achieved fi nancial about the specific processes used to achieve success and decided to reinvest that in their better results. SUBNATIONAL VARIATION IN SER VICE DELIVERY PERFORMANCE 203 Education services in Saudi highest in medium-size towns, suggesting Arabia, Oman, Qatar, the United that factors beyond town size per se are driv- Arab Emirates, Bahrain, and ing outcomes. Tunisia G ender d if ferences i n education outcomes between urban-rural settings are This section provides brief examples of sub- also seen in Oman (figure 7.23). Girls on national variation in education services in average consistently outperform boys and Saudi Arabia, Oman, Qatar, the United see only modest absolute declines in more Arab Emirates, Bahrain, and Tunisia. rural settings. Boys perform well in the Analysis of the TIMSS data from Saudi most urban settings, but also perform rela- Arabia enables exploration of the variations tively well in remote rural settings. In across the urban-rural divide and between Qatar, the gender and location gap is less genders (figure 7.22). On average, in Saudi pronounced (figure 7.24), with boys in small Arabia female students outperform male stu- towns outperforming boys elsewhere and dents, but this difference is largely driven by girls everywhere. location: girls do best in schools in cities and As in Morocco, student achievement in the suburbs. The performance of boys declines United Arab Emirates is relatively uniform markedly in remote rural areas, whereas it is across subnational areas (figure 7.25), and yet FIGURE 7.22 School performance, by sex and location: Saudi Arabia, 2011 450 400 350 TIMSS score on math, eighth graders 300 250 200 150 100 50 0 –50 –100 Urban Suburban Medium-size city Small town Remote rural Average Male Female Difference Source: TIMSS, 2011. Note: TIMSS = Trends in International Mathematics and Science Study. 204 TRUST, VOICE, AND INCENTIVES FIGURE 7.23 Education outcomes, by sex and location: Oman, 2011 900 800 700 TIMSS score on math, eighth graders 600 500 400 300 200 100 0 Urban Suburban Medium-size city Small town Remote rural Average Male Female Difference Source: TIMSS, 2011. Note: TIMSS = Trends in International Mathematics and Science Study. FIGURE 7.24 Education outcomes, by sex and location: Qatar, 2011 this is attained in the face of considerable 500 variation in the availability of instructional TIMSS score on math, eighth graders materials (figure 7.26) and teacher absentee- 400 ism (figure 7.27). A similar pattern emerges 300 in Bahrain: wide variation in instructional materials and teacher absenteeism (as broad 200 proxies for, respectively, basic resource 100 provision and provider efforts) nevertheless generates quite similar levels of student 0 achievement, especially among those attend- –100 ing public schools (figures 7.28, 7.29, and 7.30). an an n e y al ag ow cit ur b rb er Ur Finally, in the countries considered in er t ze bu all Av -si ot Su Sm m m this section, like those considered earlier, iu Re ed there often appears to be little relationship M Male Female Difference among variations in infrastructure, basic Source: TIMSS, 2011. materials, and provider efforts. These Note: TIMSS = Trends in International Mathematics and Science Study. inputs seem to be quite distinct realms of SUBNATIONAL VARIATION IN SER VICE DELIVERY PERFORMANCE 205 FIGURE 7.25 Student achievement scores, by region: United Arab Emirates, 2011 460 450 TIMSS score on math, eighth graders 440 430 420 410 400 390 380 Ajman West Sharjah Abu Dhabi Al Ain Umm al Al Fujayrah Ra’s Al Dubai Region Qaywayn Khaymah Source: TIMSS, 2011. Note: TIMSS = Trends in International Mathematics and Science Study. FIGURE 7.26 Availability of instructional materials, grades 4 and 8, by region: United Arab Emirates, 2011 Grades 4 and 8, Abu Dhabi Grades 4 and 8, Ajman Grades 4 and 8, Al Ain Grades 4 and 8, Al Fujayrah Grades 4 and 8, Ra´s Al Khaymah Grades 4 and 8, Sharjah Grades 4 and 8, West Region Grades 4 and 8, Umm al Qaywayn Grade 8, Abu Dhabi Grade 8, Ajman Grade 8, Al Ain Grade 8, Dubai Grade 8, Al Fujayrah Grade 8, Ra´s Al Khaymah Grade 8, Sharjah Grade 8, Umm al Qaywayn Grade 8, West Region 0 10 20 30 40 50 60 70 80 90 100 Percent A lot Some A little Not at all Source: TIMSS, 2011. 206 TRUST, VOICE, AND INCENTIVES FIGURE 7.27 Teacher absenteeism, grades 4 and 8, by region: United Arab Emirates, 2011 Grades 4 and 8, Umm al Qaywayn Grades 4 and 8, Al Fujayrah Grades 4 and 8, Abu Dhabi Grade 8, Dubai Grades 4 and 8, Ajman Grade 8, Al Ain Grade 8, Ajman Grades 4 and 8, West Region Grades 4 and 8, Sharjah Grade 8, Ra´s Al Khaymah Grades 4 and 8, Al Ain Grade 8, Abu Dhabi Grade 8, Umm al Qaywayn Grade 8, West Region Grade 8, Sharjah Grade 8, Al Fujayrah Grades 4 and 8, Ra's al Khaymah 0 10 20 30 40 50 60 70 80 90 100 Percent Serious problem Moderate problem Minor problem Not a problem Source: TIMSS, 2011. FIGURE 7.28 Student achievement scores, public and private activity in the provision of education, with schools, by region: Bahrain, 2011 performance (success or failure) in one domain being largely uncorrelated with TIMSS score on math, eighth graders 600 performance in another (see table 7.4, 500 which su m marizes the findings from 400 Tunisia4). This fi nding reinforces the gen- 300 eral conclusion that effective reform in the delivery of education services will require 200 providing specific responses to each type of 100 problem: one-size solutions will defi nitely not fit all. Put another way, if the core pol- 0 icy challenge is to increase the overall per- rn n te q l l er ra ta ra he iva rth formance of the education sector in each nt pi ar ut Pr Ca Ce uh No So - - M lic lic MENA country in order to provide students ic- ic- ic- b b bl bl Pu Pu bl Pu Pu with the skills and sensibilities required Pu Source: TIMSS, 2011. for the 21st century, there is no obvious Note: TIMSS = Trends in International Mathematics and Science Study. or clear place to start— each binding SUBNATIONAL VARIATION IN SER VICE DELIVERY PERFORMANCE 207 FIGURE 7.29 Availability of instructional materials, public and private schools, by region: Bahrain, 2011 100 90 80 70 60 Percent 50 40 30 20 10 0 te q rn l l n ra ta er ra iva he nt pi ar rth Ca ut Ce Pr uh No So ic- ic- M ic- ic- bl bl ic- Pu bl Pu bl bl Pu Pu Pu A lot Some A little Not at all Source: TIMSS, 2011. FIGURE 7.30 Teacher absenteeism, public and private schools, by region: Bahrain, 2011 100 90 80 70 60 Percent 50 40 30 20 10 0 Private Public-Capital Public-Central Public-Muharraq Public-Northern Public-Southern Serious problem Moderate problem Minor problem Not a problem Source: TIMSS, 2011. constraint problem is unique, seemingly Concluding observations unrelated to others. If there is an upside to this challenge, it is that the wide variation The central message of this chapter is that in key inputs is itself evidence that solutions service delivery performance varies widely lurk somewhere; the task going forward is within countries—and within provinces and to fi nd, examine, and learn from them. districts — even where service delivery 208 TRUST, VOICE, AND INCENTIVES Table 7.4 Correlation between inputs to education: Tunisia, 2011 Teacher late Instructional General School Job arrivals at Teacher materials supplies buildings satisfaction school absenteeism Instructional materials 1 General supplies 0.581* 1 School buildings 0.435* 0.526* 1 Job satisfaction −0.034 0.021 0.017 1 Teacher late arrivals at school −0.106 0.016 0.016 0.223* 1 Teacher absenteeism −0.164 −0.066 −0.025 0.231* 0.553* 1 Source: TIMSS, 2011. systems are centralized. This variation is to mechanisms discussed in chapter 5. Even be expected, and it emerges for a host of where the political and administrative reasons, some of them familiar (wealth) and accountability mechanisms are weak, some laudable (high qualifications), others less communities are able to achieve the high pro- obvious. We have also seen, within as well vider efforts and abilities needed to adhere to as across countries and sectors, that the per- standards. The existing forms and sources of formance characteristics of the three core variation allow identification of where an components of service delivery—physical effective response to the existing institutional infrastructure, basic supplies, and qualified/ constraints already exists. diligent staff—are frequently out of sync All this means that much of the time it will with one another: problems and strengths in be necessary to customize solutions to the one domain do not necessarily correlate prevailing problems because it cannot be with problems and strengths in others. We assumed that what works to address a have described several instances in this particular concern in a particular sector in a chapter of country contexts in which differ- particular context will work elsewhere. ent types of implementation problems Those seeking solutions to their specific prob- overlap little with each other. Even in com- lems, however, need not start from scratch; munities with the lowest quality of service they can exploit the existing forms and provision, problems pertaining to staff sources of variation to identify where an absenteeism frequently do not correlate with effective response to the existing constraints problems stemming from inadequate sup- is already available. And even if adopting and plies, meaning that strategies for responding adapting such insights itself proves difficult, to the former cannot also be assumed to be those desiring improvement can be assured responding to the latter. that things are not fated to be the way they Many of the processes driving subna- are: someone somewhere somehow has tional variation are deeply contingent on figured out a better way. It is hoped that context-specific combinations of different those facing their own seemingly intractable local and systemwide factors. These can challenges can learn from and be inspired by only be discerned through careful explora- those challenges. We will come back to these tion of local circumstances and the particu- issues when discussing possible solutions in la r st r uc t u res shapi ng how la rge chapter 11. bureaucracies function. Returning now to the cycle of perfor- In particular, subnational variation can mance, part IV will explore how perfor- manifest itself in how local leaders as well as mance affects citizens’ trust in institutions local institutions and accountability mecha- (chapter 8), and how this trust in turn nisms fill the gaps in the national political influences the nature of citizen action and ad ministrative accountabilit y (chapter 9). SUBNATIONAL VARIATION IN SER VICE DELIVERY PERFORMANCE 209 Annex A Table 7A.1 Number of health facilities, by region and province: Republic of Yemen, 2010 QSDS data set, Republic of Yemen Region District The Quantitative Service Delivery Survey Ibb (n = 29) Alfara (QSDS) data set is somewhat limited in its Almakhader coverage of the Republic of Yemen. Because Alnaderah i t c o v e r s o n l y 4 o f t h e c o u n t r y ’s Alodein 21 governorates, it is not possible to draw Alqafr conclusions about health care performance Alsadah in the country as a whole. These data, then, Alsayani should be interpreted with caution; they Alshaar provide only general insights into trends in Baadan Geblah the governorates, districts, and facilities Hobeish surveyed and how the measures under Ibb Rural investigation relate to each other. They also Mothikerah provide in-depth information about the Sabah per for ma nce of each of t he sa mpled Thi Sofal facilities. Yareem The QSDS assessing health care in the Al-Baidha (n = 14) Albaidhaa Republic of Yemen comprised three mod- Almalagem ules, including a survey of 82 health facili- Alqoraishiah ties (units, centers, and hospitals) across Alsawadiah f ou r gove r no r at e s : A l - B a id a h , I bb , Alssowmaah Raimah, and Sana’a. These four governor- Alttafah Mokairas ates were selected because of the particular Noaman challenges they face in delivering public Radman health services and because a pilot pro- Sana’a (n = 23) Alaymah Alkharigiah gram of expanded outreach services was to Alhaymah Aldakhiliah be implemented in each. Thus the analysis Alttyal based on this data set cannot be general- Arhab ized to the Republic of Yemen in its Bani Dhabian entirety. Bani Hosheish In the QSDS, the sampling frame for the Bani Mattar facility survey sought a census of all 108 Hamdan Khawlan health facilities for which the Ministry of Khawlan Alhesn Health was able to provide location data—8 Manakhah hospitals, 35 health centers, and 65 health Saafan units (wuhdah). The fi nal sample, as noted, Raimah (n = 16) Algaafariah was composed of 82 facilities (8 hospitals, Algobein 29 health centers, and 45 health units)—see Alsalafiah table 7A.1 for the number of facilities by Bilad Alttaam governorate and district. In this chapter, Kosmah the subnational variation is explored sys- Mazhar tematically by (1) briefly describing the Total 82 data; (2) identifying for each indicator Source: QSDS (health), Republic of Yemen, 2010. investigated the well-performing governor- ates (of the four) and districts within them; and (3) drawing conclusions for each indi- cator. The correlation analysis conducted at the governorate and district levels reveals 210 TRUST, VOICE, AND INCENTIVES the relationship between equipment avail- Table 7B.1 Health facilities surveyed by PETS, by ability and staff absenteeism. region and province/prefecture: Morocco, 2011 Province/ No. of Annex B Region prefecture facilities PETS and QSDS data sets, Morocco Chaouia-Ouardigha Settat 7 Doukkala-Abda El Jadida 7 The 2011 Public Expenditure Tracking Fès-Boulemane Fès 7 Survey (PETS) and Quantitative Service Grand Casablanca Aïn Sebaâ-Hay 3 Delivery Survey (QSDS) in Morocco covered Mohammadi 180 health centers or établissements de soins Al Fida-Mers Sultan 2 de santé de base (ESSBs) across 12 regions of Casablanca-Anfa 1 Morocco, spanning 31 provinces/prefec- Sidi Bernoussi 2 tures. This facility survey built on a user Ben M’Sick 2 Taza-Al Hoceima- Al Hoceima 7 survey conducted in 2009, in which users of Taounate both hospitals and health facilities were sur- Marrakech-Tensift-Al Marrakech 9 veyed on their experience at those facilities. Haouz The number of facilities surveyed varied by Al Haouz 8 region and province/prefecture in both sur- Chichaoua 7 veys, but the number nevertheless remained El Kelaâ 14 the same from 2009 to 2011 because the Meknès-Tafilalet Meknès 10 latter survey was based on the former. Errachidia 10 The sample varied at both times (by region Ifrane 3 Khénifra 11 and province) but was the same across times Oriental Berkane 4 (see table 7B.1). The 2009 sample was drawn Jerada 1 using a two-stage design. First, hospitals Nador 9 were randomly sampled with the goal of cov- Oujda 4 ering most of Morocco and all categories of Taourirt 4 hospitals. Health centers in the vicinity of Rabat-Salé-Zemmour- Rabat 7 sampled hospitals were then selected ran- Zaër domly, stratified by urban and rural areas. Khémisset 5 The 2011 PETS surveyed only those ESSBs Skhirate-Témara 1 surveyed in 2009, not hospitals. Because the Souss-Massa-Darâa Agadir 4 Ouarzazate 7 sampled ESSBs were in the vicinity of hospi- Tiznit 9 tals and urban areas, they were likely to pro- Tadla-Azilal Béni Mellal 7 vide a better picture of service delivery Tanger-Tétouan Tanger 6 performance than E S SB s th roug hout Larache 2 Morocco in general. Total 180 Similar to the analysis of the Republic of Sources: PETS (health), Morocco, 2011; QSDS (health), Morocco, 2011. Yemen, subnational variation was explored systematically by (1) briefly describing the data; (2) identifying for each indicator investigated the well-performing regions and provinces/prefectures; and (3) seeking Notes explanations for the subnational variation 1. Data sets focusing on health services delivery uncovered. The correlation analysis con- with enough geographic coverage were ana- ducted at the regional and provincial / lyzed at the lowest level of reasonable prefecture levels revealed the relationship confidence. between equipment availability and staff 2. The precise measure of “units of blood” is absenteeism. unclear. SUBNATIONAL VARIATION IN SER VICE DELIVERY PERFORMANCE 211 3. Beyond issues pertaining to variation in statis- Data sources tical “noise” and “power” at different units of analysis, one can speculate that a reason EHGS (Egypt Health and Governance Study), for this widely varying result—based on find- World Bank, http://documents.worldbank.org ings reported elsewhere in this chapter—is /c u r a t e d /e n / 2 0 10 / 0 6 / 16 332 5 45 /e g y p t that informal social accountability measures -management-service-quality-primary-health are more accurately captured at the district - care-facilities-alexandria-menoufia and provincial levels, and that such measures -governorates are more effective in addressing relational PETS (Public Expenditure Tracking Survey), issues (such as whether staff members show World Bank, ht tp: //web.worldbank.org up for work) than logistical ones (such as / W B SI T E / E X T ER NA L / TOPIC S/ E X T overcoming inadequate supplies and basic SOCIALDEVELOPMENT/EXTPCENG/0,, infrastructure). contentMDK:20507700~pagePK:148956~pi 4. Additional tables documenting these interac- PK:216618~theSitePK:410306,00.html tions in other MENA countries (which are QSDS (Quantitative Service Delivery Survey), remarkably similar to those reported for World Bank, ht tp: //web.worldbank.org Tunisia) are available upon request. / W B SI T E / E X T ER NA L / TOPIC S/ E X T POVERTY/EXTPSIA/0,,contentMDK:20467 190~isCURL:Y~menuPK:1108016~pagePK:14 Reference 8956~piPK:216618~theSitePK:490130,00 .html Duret, E., H. Abdulmalik, and S. Jones. 2010. TIMSS (Trends in International Mathematics and “Country Case Study: Yemen: Mid-term Science Study), Boston College, ht tp: // Evaluation of the EFA Fast Track Initiative.” timssandpirls.bc.edu/ http://www.camb-ed.com/fasttrackinitiative Tra nsit iona l G over na nce Projec t , ht t p: // /download/FTI_Yemen_CR(Feb2010y).pdf. transitionalgovernanceproject.org/ Citizens’ Trust and Engagement PART IV I n parts II and III, we covered the institutions and performance elements of the cycle of performance framework, explaining and further documenting the weaknesses in education and health services delivery in the Middle East and North Africa (MENA). We argued that these weaknesses are rooted in institutions, and we further explored how institutional, and especially accountability, weaknesses influence performance at the point of service delivery on average (chapter 6) and in specific local contexts (chapter 7). In part IV, we study how performance affects citizens’ perceptions of the state and citizens’ actions in dealing with the state. In particular, we seek to uncover how performance influences citizens’ trust in service providers (chapter 8), which in turn shapes the nature of citizens’ engagement at both the local and national levels (chapter 9). 213 MAP IV.1 Percentage of citizens voicing opinion to a public official in the last month in MENA, 2013 IBRD 41535 | MARCH 2015 Black Sea Ca s pi an Sea M ed it e r r an ea n Sea ATLANTIC Me SYRIAN TUNISIA MALTA di t e OCEAN r ra n e an Se a LEBANON A.R. IRAQ ISLAMIC REP. MOROCCO WEST BANK AND GAZA OF IRAN JORDAN KUWAIT A L G E R I A L I B Y A ARAB REP. BAHRAIN OF EGYPT SAUDI QATAR UNITED Percentage of citizens voicing ARABIA ARAB EMIRATES opinion to a public official in OMAN Re the last month, 2013 d Se a More than 20 16 – 20 REP. OF YEMEN 10 – 15 d en Arabian f A MENA average 15 lf o Sea Gu Less than 10 Non-MENA/non-OECD average 19 DJIBOUTI No data available OECD average 25 Data source: Gallup World Poll, 2013. Source: World Bank (IBRD 41535 | March 2015). Citizens’ Response to Poor Performance and Unresponsive 8 Institutions? Lower Trust • Citizens view the poor quality of services as an indication of the inability or unwillingness of state institutions to meet their needs. • If citizens believe that the ruling elites are either incapable or uncaring, they may lose trust in public institutions. A s we saw in the preceding chapters, citizens are frequently dissatisfied with the quality of the education, health, and other services they receive. Citizens’ trust in Schools are overcrowded and underperform- public institutions ing; doctors and nurses are frequently absent from clinics, and, when they are present, they often do not follow the standard protocols; garbage stays in the streets; and streets are unpaved. Because citizens expect the state to provide these services, they equate the fail- ures in service provision with the failure of the state. The object of their blame varies to some extent, with some finding fault with service providers or local officials and others blaming the central government. Whatever the case, poor performance undermines trust in the state. This chapter explores the link between performance and trust. It begins with a brief 215 216 TRUST, VOICE, AND INCENTIVES discussion of the role of trust in the state. It societies, Mishler and Rose (2001) found that then looks at how citizens place their blame citizens’ trust can be earned by the state on the state and the ways in which it under- responding efficiently to public priorities. mines trust in state institutions and actors Along similar lines, Yang (2013) has argued, more generally. after studying both authoritarian and demo- Many studies have recognized the impor- cratic regimes, that trust is a function of citi- tant role that political trust plays in effective zens’ perceptions of how capable their government operations and regime stability. institutions are and how committed those High levels of trust reduce government fail- institutions are to fulfilling their function. ures and enforcement and transaction costs; In line with the literature, performance- lack of trust is a breeding ground for oppor- based perceptions of how governments are tunism, informality, and free-riding, seri- responding to the needs of ordinary citizens ously compromising the effectiveness of also appear to drive trust in the Middle East public policies. Trust is positively associated and North Africa (MENA). Indeed, the link- with the support of democratic values and age between performance and trust may be political involvement (Inglehart 1990; Muller even more evident in the MENA region and Seligson 1994; Rothstein and Uslaner because its citizens, as discussed in chapter 1, 2005). Trust can also encourage civic coop- have high expectations that the government eration and the collective mobilization of citi- will provide for them and yet seem to be zens, freedom of expression and association, deeply dissatisfied with the services they and the right to pursue economic opportuni- receive and in turn blame state institutions ties, all of which are crucial for democratic for failing them. Therefore, satisfaction with sustainability and legitimacy. Authoritarian services and trust in state institutions appear regimes also seek to induce trust among the to move in parallel in the MENA region. populace as a way of asserting their legiti- macy and regime survival. They employ clientelism and the distribution of patronage Blaming the state for poor to increase vertical trust among narrow inter- services est groups or nationalism to rally support Citizens’ experiences when they visit a health from the masses (Jamal 2007; Jamal and clinic, send their children to school, try to Nooruddin 2010). turn on their lights, or try to access clean Trust is driven by individual-level judg- water affect not only their view of perfor- ments and perceptions of how the govern- mance but also their attitude toward the ment and various political actors are state. Even when they access these services performing (Hetherington 1998; Norris through private providers, they may resent 1999; Levi and Stoker 2000; Jamal 2007; having to do so. Their confidence in the state Hakhverdian and Mayne 2012). In general, may further decline if they have to turn to however, citizens evaluate their governments using wasta or informal payments to gain the based on two different sets of criteria: those services they believe they rightfully deserve. related to “input” or procedural performance Recent data from Tunisia reveal that a strong, such as the implementation of laws and regu- statistically significant correlation exists lations that guide the functioning of institu- between citizens’ unfavorable evaluations of tions, and those related to “output” or policy how well their government is providing performance and the provision of services services and their deep belief in the impor- responsive to citizens’ preferences and needs. tance of wasta and informal payments in Whenever state institutions perform poorly obtaining services (table 8.1). The belief that on these two main fronts, citizens’ trust in the government is doing a bad job in provid- institutions is likely dampened (Hakhverdian ing health services is correlated with the and Mayne 2012). In their study of the ori- belief that wasta matters in obtaining medi- gins of political trust in postcommunist cal treatment (0.53), school-related services CITIZENS’ RESPONSE TO POOR PERFORMANCE AND UNRESPONSIVE INSTITUTIONS? LOWER TRUST 217 Table 8.1 Correlation between evaluation of various services and belief in the importance of wasta and bribes: Tunisia, 2014 Wasta: medical Wasta: Wasta: Bribes: medical Bribes: Bribes: treatment schools construction treatment schools construction Evaluation: health 0.53*** 0.50*** 0.48*** 0.52** 0.51*** 0.47*** Evaluation: education 0.57*** 0.55*** 0.54*** 0.57*** 0.56*** 0.54*** Evaluation: roads 0.57*** 0.55*** 0.55*** 0.53*** 0.55*** 0.53*** Source: Transitional Governance Project, 2014. Note: The evaluation variables are categorical variables in which higher values imply worse evaluations. The wasta and bribes variables are also categorical variables in which higher values imply a stronger belief in their importance and influence in obtaining services. *p < .10 **p < .05 ***p < .01 (0.50), and construction (0.48), and with the Nonstate actors can and do take on some belief that bribes or informal payments mat- of the responsibility for providing services. At ter in medical treatment (0.52), schools times, the private sector, charities, or local (0.51), and construction (0.47). Similarly social elites can help relieve the demand for strong, statistically significant correlations state-provided services and increase citizens’ also exist between poor evaluations of the trust in state institutions. However, even government’s performance in education and when private providers deliver services, the construction and the belief in the importance state is still responsible for ensuring that the of wasta and informal payments or bribes in provision of quality services is fair and equi- obtaining services. table. When it fails to uphold this responsibil- When the state seems uncaring about the ity, allowing unqualified providers to deliver welfare of ordinary citizens, notions of fair- services based on political allegiance, ethnic ness and equity are heavily compromised. In or religious belonging, or other discrimina- some cases, the state simply excludes some tory criteria (Cammett 2014), 2 it undermines users from accessing services—one example citizens’ trust in the state as well. As one is the marginalized groups excluded by the Yemeni man from Al-Hodeidah told his focus state in the Republic of Yemen (Aslam 2014). group, local elites “control people whenever At other times, those who are unable to they provide a simple service,” while one gain privileged access or who lack informa- from Mareb complained, “We are not satis- tion (often because fees and regulations fied [with the role of elders in dispute resolu- are not posted) feel that they face higher tion] but there isn’t a state” (Aslam 2014). costs than others. For example, in Sana’a in This is not unique to the MENA region. the Republic of Yemen a woman complained Indeed, there is evidence from Europe that that “the I.D. card officially costs 300 riyals, at-risk individuals in systems with higher reli- but in reality it may cost 2,700 riyals.”1 For ance on private health provision have lower many women, for the poor, and for mem- trust in their government (Cammett, Lynch, bers of other marginalized groups, this situa- and Bilev 2014). tion puts services out of reach. As another young woman from the Republic of Yemen stated, voicing her frustration about the Lacking trust in institutions and access to health care, “Basically there are no actors health services for the people.” And in the Citizens are not simply dissatisfied with the Arab Republic of Egypt, a middle-aged state’s failure to provide services; for them, woman wryly observed about the lack of real that failure reflects the nature of the institu- options, “A public hospital is where you lose tions themselves. When they have to pay your life, . . . a private one is where you lose additional fees to put their children in the your money” (World Bank 2013). best schools, when they have to hire tutors to 218 TRUST, VOICE, AND INCENTIVES ensure that their children receive good marks, corruption, and 59 percent blamed senior when the clinic doctor asks them to visit his officials.4 Jordanians primarily blamed senior private clinic in order to undergo the medical public officials, but also held politicians, tests needed, they do not believe such scenar- lower-level public servants, and businessmen ios reflect only on the directorate, school, or responsible.5 As concluded in a 2007 World clinic; rather, these experiences create or rein- Bank report, “Tales abound of irresponsible force the notions that the state is corrupt, behavior by local governments, such as keep- incompetent, or uncaring. ing the street lights on all day, or of ‘big shots’ Citizens find the experience of dealing who are accumulating enormous water or with the bureaucracy daunting and frustrat- electricity bills and not being prosecuted. ing. They feel “constantly exposed to corrup- These stories, true or false, contribute to a tion, favoritism, poor customer service and general sense of unfairness and cynicism deficient information” and complain about about public life” (World Bank 2007, 18–19). the “crazy machine of bureaucracy” (World The relationship between perceptions of Bank 2007, 10). As a Yemeni from the focus institutions and the context governing them group in Mareb noted, “Government is very on the one hand, and citizens’ confidence and far from us.” Indeed, even local council trust in the state on the other can be observed. members echo that frustration. One council It is evidenced, for example, by the high cor- member from Al-Hodeidah complained, relation between the percentage of respon- “I exhaust all efforts in serving and helping dents who trust their national government the citizens, but I am helpless when there is and the Worldwide Governance Indicators no response.” Citizens agree; elected officials (WGI), which include underlying measures of may be closer to the people, but “they are not state institutional quality and performance the decision makers,” as a man from Taiz such as government effectiveness, rule of law, explained (Aslam 2014). and control of corruption. Table 8.2 reports Furthermore, surveys find that most people the correlations between WGI measures and believe corruption flourishes in government trust. Trust is highly correlated with political institutions and agencies. 3 A 2009 survey stability (0.9, p < .01), government effective- found that 92 percent of Egyptians believed ness (0.8, p < .01), regulatory quality (0.6, corruption had become an “undeniable part p < .10), rule of law (0.9, p < .01), and control of life” (Al- Gharini, Al-Rashidi, and of corruption (0.9, p < .01). MENA countries Al-Gamal 2009). One in four respondents in the Gulf Cooperation Council (GCC) have believed that it was more widespread in the very high levels of trust in government when central ministries; nearly one in five believed it compared with MENA non-GCC countries. prevailed at the level of localities and director- Although the high levels of trust could be ates; and two in five felt it existed equally at attributed to more than one factor, such as both levels. When it came to allocating the the nature of their political economies and blame for corruption, 61 percent of respon- composition of their populations, GCC dents stated that public servants were respon- countries score “very satisfactory” on the sible for the spread and high prevalence of Corruption Perceptions Index (as illustrated Table 8.2 Correlation between percentage of respondents who trust their national government and WGI measures: MENA region, 2013 Voice and Political Government Regulatory Rule of Control of accountability stability effectiveness quality law corruption Percentage of respondents who trust their national government 0.1964 0.8844*** 0.7685*** 0.5721* 0.8928*** 0.9071*** Sources: Gallup World Poll, 2013; Worldwide Governance Indicators (WGI), 2013. Note: MENA = Middle East and North Africa. *p < .10 **p < .05 ***p < .01 CITIZENS’ RESPONSE TO POOR PERFORMANCE AND UNRESPONSIVE INSTITUTIONS? LOWER TRUST 219 in chapter 1), have stronger institutions, and FIGURE 8.1 Perceptions of the effectiveness of governments: in general enjoy very favorable citizen evalua- MENA region, 2006–08, 2010–11 tions of their governments’ performance. Citizens’ trust can be directed toward a To what extent do you agree that the government does all it can to provide its citizens with all services? wide range of institutions, such as local gov- 100 ernment, the civil service administration, the 90 80 legislature, political parties, civil society orga- 70 nizations (CSOs), or the judicial system. Public 60 Percent 50 opinion polls reflect the low levels of trust and 40 confidence among citizens in the willingness 30 20 and ability of governments (national or local) 10 and legislatures to solve citizens’ problems. In 0 za . p. on co sia an ep q ria a 5 of the 10 countries surveyed by the Arab bi Re Ga Ira oc rd n ni ,R ge ra ba Tu ab Jo or nd en iA Al Barometer in 2010–11, nearly 50 percent of Le M Ar ka m ud Ye t, an Sa yp respondents disagreed with the statement “the tB Eg es government does all it can to provide its citi- W zens with services” (figures 8.1 and 8.2). Absolutely disagree Disagree Moreover, people believe elected parliaments Agree Very much agree are not able to enact policies that improve Sources: All countries except Morocco: Arab Barometer, 2010–11 (Wave II); Morocco: Arab Barometer, service delivery. Indeed, majorities across the 2006–08 (Wave I). Arab world do not believe that their parlia- Note: MENA = Middle East and North Africa. ment is performing well.6 Local governments are also seen as toothless and ineffective FIGURE 8.2 Perceptions of the effectiveness of parliaments: in carrying out their tasks and duties. For MENA region, 2010–11 example, in the 2010–11 Arab Barometer, To what extent do you think parliament has 42.3 percent of respondents in Iraq evaluated a role in the formation of policies? the performance of their local government as 100 bad or very bad, while only 24.4 percent evalu- 90 ated it as good or very good. 80 Similarly, citizens have little trust in the 70 60 Percent effectiveness of political parties and CSOs. 50 According to the 2010–11 Arab Barometer 40 surveys, only in Egypt and Morocco did a 30 majority of citizens believe political parties 20 cared about the needs of ordinary citizens, 10 and political parties were generally seen as 0 . ep za ia n n q a no Ira er ,R Ga more concerned about their leaders’ benefits rd g ba en Jo Al nd m Le than public welfare.7 Citizens voiced higher ka Ye an tB trust in CSOs, but there were certainly differ- es W ences in the extent to which citizens trusted different associations. Still, people were often No extent Little extent unaware of the CSOs and charities in their Medium extent Great extent own areas.8 Moreover, citizens recognized the Source: Arab Barometer, 2010–11 (Wave II). constraints that CSOs faced. For example, Note: MENA = Middle East and North Africa. fewer than one in six citizens in Egypt believed civil society had an impact on local politics state institutions. For example, Tunisia’s (Transitional Governance Project, 2012). administrative divisions (mu’atamadiya) As noted earlier for service delivery perfor- vary in the levels of trust they place in the mance, there is significant subnational varia- Constituent Assembly and the local govern- tion in the levels of trust that citizens have in ment (figures 8.3a and 8.3b). Although data 220 TRUST, VOICE, AND INCENTIVES FIGURE 8.3 Subnational variation in trust in state institutions, by region: Tunisia, 2014 a. Trust in constituent assembly 100 90 80 70 60 Percent 50 40 30 20 10 0 So s Ch uth en Fer ka eS a M uth i A e K tir be ram Ha asr id i E ah iro B rid nS o Z th ok s Bi Me ine So te ine se rth Bi Ha adh ch ouz er o a Be idou j n n Sid en us i H zah Sid Mg e l ra sr hir M llal Ka uh r Ke la Om bira zo e Sa our So a Ja ka He mel Gh ou ha Sid l M el i B elh Sid Sak Sak i i B iet en M d Ea t az st Sb a Ou Je ba As a Th r ala Ba El A f ’ l M ou Bo ateu ke zi zi a rte Fah M arzi M uA in ian Dj El M rgu b mr un led lm Al assin Ez ran ua ard laa aj ar mh N am i E hi ou ou Sid L nas El Aro i ui ou Ez ss all nc i Ka Bec e eb He us No n ze gr ba ur uh o r M ff m zo Ri lF o e lh o M E B r ze Bi ed let Ka M ba se Sa us So b. Trust in local government 100 90 80 70 60 Percent 50 40 30 20 10 0 ika Fe ahs ou a m j El sral l Bi M lah Ba her h So a Le uika M m Be egri a n ne ze aff s So z Sid Ma th as r on e iro S stir t h uh it Bi M’ ajla Ez No n K uh h Sid asr our l B lal M u Ba ir en d A o er e S er Ha M uth Ka i El un Ke rid Z ra Sid um is o l i ok e Gh J ne So r El lma Sid M Ria h i B azz dh zid na El East i A He ra lh ha Sb id Th a ala n af Omu Al e i B he Ja rou i H teu Bi H rou rte ou M an b gu M hir ib M sin M ran ed le rd c za kie ut rte ke zo rt se el h Bo Ezz Bo arz Na m am bi Am ua aou Dj in sk Al Kra u i E Hel ou ou be nc ni ss ido ec laa Fe eb a us M F Sa So ba o r M en El ri ze Sa a e g A Ch tO Sid Ka le ba se Sa us So Very high High Average Low Very low Source: Transitional Governance Project, 2014. limitations do not allow us to attribute this health services and their perceptions of the wide variation in trust to service delivery per- pervasiveness of corruption within their gov- formance by the state in each division, the evi- ernment and state institutions. Estimating a dence is complementary to our earlier analysis binary logit model9 (which includes the demo- and suggests that state performance and sub- graphic variables likely to affect trust) reveals sequently trust in it are not uniform within that the probability of trusting the national countries. government is significantly higher among Empirical analyses, relying on perceptions- respondents who are satisfied with education based data from the Gallup World Poll (2013) services and the availability of quality health and the Arab Barometer (2010–11), lend addi- care and who believe that corruption is not tional support to the notion that performance pervasive within their government. For exam- legitimacy in the eyes of citizens affects trust ple, in Egypt the predicted probability of levels. Based on data from the 2013 Gallup trusting the national government is 77 percent World Poll, trust in national government among those who are satisfied with education seems to be highly associated with citizens’ services compared with 67 percent among satisfaction with education and those who are not satisfied. The results are CITIZENS’ RESPONSE TO POOR PERFORMANCE AND UNRESPONSIVE INSTITUTIONS? LOWER TRUST 221 statistically significant for other MENA econ- respondents believe that corruption is wide- omies such as Iraq, the Republic of Yemen, spread, whereas it is 73 percent among those Tunisia, and West Bank and Gaza. The results who believe it is not widespread (see figure 8.4 seem to be quite similar when respondents are and tables 8A.1, 8A.2, and 8A.3 in the satisfied with health services, where the prob- annex). ability of trusting the government is statisti- Using trust in the judiciary as a second cally significantly higher for those who are measure of trust in state institutions, the satisfied with health services in the five sam- results hold and, as with the analysis using pled MENA countries. There are similar trust in national government as a measure results regarding belief in the pervasiveness of for state institutions, the magnitude appears corruption, but with a larger magnitude. For larger for belief in pervasiveness of corrup- example, the probability of trusting the tion. For example, in Tunisia the probability national government is only 40 percent when of trusting the judiciary is 61 percent FIGURE 8.4 Predicted probability of trusting the national government as a function of satisfaction with education and health services and believing that corruption is pervasive: Selected MENA economies, 2013 a. Satisfaction with education services b. Satisfaction with availability of quality health care Adjusted prediction with 95% Cls Adjusted prediction with 95% Cls 0.8 0.8 Pr (trust in national government) Pr (trust in national government) 0.7 0.7 0.6 0.6 0.5 0.5 0.4 0.4 0.3 0.3 Not Satisfied Not Satisfied Satisfaction with education services Satisfaction with availability of Satisfied Satisfied quality health care c. Belief that corruption is pervasive within the government Adjusted prediction with 95% Cls 1.0 Pr (trust in national government) 0.8 0.6 0.4 0.2 Not Belief in pervasiveness of corruption Pervasive pervasive Egypt, Arab Rep. Tunisia West Bank and Gaza Iraq Yemen, Rep. Source: Gallup World Poll, 2013. Note: CI = confidence interval; MENA = Middle East and North Africa; Pr = probability. 222 TRUST, VOICE, AND INCENTIVES when respondents are satisfied with health As expected, perceptions of pervasiveness services, compared with 53 percent when of corruption have a remarkably strong and they are not. The magnitude (that is, the dif- significant association with trust in govern- ference in probabilities) seems higher for the ment. In probing more on the elements of belief in the pervasiveness of corruption, corruption, the Arab Barometer asked in a as the probability of trusting the judiciary is categorical variable about whether respon- 71 percent when respondents do not believe dents believed that qualifications were more that corruption is widespread, whereas it important than connections in obtaining drops to 53 percent when they believe it is employment, as important as connections, or (see figure 8.5 and tables 8A.4, 8A.5, and less important. Trust in government is a cat- 8A.6 in the annex). egorical variable that ranges from trusting FIGURE 8.5 Predicted probability of trusting the judiciary as a function of satisfaction with education and health services and believing that corruption is widespread: Selected MENA economies, 2013 a. Satisfaction with education services b. Satisfaction with availability of quality health care Adjusted predictions with 95% Cls Adjusted predictions with 95% Cls 1.0 1.0 0.8 0.8 Pr (trust in the judiciary) Pr (trust in the judiciary) 0.6 0.6 0.4 0.4 0.2 0.2 Not Satisfaction with education services Satisfied Not Satisfaction with availabitiy of Satisfied satisfied satisfied quality health care c. Belief in pervasiveness of corruption Adjusted predictions with 95% Cls 1.0 0.8 Pr (trust in the judiciary) 0.6 0.4 0.2 Not Belief in pervasiveness of corruption Pervasive pervasive Egypt, Arab Rep. Tunisia Iraq West Bank and Gaza Syrian Arab Republic Yemen, Rep. Source: Gallup World Poll, 2013. Note: CI = confidence interval; MENA = Middle East and North Africa; Pr = probability. CITIZENS’ RESPONSE TO POOR PERFORMANCE AND UNRESPONSIVE INSTITUTIONS? LOWER TRUST 223 the government to a great extent to absolutely FIGURE 8.6 Predicted probability of trusting the government not trusting it. Estimating a multinomial to a great extent compared with absolutely not trusting it as a logit model10 reveals that the probability of function of perceptions of the importance of connections in trusting the national government to “a great obtaining employment: Selected MENA countries, 2010–11 extent” compared with “absolutely not Adjusted predictions with 95% Cls trusting it” increases significantly when 1.0 respondents believe that qualifications Pr (trust government to a great extent) are more important than connections and is 0.8 higher than the probability when respon- dents believe that connections are more 0.6 important than qualifications. This effect tends to be particularly strong for countries 0.4 such as Egypt, Jordan, and Saudi Arabia (see figure 8.6 and table 8A.7 in the annex). For 0.2 example, in Saudi Arabia, the probability of trusting the government is 41 percent when respondents think that connections are more 0 important than qualifications. It then Connections Connections as Qualifications more important more increases to 58 percent when they believe it is important important as important, and then jumps to 71 percent Importance of connections versus when they believe it is less important than qualifications in obtaining employment qualifications. The impact of the importance Saudi Arabia Egypt, Arab Rep. Jordan of tribal affinities on trust in national govern- Lebanon Yemen, Rep. Algeria ment was found to be very modest, with little Iraq or no impact on trust, suggesting that in many cases citizens do not view tribal affini- Source: Arab Barometer, 2010–11 (Wave II). Note: Reference category for the dependent variable is “absolutely do not trust the national ties in a negative light. government.” CI = confidence interval; MENA = Middle East and North Africa; Pr = probability. These results are similar to fi ndings else- where that corruption and perceptions of there may be some endogeneity between citi- government performance in service delivery zens’ evaluations and trust—that is, those in the MENA region are significantly associ- who do not trust the government are more ated with lower trust (Rothstein 2011). The likely to view its performance negatively, impact of political corruption on erosion of and they also may be more likely to engage the trust of citizens has been supported in corrupt practices, which in turn under- empirically by various regional barometers. mines trust in government and its institu- Using data on perceptions of corruption tions (Cleary and Stokes 2006; Morris and and trust in state institutions from the Klesner 2010). Furthermore, perceptions of Eurobarometer, Della Porta (2000) found an corruption and service delivery are subjec- inverse relationship between corruption and tive and affected by many factors. For exam- trust. Similarly, Chang and Chu (2006) used ple, they may be picking up some of the the Asian Barometer to test whether trust was effects of citizens’ unhappiness with their more likely to be eroded by corruption, and country’s broader economic situation and they found a strong significant impact across the overall political performance of the gov- the five Asian countries in their sample. ernment such as perceived fairness, freedom, Although this correlation between per- and satisfaction with the democratic process ceptions of service quality and trust in gov- (Chang and Chu 2006). Measuring actual ernment may not appear to be an absolute government performance in service delivery indication that poor quality service delivery objectively rather than relying on percep- undermines trust, there are reasons to tions to understand its actual impact on believe this relationship holds. Certainly, trust would be ideal, but it is empirically 224 TRUST, VOICE, AND INCENTIVES difficult (Yang and Holzer 2006). Instead, of government’s efforts to provide services in we remind readers of the earlier findings a fair and equitable manner. that individuals’ perceptions of service deliv- Citizens cite the need for connections and ery vary to some extent across services, even informal payments to access services, the when their evaluation of and engagement widening inequities, and the fact that the with the state remain the same. This sug- most vulnerable segments have been left gests that citizens’ perceptions do mirror uncovered. Their trust levels are also heavily reality to an extent and gives us confidence affected by their perceptions of the institu- that the quality of service delivery contrib- tions governing them. They find dealing with utes to citizens’ low trust in the state. the bureaucracy daunting, and they fi nd the ministries crippled with corruption, staffed by unresponsive and undermotivated civil Conclusion servants. The evidence largely suggests that citizens’ The strategies long deployed by the states trust in public institutions in the MENA with deep colonial legacies such as clientelism countries is similar to what has been sug- and patronage have gained the vertical trust gested in the literature and found in other of a narrow set of groups, but they have regions. Their trust in the state is a function alienated the populace in general, resulting in of their perceptions of the quality of the inequitable social institutions and a largely services offered, as well as their evaluations dissatisfied population. CITIZENS’ RESPONSE TO POOR PERFORMANCE AND UNRESPONSIVE INSTITUTIONS? LOWER TRUST 225 Annex Predicted probabilities tables on trusting the national government Table 8A.1 Predicted probabilities of satisfaction with education services on trusting the national government: Selected MENA economies, 2013 Predicted probability on trusting Independent variable the national government 0 = not satisfied, Arab Republic of Egypt 0.665*** (0.00815) 0 = not satisfied, West Bank and Gaza 0.406*** (0.0142) 0 = not satisfied, Iraq 0.361*** (0.0136) 0 = not satisfied, Tunisia 0.412*** (0.0114) 0 = not satisfied, Republic of Yemen 0.339*** (0.0190) 1 = satisfied, Arab Republic of Egypt 0.771*** (0.00708) 1 = satisfied, West Bank and Gaza 0.537*** (0.0139) 1 = satisfied, Iraq 0.490*** (0.0148) 1 = satisfied, Tunisia 0.543*** (0.0114) 1 = satisfied, Republic of Yemen 0.465*** (0.0209) No. of observations 16,812 Source: Gallup World Poll, 2013. Note: All predictors are at their mean value. Standard errors are in parentheses. *p < .10 **p < .05 ***p < .01 Table 8A.2 Predicted probabilities of satisfaction with availability of quality health care on trusting the national government: Selected MENA economies, 2013 Predicted probability on trusting Independent variable the national government 0 = not satisfied, Arab Republic of Egypt 0.661*** (0.00840) 0 = not satisfied, West Bank and Gaza 0.408*** (0.0143) 0 = not satisfied, Iraq 0.358*** (0.0134) 0 = not satisfied, Tunisia 0.408*** (0.0116) 0 = not satisfied, Republic of Yemen 0.337*** (0.0184) (continued next page) 226 TRUST, VOICE, AND INCENTIVES Table 8A.2 Predicted probabilities of satisfaction with availability of quality health care on trusting the national government: Selected MENA economies, 2013 (continued) Predicted probability on trusting Independent variable the national government 1 = satisfied, Arab Republic of Egypt 0.749*** (0.00842) 1 = satisfied, West Bank and Gaza 0.513*** (0.0150) 1 = satisfied, Iraq 0.461*** (0.0158) 1 = satisfied, Tunisia 0.514*** (0.0125) 1 = satisfied, Republic of Yemen 0.438*** (0.0219) No. of observations 10,821 Source: Gallup World Poll, 2013. Note: All predictors are at their mean value. Standard errors are in parentheses. *p < .1 **p < .05 ***p < .01 Table 8A.3 Predicted probabilities of belief in pervasiveness of corruption within government on trusting the national government: Selected MENA economies, 2013 Predicted probability on trusting Independent variable the national government 0 = not pervasive, Arab Republic of Egypt 0.887*** −0.00758 0 = not pervasive, West Bank and Gaza 0.734*** −0.0155 0 = not pervasive, Iraq 0.691*** −0.0172 0 = not pervasive, Tunisia 0.735*** −0.0154 0 = not pervasive, Republic of Yemen 0.671*** −0.0226 1 = pervasive, Arab Republic of Egypt 0.658*** −0.00743 1 = pervasive, West Bank and Gaza 0.404*** −0.0136 1 = pervasive, Iraq 0.355*** −0.0131 1 = pervasive, Tunisia 0.405*** −0.0103 1 = pervasive, Republic of Yemen 0.334*** −0.0184 No. of observations 10,821 Source: Gallup World Poll, 2013. Note: All predictors are at their mean value. Standard errors are in parentheses. *p < .1 **p < .05 ***p < .01 CITIZENS’ RESPONSE TO POOR PERFORMANCE AND UNRESPONSIVE INSTITUTIONS? LOWER TRUST 227 On trusting the judiciary Table 8A.4 Predicted probabilities of satisfaction with education services on trusting the judiciary: Selected MENA economies, 2013 Predicted probability on Independent variable trusting the judiciary 0 = not satisfied, Arab Republic of Egypt 0.748*** (0.00730) 0 = not satisfied, Syrian Arab Republic 0.339*** (0.0158) 0 = not satisfied, West Bank and Gaza 0.377*** (0.0105) 0 = not satisfied, Iraq 0.437*** (0.0120) 0 = not satisfied, Tunisia 0.471*** (0.0117) 0 = not satisfied, Republic of Yemen 0.247*** (0.00976) 1 = satisfied, Arab Republic of Egypt 0.859*** (0.00543) 1 = satisfied, Syrian Arab Republic 0.514*** (0.0172) 1 = satisfied, West Bank and Gaza 0.555*** (0.0102) 1 = satisfied, Iraq 0.615*** (0.0117) 1 = satisfied, Tunisia 0.648*** (0.0108) 1 = satisfied, Republic of Yemen 0.403*** (0.0125) No. of observations 15,472 Source: Gallup World Poll, 2013. Note: All predictors are at their mean value. Standard errors are in parentheses. *p < .1 **p < .05 ***p < .01 Table 8A.5 Predicted probabilities of satisfaction with availability of quality health care on trusting the judiciary: Selected MENA economies, 2013 Predicted probability on Independent variable trusting the judiciary 0 = not satisfied, Arab Republic of Egypt 0.786*** (0.00666) 0 = not satisfied, Syrian Arab Republic 0.389*** (0.0163) 0 = not satisfied, West Bank and Gaza 0.429*** (0.0106) 0 = not satisfied, Iraq 0.490*** (0.0117) 0 = not satisfied, Tunisia 0.525*** (0.0116) 0 = not satisfied, Republic of Yemen 0.289*** (0.0101) (continued next page) 228 TRUST, VOICE, AND INCENTIVES Table 8A.5 Predicted probabilities of satisfaction with availability of quality health care on trusting the judiciary: Selected MENA economies, 2013 (continued) Predicted probability on Independent variable trusting the judiciary 1 = satisfied, Arab Republic of Egypt 0.838*** (0.00610) 1 = satisfied, Syrian Arab Republic 0.473*** (0.0179) 1 = satisfied, West Bank and Gaza 0.514*** (0.0107) 1 = satisfied, Iraq 0.575*** (0.0127) 1 = satisfied, Tunisia 0.609*** (0.0115) 1 = satisfied, Republic of Yemen 0.364*** (0.0130) No. of observations 15,472 Source: Gallup World Poll, 2013. Note: All predictors are at their mean value. Standard errors are in parentheses. *p < .1 **p < .05 ***p < .01 Table 8A.6 Predicted probabilities of belief in pervasiveness of corruption within government on trusting the judiciary: Selected MENA economies, 2013 Predicted probability on Independent variable trusting the judiciary 0 = not pervasive, Arab Republic of Egypt 0.889*** (0.00602) 0 = not pervasive, Syrian Arab Republic 0.582*** (0.0187) 0 = not pervasive, West Bank and Gaza 0.621*** (0.0132) 0 = not pervasive, Iraq 0.677*** (0.0136) 0 = not pervasive, Tunisia 0.707*** (0.0134) 0 = not pervasive, Republic of Yemen 0.470*** (0.0163) 1 = pervasive, Arab Republic of Egypt 0.790*** (0.00613) 1 = pervasive, Syrian Arab Republic 0.395*** (0.0163) 1 = pervasive, West Bank and Gaza 0.434*** (0.00983) 1 = pervasive, Iraq 0.496*** (0.0115) 1 = pervasive, Tunisia 0.531*** (0.0107) 1 = pervasive, Republic of Yemen 0.294*** (0.0102) No. of observations 15,472 Source: Gallup World Poll, 2013. Note: All predictors are at their mean value. Standard errors are in parentheses. *p < .1 **p < .05 ***p < .01 CITIZENS’ RESPONSE TO POOR PERFORMANCE AND UNRESPONSIVE INSTITUTIONS? LOWER TRUST 229 On trusting the government to a great extent Table 8A.7 Predicted probabilities of perceptions of importance of connections in obtaining employment on trusting the government to a great extent compared with absolutely not trusting it: Selected MENA countries, 2010–11 Predicted probability on trusting the Independent variable government to a great extent 1 = connections are more important than qualifications, Algeria 0.0289*** (0.00428) 1 = connections are more important than qualifications, Arab Republic of Egypt 0.322*** (0.0176) 1 = connections are more important than qualifications, Iraq 0.0266*** (0.00392) 1 = connections are more important than qualifications, Jordan 0.215*** (0.0136) 1 = connections are more important than qualifications, Lebanon 0.0517*** (0.00613) 1 = connections are more important than qualifications, Saudi Arabia 0.409*** (0.0195) 1 = connections are more important than qualifications, Republic of Yemen 0.0416*** (0.00570) 2 = connections are as important as qualifications, Algeria 0.0841*** (0.0116) 2 = connections are as important as qualifications, Arab Republic of Egypt 0.497*** (0.0219) 2 = connections are as important as qualifications, Iraq 0.0695*** (0.00975) 2 = connections are as important as qualifications, Jordan 0.357*** (0.0199) 2 = connections are as important as qualifications, Lebanon 0.161*** (0.0187) 2 = connections are as important as qualifications, Saudi Arabia 0.576*** (0.0204) 2 = connections are as important as qualifications, Republic of Yemen 0.114*** (0.0145) 3 = qualifications are more important than connections, Algeria 0.125*** (0.0204) 3 = qualifications are more important than connections, Arab Republic of Egypt 0.647*** (0.0318) 3 = qualifications are more important than connections, Iraq 0.110*** (0.0188) 3 = qualifications are more important than connections, Jordan 0.504*** (0.0351) 3 = qualifications are more important than connections, Lebanon 0.241*** (0.0337) 3 = qualifications are more important than connections, Saudi Arabia 0.704*** (0.0290) 3 = qualifications are more important than connections, Republic of Yemen 0.169*** (0.0266) No. of observations 7,311 Source: Arab Barometer, 2010–11 (Wave II). Note: All predictors are at their mean value. Standard errors are in parentheses. *p < .1 **p < .05 ***p < .01 230 TRUST, VOICE, AND INCENTIVES Notes (2011); Jordan, 49 percent (2010); Lebanon, 15 percent (2010); Libya, 30 percent (2013); 1. Member of a focus group convened for the Morocco, 9 percent (2007); Republic of social development study conducted by the Yemen, 21 percent (2011); Tunisia, 8 percent World Bank (2007). (2012); and West Bank and Gaza, 39 percent 2. Also see Cammett and MacClean (2014) for (2010). In Libya, only one in seven citizens a more general discussion of how the charac- believed their parliament had taken the teristics of nonstate providers—including the right steps to fight corruption (Transitional degree of formalization, degree of embedded- Governance Project, 2013). ness/locus of operation, profit orientation, 7. In probing citizens’ trust of political parties, and eligibility criteria—combine with the the Arab Barometer (2010–11) used a four- characteristics of private provider-state rela- scale categorical variable: trust them to a great tions to affect the effects of nonstate provi- extent, trust them to medium extent, trust sion on citizens’ trust. them to a limited extent, and absolutely do 3. In the 2010–11 Arab Barometer, the percent- not trust them. The following responded that ages of respondents answering yes to the ques- they absolutely do not trust political parties in tion “Do you think that there is corruption their economies: Algeria, 53.4 percent; Egypt, within the state’s institutions and agencies?” 41.1 percent; Iraq, 52.4 percent; Jordan, were as follows: Algeria, 93 percent (n = 1,178, 41.6 percent; Lebanon, 60 percent; Republic 2011); Egypt, 82 percent (n = 1,193, 2011); of Yemen, 46.2 percent; Tunisia, 48.8 per- Iraq, 97 percent (n = 1,215, 2011); Jordan, cent; and West Bank and Gaza, 52.7 percent. 74 percent (n = 1,065, 2010); Lebanon, 8. The Gallup World Poll (2013) reported that the 97 percent, (n = 1,379, 2011); Republic of following were unaware of social and nongov- Yemen, 94 percent (n = 1,137, 2011); Tunisia, ernmental organizations in their areas that offer 79 percent (n = 1,047, 2011); and West Bank people opportunities to serve the community by and Gaza, 77 percent (n = 1,054, 2010). In volunteering their time: Algeria, 55.8 percent; two-thirds of the countries surveyed by the Bahrain, 19.8 percent; Egypt, 60 percent; Iraq, Arab Barometer, less than 15 percent of respon- 67.6 percent; Jordan, 49.6 percent; Kuwait, 43 dents believed that the government did a great percent; Lebanon, 44.2 percent; Morocco, deal to eliminate corruption. In Algeria, Egypt, 56.8 percent; Oman, 43.7 percent; Qatar, and Tunisia, more than two-thirds believed 37 percent; Republic of Yemen, 81.6 percent; that corruption was widespread or very wide- Saudi Arabia, 26.9 percent; Syrian Arab spread within the government. Even in coun- Republic, 68.6 percent; Tunisia, 66.9 percent; tries that are experiencing a transition—Egypt United Arab Emirates, 35.6 percent; and and Tunisia—citizens do not believe the gov- West Bank and Gaza, 62.2 percent. ernment handles corruption well. Only in those 9. The binary logit model included trust in two countries when surveyed in 2011 did a national government (a binary variable coded majority of respondents believe the government 1 if yes and 0 if no) as a dependent variable and was working to eliminate corruption to a great the following explanatory variables: (1) satis- or medium extent. faction with education (a binary variable coded 4. Another 26 percent blamed citizens themselves, 1 if yes and 0 if no); (2) satisfaction with the 16 percent blamed municipalities, 15 percent availability of quality health care (a binary blamed businessmen and the private sector, variable coded 1 if yes and 0 if no); and (3) a and 14 percent pointed to the police (see belief in the pervasiveness of corruption within Al-Gharini, Al-Rashidi, and Al-Gamal 2009). the government (a binary variable coded 1 if 5. Results of a 2007 Jordanian public opinion yes and 0 if no), in addition to age, gender, poll from the Center for Strategic Studies at income quintile, and employment status. the University of Jordan, cited in Al-Gharini, 10. The multinomial logit model included as a Al-Rashidi, and Al-Gamal (2009). dependent variable trust in national govern- 6. Specifically, when asked by the Arab ment (a categorical variable coded as 1 = trust Barometer how they would evaluate the per- the government to a great extent; 2 = trust the formance of their parliament in carrying out government to a medium extent; 3 = trust the its tasks and duties, the percentage that government to a limited extent; 4 = abso- answered “good” or “very good” was in lutely do not trust the government). It also Algeria, 6 percent (2011); Iraq, 23 percent included the following explanatory variables: CITIZENS’ RESPONSE TO POOR PERFORMANCE AND UNRESPONSIVE INSTITUTIONS? LOWER TRUST 231 (1) the role of connections in gaining employ- Asian Democracies?” Journal of Politics ment (a categorical variable coded as 1 = 68 (2): 259–71. connections are more important than qualifi- Cleary, M. R., and S. C. Stokes. 2006. Democracy cations in gaining employment; 2 = connec- and the Culture of Skepticism: Political Trust tions are as important as qualifications in in Argentina and Mexico. New York: Russell gaining employment; 3 = qualifications are Sage. more important than connections in gaining Della Porta, D. 2000. “Social Capital, Beliefs in employment); (2) the role of tribal affinity in Government, and Political Corruption.” In gaining employment (a categorical variable Disaffected Democracies: What’s Troubling coded as 1 = tribal affinities are more impor- the Trilateral Countries? edited by Susan Pharr tant than qualifications in gaining employ- and Robert Putnam. Princeton, NJ: Princeton ment; 2 = tribal affinities are as important as University Press. qualifications in gaining employment; 3 = H a k hverd ia n , A ., a nd Q. M ay ne. 2012 . qualifications are more important than tribal “ I n s t it ut ion a l Tr u s t , E du c at ion , a nd affinities in gaining employment); (3) age; Corruption: A Micro-Macro Interactive (4) educational attainment; (5) gender; and Approach.” Journal of Politics 74 (3): 739–50. (6) urban/rural status. Hetherington, M. J. 1998. “The Political Relevance of Political Trust.” American Political Science Review 92 (4): 791–808. References Inglehart, R. F. 1990. Culture Shift in Advanced A l- Gharini, K., I nas A . A l-Rashidi, and Industrial Society. Princeton, NJ: Princeton M. Al-Gamal. 2009. “The Report of the University Press. National Survey of the Citizens’ Opinions on Jamal, A. 2007. “When Is Social Trust a Desirable Corruption, the Judicial System and the Outcome? Examining Levels of Trust in the Quality of Governmental Services in Egypt.” Arab World.” Comparative Political Studies Research, Monitoring and Governance Unit, Social 40 (11): 1328–49. Contract Center, Washington, DC. http:// Jamal, A., and I. Nooruddin. 2010. “The siteresources.worldbank.org/INTLAWJUSTINST Democratic Utility of Trust: A Cross-National / Resources/Egypt_Corruption_JudicialSystem Analysis.” Journal of Politics 72 (1): 45–59. _Government_Services_Report.pdf. Levi, M., and L. Stoker. 2000. “Political Trust Aslam, G. 2014. “Notes from Fieldwork in Yemen and Trustworthiness.” Annual Review of Conducted for SA Flagship.” Unpublished raw Political Science 3: 475–507. data, World Bank, Washington, DC. Mishler, W. , and R. Rose. 2001. “What Are the C a m m e t t , M . 2 0 14 . C o m p a s s i o n a t e Origins of Political Trust? Testing Institutional Communalism: Welfare and Sectarianism in and Cultural Theories in Post-communist Lebanon. Ithaca, NY: Cornell University Press. Societies.” Comparative Political Studies 34 Cammett, M., J. Lynch, and G. Bilev. 2014. (1): 30–62. “Does Non-State Provision of Social Services Morris, S. D., and J. L. Klesner. 2010. “Corruption Promote or Undermine Citizen Trust in and Trust: Theoretical Considerations and Government? The Case of Health Care in Evidence from Mexico.” Comparative Political Europe.” Unpublished manuscript. https:// Studies 43 (10): 1258–85. www.researchgate.net/publication/254251158 Muller, E., and M. Seligson. 1994. “Civic Culture _ Does _ Non- St ate _ Prov ision _ of _ S o cia l and Democracy: The Question of Causal _ Services_Promote_or_Undermine_Citizen Relationship.” American Journal of Political _Trust_in_Government_The_Case_of _Health Science 88: 635–52. _Care_in_Europe. Norris, P., ed. 1999. Critical Citizens: Global C a m me t t , M ., a nd L . M ac C le a n. 2014. Support for Democratic Government. New “Introduction: The Political Consequences of York: Oxford University Press. Non-State Social Welfare in the Global Rothstein, B. 2011. The Quality of Government: South.” In Compassionate Communalism: Corruption, Social Trust, and Inequality in Welfare and Sectarianism in Lebanon, edited International Perspective. Chicago: University by Melani Cammett. Ithaca, NY: Cornell of Chicago Press. University Press. Rothstein, B., and E. M. Uslaner. 2005. “All for Cha ng, E . C . C ., a nd Y.-h. Chu. 20 06. One: Equality, Corruption, and Social Trust.” “Corruption and Trust: Exceptionalism in World Politics 58 (1): 41–72. 232 TRUST, VOICE, AND INCENTIVES World Bank. 2007. Decentralization and Local Data sources Governance in MENA: A Survey of Policies, Institutions, and Practices. A Review of A rab B ., 20 06 – 08 ( Wave I), ht t p: //w w w Decentralization Experiences in Eight Middle .arabbarometer.org/content East and North Africa Countries. Washington, /arab-barometer-i DC: World Bank. A rab B ., 2010 –11 (Wave I I), ht tp: //w w w ———. 2013. Fairness and Accountability: .arabbarometer.org/content/arab-barometer-ii Engaging in Health Systems in the Middle A r ab B . , 2 013 ( Wave I I I ), ht t p: // w w w East and North Africa. Washington, DC.: .arabbarometer.org/content/arab-barometer Wo r l d B a n k . h t t p s : / /o p e n k n o w l e d g e -iii-0 .worldbank.org/handle/10986/16109. Eurobarometer, http://ec.europa.eu/public_opinion Ya ng , K . , a nd M . H ol z er. 2 0 0 6 . “ T he /index_en.htm Performance–Trust Link: Implications for Gallup W. Poll, http://www.gallup.com/services P e r f o r m a n c e M e a s u r e m e n t .” P u b l i c /170945/world-poll.aspx Administration Review 66 (1): 114–26. Transitional Governance Project, http://transitional Yang, Q. 2013. “A Political Story of Trust: governanceproject.org/ Institutional Settings, Political Performance, Worldwide Governance Indicators (database), and Political Trust in East Asia.” PhD diss., World Bank, http://data.worldbank.org/data University of Pittsburgh. -catalog/worldwide-governance-indicators Effects of Lower Citizen Trust on Citizen Engagement: Circumventing the State, Relying on Wasta, and Pursuing Conflict • Citizens’ trust (and lack of trust) in public institutions shapes their behavior. • Citizens of the Middle East and North Africa rarely engage formally with the state or use formal institutions to demand accountability. • Instead, they circumvent the state when possible, turning to private providers, charitable organizations, and informal pathways (such as wasta or informal payments) to access state services, or they engage in head-on conflict with the state. O n the face of it, the Middle East and North Africa (MENA) region seems Citizens’ engagement: to suffer from a paradox: citizens formal and informal have high expectations of their governments, and yet they do not pursue any formal means of demanding change through action. In this chapter, we provide one potential explana- tion for this paradox. We argue that the key element in understanding citizens’ lack of for- mal engagement is low trust, in many ways driven by lack of an institutional capacity for accountability and voice empowerment even if citizens were to demand accountability through formal channels. The challenge in the MENA region in reviving bottom-up approaches and social accountability tools to put more pressure on state institutions lies in the fact that citizens 233 234 TRUST, VOICE, AND INCENTIVES do not trust that their formal institutions are Citizens also tend not to file complaints or capable of reforms or of making any tangible directly challenge authorities because of their improvements on the ground. They also tend feelings of disempowerment. Analyses at the to generally feel that they have no power over individual level using Arab Barometer data the decision-making process. Therefore, they found that those least satisfied with their gov- disengage because they believe there are no ernment’s efforts to provide services are also of f icia l for m a l pro c e sse s or red re ss the ones least likely to file a complaint mechanisms where their voices can be heard, through formal means when their rights are and indeed rightfully so. The preceding violated.2 chapters illustrated how weak jurisdictions, In the Governance and Anti-Corruption inadequate financial support, and lack of (GAC) survey conducted in 2009 in the independence from political pressure Republic of Yemen, only 10 percent of compromise the effectiveness of many respondents who answered that they had a political and administrative accountability valid reason to make a complaint actually did mechanisms—a situation that is well known so (figure 9.1). This finding suggests that to most citizens, who, as we have noted, see these citizens recognized grievances but felt state institutions as politically captured, unable to address them. A service provider in toothless, and riddled with corruption. Thus Ramallah in the West Bank reflected similar rather than engage directly with the state, concerns in a focus group, saying, “We can citizens often circumvent the state, resort to see injustice and inefficiency, and understand survival mechanisms such as wasta or infor- what causes them; but that in itself does not mal payments, or at times confront the state. lead to action. I actually mean, whatever we Unfortunately, in doing so they exacerbate are discussing right now, it will not solve our the problem and widen the inequities in problems. We are sitting here, and none of us service provision. has the power to make decisions” (World This chapter explores these dynamics. It Bank 2007). As Ringold, Holla, and Koziol begins with a discussion of the failure of citi- (2012, 12) note, such problems are “particu- zens to engage directly with the state. It then larly salient in low-income countries [and turns to a discussion of how citizens circum- subregions], where providers may come from vent the state (primarily seeking individual more affluent backgrounds and citizens may solutions) or engage in confl ict. It concludes not feel in a position to question them . . . by considering the impact that the expansion because of their status, credentials, or knowl- of social media can have on engagement. edge, or they may be concerned about the repercussions of giving negative feedback”(also see World Bank 2010).3 Directly engaging the state and Engagement with the state in the form of public services delivery political action does not happen frequently, institutions and citizens seldom see voting, signing peti- In much of the MENA region, people are tions, and other actions as effective means of unlikely to use official channels to obtain ser- changing policies in ways that improve the vices because they believe they have little lives of ordinary citizens. Thirty-four percent chance of succeeding by simply following the of all respondents in sampled MENA coun- rules. For example, in surveys conducted in tries of the sixth wave of the World Values Morocco and Algeria in 2007, only about Survey answered that they had never partici- half of citizens said they would go directly to pated in a local election and 32 percent a government agency to obtain services, and responded the same about a national one. only about a quarter thought it would be The percentage of respondents never voting effective to do so. More citizens believed they in a local election was as high as 72 percent stood a better chance if they went through in Tunisia and as low as 20 percent in friends or family.1 the Arab Republic of Egypt. However, EFFECTS OF LOWER CITIZEN TRUST ON CITIZEN ENGAGEMENT 235 FIGURE 9.1 Rate of respondents who had and acted on a complaint about health services: Republic of Yemen, 2009 69 Percent Among these Among these 18 10 Have had contact with health Had a valid reason to make a Actually made a complaint service provider in past 6 months complaint Source: GAC, Republic of Yemen, 2009. as indicated earlier, the picture also varies their analysis of the 2010 Qatar World within countries. For example, in Jordan, Values Survey that the 20 percent of the resi- participation levels in parliamentary elections dent population who do engage in civic asso- can range from 20 percent to more than ciations “seem to engage in association life 90 percent at the local level (figure 9.2a). In primarily in order to seek their private Tunisia, some degree of variation in voting is advantage and interact with like-minded found as well at the level of administrative individuals, ends that serve exactly to rein- districts, although it is not as high as that force rather than challenge the established found in Jordan (figure 9.2b). social and political system.” Citizens rarely join political parties or Evidence from the Arab Barometer indi- civil society organizations (CSOs) in order to cates that in some countries, citizens’ limited influence policy because they do not trust the engagement stems from the little trust they efficacy of these groups. 4 For example, have in the efficacy of organizations, includ- the World Values Survey found that in the ing political parties (see figure 9.3 and 10 economies surveyed, only in Lebanon, the table 9A.1 in the annex). Estimation of a Republic of Yemen, and West Bank and binary logit model7 reveals that the predicted Gaza were more than one in five citizens probability of becoming a member of a politi- likely to be a party member, and only in the cal party is higher among respondents who Republic of Yemen did party membership trust political parties and their efficacy. surpass 20 percent. 5 Citizen engagement in The effects are noticeably strong in CSOs in MENA countries is also low. The Lebanon and the Republic of Yemen. In the Republic of Yemen again leads, with latter, the probability of becoming a member 28 percent of respondents claiming to be of a political party is 60 percent when members of CSOs, but most of these associa- respondents trust parties “to a great extent,” tions are not aimed at pressuring policy whereas it is only 18 percent when they makers.6 Gengler et al. (2013) concluded in “absolutely do not trust” parties. Similarly, in FIGURE 9.2 Percentage of respondents participating and voting in elections at the subnational level: Jordan and Tunisia, 2014 a. Jordan: Percentage of respondents participating in elections at the local level 100 90 80 70 60 50 Percent 40 30 20 10 0 A W A A U A B R A S A J A A A H A A K A A A A A D A A A A A W W A K B A A K A K W A G A A A B S K S W A A A A A S M S A R l T a l J nn m l A ay ag l T am l M ab l M l Q l 'A am l A l T af l T ssh l S l M l M ei l Z l K l R l 'O sk a a l T afr as l H l B afr l Q afr a l M hr l 'E l A ss as hm ho ak a l T rro ss ss l M am u uf l Ta aj ha di or ah N fra t R hd ha m a al a ad y z m ha ra ay u hu a a r A ab ar ab u an di di ay an ira as ala an ad an di u an es br ak ira a w ib di ay w ak hu a m kh yb ib lit M n ar iwa h as an ni a 2 shm Ann qab isi n A a ir H lith nje ba na bik nar shm ss'a iria am 'i da Al Mo Mo ba je 1 him d je isia je Mo tha da ar hn 2 kh fa M ba da hn na ns a ay a2 h 1 ou ia ra am 2 4 wa Al a ia n a a M us us 1 3 iA 1 2 2 us lat l a2 Al K ou 3 s a 1 A ou 1 im j r sa ia az ilin a 1 l Ba 1 h J 2 a a a a h a u a S 4 a yd a za Kh ad ar ida sh 2 5 m lS ha 3 af rin 1 nu uf m bi ib A a A ss ia a ali ss uk uq un b. Tunisia: Percentage of respondents planning on voting in elections at the administrative district level 100 90 80 70 60 50 Percent 40 30 20 10 0 C M S M B E M S O B E N D F S S B B M J E G S B S A F Z M S B H S Sa J 'Sa heb idi ok ab l M eg idi Haff mr Le K ou l M asr Bard Hen jer El A ahs aki idi ir M Bize Sao en Kala az Kair Kasr am Mat zzo har idi Mon ou ous Tha l M eria arz ed ous ize ass bib As bal elm ke ika Bou nin So our rin El B ou ane ram haj enz alla o cha ba mra et Ha c rt ua A B m l r n ec z la ah h M Ez ss he e N f r ro a Ke zoun oua He mel eur uho El M ou asti he se S a ghir na is enin se R te S i El a ker et O a u n l l u z r l i a o F Al e uika ouj e i hi id zit ine g s bi ra a So al r elh id di e S iad u er ul ed r ou n ua orth ut Ea Ab ou h th id h st elh th am id Mean Standard deviation Source: Governance and Local Development survey, 2014. 236 Note: In panel a, the numbers that appear next to some localities are sublocalities. Numbers were added for survey sampling purposes and do not represent official administrative boundaries. EFFECTS OF LOWER CITIZEN TRUST ON CITIZEN ENGAGEMENT 237 FIGURE 9.3 Predicted probability of becoming a member of a political party as a function of trust in political parties: Selected MENA countries, 2010–11 Adjusted predictions with 95% Cls 1.0 0.8 Pr (membership in a political party) 0.6 0.4 0.2 0 To a great To a medium To a limited Absolutely extent extent extent do not trust Trust political parties Yemen, Rep. Lebanon Iraq Algeria Jordan Source: Arab Barometer, 2010–11 (Wave II). Note: MENA = Middle East and North Africa; CI = confidence interval; Pr = probability. Lebanon the probability of becoming a mem- Republic of Yemen, the probability of becom- ber increases from 10 percent to 42 percent ing a member is 14 percent among those who when respondents trust political parties “to a trust such organizations, where it is only great extent.” The results are also significant 10 percent among those who do not. for Algeria, Iraq, and Jordan, although of a Citizens’ quests for accountability are smaller magnitude. further hampered by the institutional Similar to the lack of engagement in politi- constraints (so-called supply-side account- cal parties, the limited trust in CSOs or their ability) that do not allow CSOs to function capacity to implement changes is significantly freely without state intervention and con- related to lack of membership among citizens tinual crackdowns. These in turn limit citi- in such movements, although the effect is zens’ effectiveness in influencing policy and smaller in magnitude (see figure 9.4 and can explain why they do not trust and do table 9A.2 in the annex). Estimation of a not join such organizations. Indeed, the binary logit model8 reveals that the probabil- social accountability literature argues that ity of becoming a member of a CSO such as a citizens need to supplement their demands youth or cultural group is lower among those for accountability and their capacity for who do not trust such organizations. In the collective ac tion with accessible and 238 TRUST, VOICE, AND INCENTIVES FIGURE 9.4 Predicted probability of becoming a member of a youth/civil society organization (CSO) as a function of trusting CSOs: Selected MENA countries, 2010–11 Adjusted predictions with 95% CIs 0.20 Pr (membership in a youth/civil society 0.15 organization) 0.10 0.05 0.00 To a great To a medium To a limited Absolutely extent extent extent do not trust Trust youth/civil society organizations Yemen, Rep. Saudi Arabia Iraq Algeria Lebanon Egypt, Arab Rep. Jordan Source: Arab Barometer, 2010–11 (Wave II). Note: MENA = Middle East and North Africa; CI = confidence interval; Pr = probability. responsive accountability institutions to A report by the Carnegie Middle East Center, move from the “‘accountability trap” to The Arab World’s Education Report Card: improved public sector performance, or School Climate and Citizenship Skills, con- else, as Fox (2014) has written, “voice cluded, “The effective school-family partner- becomes toothless.” ship that is present in many non-Arab Even parent-teacher organizations and countries is virtually absent in the Arab boards of trustees are often weak. Education region. This has negative implications for the is well suited to community action aimed at school climate as well as student achieve- improving service quality because many ment” (Faour 2012, 24). parents view the education of children as Often, citizens contact local governments extremely important. Schooling (unlike (and particularly mayors) for help in obtain- health care, for example) accommodates ing jobs and accessing education, health care, continual, frequent engagement among and other services, because they view such parents, teachers, and school administrators. officials as important interlocutors with the Moreover, a 2007 World Bank report state. However, as elaborated earlier, local concluded that these parties do want more governments are often weak in the MENA communication between parents and teach- region and do not have significant autonomy ers. And yet in public schools such councils over financial resources and decision-making remain weak or absent (World Bank 2007). power. And yet high volumes of requests are EFFECTS OF LOWER CITIZEN TRUST ON CITIZEN ENGAGEMENT 239 still received even when the services sought residents personally—helped to build the are not under local officials’ purview or the legitimacy and trust needed to mobilize local government’s resources are inadequate. community support. As a result, Qena saw For example, in the Republic of Yemen citi- considerable gains in the six years of Labib’s zens petitioned local council members for tenure: his efforts generated local funds to services even though both council members improve hospitals, schools, and roads and and their constituents acknowledged the created programs to lower illiteracy and council members were “helpless” and “not unemployment rates (Bennet 2011). the decision makers” (Aslam 2014). In Jordan, mayors and local council members bemoaned that even when they are over- Circumventing the state whelmed with enormous challenges, limited In authoritarian states plagued with politi- time, and few material resources, they are cally captured institutions that cater only to inundated with requests for all types of the interests of certain groups and constitu- services.9 ents without any guarantees of equal protec- At times, however, engagement in local tion or fair judicial proceedings, citizens are CSOs and collective action aimed at obtain- more likely to rely less on formal processes to ing better services are occurring. The case obtain services and secure their rights studies in chapter 3 illustrate how in some (Hardin 1996). In the MENA region, as else- instances citizens have been able to mobilize where, citizens turn to nonstate actors for themselves and fi nd ways to hold providers services. Those who can afford to do so buy and their local representatives accountable, water from private sources, put their children albeit their practices remain context-specific in private schools, and see doctors in private and have yet to be institutionalized. The clinics. The wealthy and poor alike also turn 2000s saw an increase in strikes, primarily to each other (World Bank 2007). The 2013 for better wages and working conditions Transitional Governance Project survey in (see Ortiz et al. 2013; Khatib and Lust 2014), Libya found that 62 percent of respondents and since 2011, from Bahrain to Oman, helped out neighbors at least once a month, there have been protests over school and in Egypt helping those in need is consid- fees, wages, better living standards, and gov- ered the most important citizenship norm. ernment corruption.10 There are also impor- Local social elites, religious leaders, and tant national, regional, and local variations. charities also help citizens meet their needs. For example, in the World Bank’s 2008 For example, in Egypt wealthy religious peo- Public Expenditure Tracking Survey (PETS) ple sponsor a chain of religious Al-Azhar in Egypt, nearly two-thirds of eighth-grade schools in Zagazig. These schools—which students reported that their parents had been are free, provide a daily meal, and emphasize invited to participate in committees (far religious education—provide an alternative above the regional average), and one-third of education for the rural poor. But such alter- the schools in the six governorates included natives are not available everywhere (World in PETS received extra resources from com- Bank 2007). munity sources.11 A governor in Qena, As described in chapter 5, citizens of the Egypt, Adel Labib, demonstrated how par- MENA region tend to use personal connec- ticipation can be achieved. His firm belief tions, wasta, to get things done, even when that community participation is critical to turning to officials. Those seeking to change development led him to form local expert their children’s school or find a hospital bed and community councils to support school for a sick parent, for example, often need development, and his willingness to engage someone in the right place willing to pave the in the communities—walking through towns way.12 Many turn to ministers of parliament and holding meetings to get to know (MPs, called “service parliamentarians,” 240 TRUST, VOICE, AND INCENTIVES FIGURE 9.5 Responsiveness of members of parliament (MPs): services within the bureaucracy (Morris and Arab Republic of Egypt (2012), Libya (2013), and Tunisia (2012) Klesner 2010). According to the evidence in the MENA 90 region, citizens use informal side payments 80 (or bribes) to obtain services, although the 70 prevalence of bribes varies across services as 60 well as countries and subnational regions. For example, in the Republic of Yemen nearly Percent 50 two-thirds of respondents said that when the 40 traffic police stopped them, they were asked 30 to make a side payment to avoid fines (World 20 Bank 2010). About one-third of respondents 10 were asked for additional payments for their eldest child’s schooling in the past year 0 Egypt, Arab Rep. Libya (2013) Tunisia (2012) (33.5 percent in rural areas and 22.8 percent (2012) in urban areas), and the average fees were Agree that MPs respond more Agree that MPs respond more nearly one-third of the respondent’s average quickly to voters quickly to friends or family monthly salary (World Bank 2010). 14 members Similarly, in 2012, 48 percent of Egyptians Source: Transitional Governance Project. completely agreed, and another 15 percent somewhat agreed, with the statement “A bribe should be paid to get access to health care” (Transitional Governance Project, na’ib khadmat), school principals, clinic 2012). The World Bank’s 2009 Governance directors, or even ministry officials to appeal and Anti-Corruption Country Diagnostic for a school transfer, a faster referral for Survey in the Republic of Yemen found that clinic services, or other solutions to individ- 39 percent of respondents believed that “fam- ual problems (figure 9.5). But they make such ilies like theirs” needed to pay bribes in order appeals on a personal basis. For example, to obtain education services (World Bank MPs are likely to respond more readily to 2010). In Tunisia, informal payments are also people they know and who voted for them deemed useful to obtain a government job or than to the average citizen.13 More generally, navigate the bureaucracy. By contrast, in the Arab Barometer surveys found that, Libya (and possibly in Gulf countries where across countries, 16–40 percent of citizens, rents help support services that are provided depending on the country, used wasta within by expatriate communities), citizens believe during the last five years in filling a need. informal payments are less necessary. A 2013 Exploiting personal connections to obtain survey found that 87 percent of Libyans com- services and resources is for many not a form pletely disagreed with the statement “People of corruption but simply a way to do business like me have to pay bribes for medical treat- (Khatib 2013). ment in local hospitals” (Transitional At times, the practical path to obtaining Governance Project, 2013). Nevertheless, services more quickly means engaging in cor- bribery is often so common that it is an rupt activities such as giving out bribes or expected part of the transaction.15 It may fos- making payments under the table. In this ter additional disincentives to follow the writ- case, the effect of corruption becomes self- ten rule since the general perception is that reinforcing. It reduces trust in the regime’s “everyone is doing it,” thereby lowering the ability to respond to citizens’ concerns (as risks associated with offering bribes. demonstrated earlier), which in turn breeds Although side payments could be viewed corruption because it could drive citizens to as positive contributions from the local com- offer bribes to secure their rights and public munity to help financially struggling schools EFFECTS OF LOWER CITIZEN TRUST ON CITIZEN ENGAGEMENT 241 and local clinics, the process through which Such actions address basic needs, and they contributions are made is suboptimal. Such are certainly reasonable responses to the payments are an inefficient way to obtain suf- existing circumstances. People have little ficient resources. They are opaque, they reason to believe that if they try to obtain increase transaction costs, and they foster services through direct means they will suc- dissatisfaction. As one parent explained, ceed. Moreover, in the authoritarian regimes that prevail across the MENA region, If you want to transfer your child from demanding accountability and better services a school in one place to another place, through more direct, confrontational appeals you have to pay a lot of expenses, could have serious negative consequences. around 500–600 pounds without any Finally, as Olson (1965) noted long ago, col- reason, even if it is your right to lective action is difficult to organize, particu- transfer and it is in your residence larly among large groups. district. If you refuse to pay, they Individual solutions are reasonable, but would say the location is not suitable, they also can be problematic. First, they and you have to pay donations to the exacerbate social inequalities because some school and gifts to the managers, and citizens (such as the less educated or urban to pay for construction. My daughter migrants) have fewer economic resources and was not transferred until I bought less access to the social and political net- photocopying papers for 100 pounds, works that can provide access to services. a carpet, and an electric fan, although Such actions also undermine citizen agency. it was her right to be transferred. I do People are able to get what they need not not mind paying but with my own free because they communicate with the state to will. (World Bank 2007) demand better services, but because they As this respondent suggests, even when accept the status quo and find ways to work side pay ment s a re reque sted for t he around it. Finally, such actions quash collec- “greater good,” such payments appear to tive action. It is not surprising, then, that increase dissatisfaction. The Yemeni GAC there is a significant negative correlation survey found that those who were asked between clientelistic practices (and particu- for a bribe were more likely to complain larly vote buying) and the provision of public about the service (World Bank 2010).16 services (Khemani 2013). Bribery deflates a citizen’s satisfaction, which likely also further degrades his or her engagement in strengthening service Clashing with the state delivery systems. Finally, at times pent-up grievances provoke Citizens’ coping strategies also include citizens to engage in head-on collisions with actions that undermine public welfare, the state. In Tunisia and Egypt, for exam- circumvent the law, and quietly challenge the ple, poor service provision provided an state (World Bank 2007; Bayat 2010). People impetus for citizens to take to the streets in fi nd ways to divert electric lines, pull their 2010 –11, demanding change. The Arab children out of school, or exploit subsidized Uprisings were remarkable moments of medications or foodstuffs. Often they justify mass mobilization in opposition to the state such actions by pointing to the injustices of that opened up a transition period offering the system and the ruling elites. For example, citizens new reasons for hope. As we dis- one provider in the West Bank noted, “One of cuss in chapter 10, the uprisings led to (at the main factors which contribute to the least temporary) greater trust and engage- increasing rate of power line theft is that ment and opened up possibilities for posi- people consider us thieves, blood suckers tive institutional reform. and that it is OK to steal from us” (World At other times, the mounting dissatisfac- Bank 2007). tion leads to longer and more violent clashes 242 TRUST, VOICE, AND INCENTIVES between citizens and the state. In the of Tunisians and 30 percent of Egyptians said Republic of Yemen, for example, violence it was to organize actions and manage between the Houthis and the state has been activists. growing. According to a 2013 survey in the Users across the MENA region are opti- Republic of Yemen, poor government mistic about the use of new technology and services are viewed as a major reason for the social media to improve service delivery. Houthis’ grievances.17 Similar grievances, Seventy-five percent of the Internet users sur- driven by poor governance and inequality, veyed for the Arab Social Media Report appear to underlie support for the Islamic (2014) believed that one of the benefits of State of Iraq and Syria (ISIS) and other social media was better quality of service movements that are challenging delivery because the flow of information states across the region (Gerges 2014; between customer and government was Al Makhtoum 2014). improved; 78 percent believed it would lead to services that better meet citizens’ needs; 79 percent agreed that social media would The use of social media reduce service delivery costs; and 76 percent Social media are widely understood to be an mentioned the increased inclusiveness of effective means of catalyzing collective service delivery. In addition, 60 percent of action, but they can also be a powerful tool respondents from high-income MENA coun- for generating discussions of government tries believed social media made their govern- services, local issues, and politics. A survey ment more accessible, compared with an conducted by the Arab Social Media Report average of 40 percent in middle-income (2014) found that about 50 percent of respon- countries. dents were using personal social media And yet it appears that most of the con- accounts such as Facebook, Twitter, and versations are limited to social groups and LinkedIn to talk about government ser- not between citizens and their government vices.18 In 2012 the Pew Research Center (Shediac et al. 2013). Thus citizens are skep- compared social media usage internationally tical about the impact social media will have and found that those living in the MENA on their government’s responsiveness to their region were among the top users of social needs. If people do engage with their govern- media to debate politics, religion, and local ment, the majority of youth prefer to do so issues.19 When asked, 64.5 percent of MENA via blogs and social media, whereas those in social media users reported discussing poli- the older generation (49 and above) prefer to tics, compared with 34 percent internation- communicate face to face and through ally, and 79.2 percent reported talking about gatherings such as majalis. Few citizens use local issues, compared with 46 percent inter- their government’s official social media page nationally (Pew Research Center, 2012). This as a means of communicating feedback or was most evident during the Arab Uprisings, suggestions. Instead, 73.8 percent of those when activists used social media for dissemi- who do so use such pages to gather informa- nating or sharing information and mobilizing tion on services or entities (figure 9.6). And groups. For example, when Tunisian and when given other options for seeking infor- Egyptian Facebook users were asked about mation about government services, citizens their main usage of Facebook during early are less likely to fi rst choose the government 2011, 33 percent of Tunisian users and service website or social media platform, 24 percent of Egyptian users said it was to preferring instead to conduct a general spread information to the world about the online search—indeed, in one survey movement and related events; 31 percent of 30 percent of respondents preferred this Tunisians and 31 percent of Egyptians said it option (figure 9.7). This is likely an indica- was to raise awareness inside the country on tion of the low levels of trust citizens have in the causes of the movement; and 22 percent government responsiveness through social EFFECTS OF LOWER CITIZEN TRUST ON CITIZEN ENGAGEMENT 243 media channels or a reflection of their lack responsive and accessible mechanisms even if of interest in communicating with govern- they were to voice their concerns. After the ment (Shediac et al. 2013). recent upheavals in the region, there were There is no denying that connectivity is attempts to expand political mobilization having a bigger influence on citizen and gov- through CSOs and more widespread use of ernment engagement. Promising efforts have social media, but such attempts mainly been made in areas such as e-governance in some MENA countries, but it is still not clear FIGURE 9.6 Primary uses of government social media: MENA region, 2014 whether other countries in the MENA region will follow this path and how governments My primary use of government social media pages and citizens in the region can harness these is for the following purposes technologies to improve accountability and 80 the quality of service delivery. In the absence 70 of strong institutions and trust in govern- ment, even the impact of social media may be 60 limited. Percent 50 40 Conclusion 30 20 In the MENA region, poor-quality service 10 provision is affecting citizens’ distrust of the state, which in turn has shaped their engage- 0 ment with the state. In most MENA coun- Complain about government services tries, citizens do not formally engage with the Communicate with senior government officials directly state to demand accountability. This is in Suggest new ideas or improvements for government services Give feedback on or evaluation of government services I have used part because they do not believe their institu- Access information on government services and entities tions are capable of implementing any credi- ble improvements on the ground and in part Source: Arab Social Media Report 2014. because of their awareness of the lack of Note: MENA = Middle East and North Africa. FIGURE 9.7 Sources of information about government services: MENA region, 2014 40 30 Percent 20 10 0 Official government social media pages Government mobile apps on my phone My own personal social media accounts to A phone call to the relevant government agency ask friends and followers for advice An official electronic government portal Ask a colleague, friend, or family member The website of the relevant government A general online search engine department Source: Arab Social Media Report 2014. Note: MENA = Middle East and North Africa. 244 TRUST, VOICE, AND INCENTIVES occurred on ad hoc basis, and it is still not In more extreme cases, confrontations, clear how far activists can push for account- which spiked in 2011, can take various ability and improvements in the quality of forms—from toppling autocrats in Egypt and service delivery. Tunisia, to civil wars in Libya, Syria, and the Overall, most citizens have turned away Republic of Yemen, to state-led reforms in from any formal engagement with the state. Morocco and much of the Gulf. These con- Instead, they have sought services from pri- frontations have an impact across the cycle of vate providers, used mechanisms that bypass performance. They dislodge institutions, formal state pathways in an attempt to alter the demand for and supply of services, benefit from public services, and, at times, alter citizens’ trust in institutions, and spur confronted the state. In normal times, such new engagement—that is, they can reshape survival mechanisms further reduce the effec- the cycle of performance. How such major tiveness and credibility of public institutions. shocks, as well as more gradual reforms and Meanwhile, the lack of collective action donor incentives, can shift the cycle of per- contributes to the problems of nepotism, formance out of a low-equilibrium state is the cronyism, and capture. subject of chapters 10–12. Annex Predicted probabilities tables Table 9A.1 Predicted probability of trusting a political party on becoming a member of one: Selected MENA countries, 2010–11 Independent variable Predicted probability 1 = trust a political party to a great extent, Algeria 0.0578*** (0.0140) 1 = trust a political party to a great extent, Iraq 0.142*** (0.0268) 1 = trust a political party to a great extent, Jordan 0.0300*** (0.00997) 1 = trust a political party to a great extent, Lebanon 0.420*** (0.0411) 1 = trust a political party to a great extent, Republic of Yemen 0.600*** (0.0421) 2 = trust a political party to a medium extent, Algeria 0.0269*** (0.00609) 2 = trust a political party to a medium extent, Iraq 0.0693*** (0.0117) 2 = trust a political party to a medium extent, Jordan 0.0137*** (0.00453) 2 = trust a political party to a medium extent, Lebanon 0.246*** (0.0225) 2 = trust a political party to a medium extent, Republic of Yemen 0.403*** (0.0301) 3 = trust a political party to a limited extent, Algeria 0.0129*** (0.00298) 3 = trust a political party to a limited extent, Iraq 0.0340*** (0.00595) (continued next page) EFFECTS OF LOWER CITIZEN TRUST ON CITIZEN ENGAGEMENT 245 Table 9A.1 Predicted probability of trusting a political party on becoming a member of one: Selected MENA countries, 2010–11 (continued) Independent variable Predicted probability 3 = trust a political party to a limited extent, Jordan 0.00654*** (0.00221) 3 = trust a political party to a limited extent, Lebanon 0.134*** (0.0151) 3 = trust a political party to a limited extent, Republic of Yemen 0.242*** (0.0225) 4 = absolutely do not trust political parties, Algeria 0.00902*** (0.00203) 4 = absolutely do not trust political parties, Iraq 0.0239*** (0.00406) 4 = absolutely do not trust political parties, Jordan 0.00457*** (0.00152) 4 = absolutely do not trust political parties, Lebanon 0.0971*** (0.00904) 4 = absolutely do not trust political parties, Republic of Yemen 0.182*** (0.0161) No. of observations 5,793 Source: Arab Barometer, 2010–11 (Wave II). Note: All controls are at their mean value. Standard errors are in parentheses. MENA = Middle East and North Africa. *p < .10 **p < .05 ***p < .01 Table 9A.2 Predicted probability of trusting a youth/civil society organization on becoming a member of one: Selected MENA countries, 2010–11 Independent variable Predicted probability 1 = trust a youth/civil society organization to a great extent, Algeria 0.0878*** (0.0128) 1 = trust a youth/civil society organization to a great extent, Arab Republic of Egypt 0.0365*** (0.00710) 1 = trust a youth/civil society organization to a great extent, Iraq 0.0980*** (0.0134) 1 = trust a youth/civil society organization to a great extent, Jordan 0.0346*** (0.00632) 1 = trust a youth/civil society organization to a great extent, Lebanon 0.0807*** (0.00954) 1 = trust a youth/civil society organization to a great extent, Saudi Arabia 0.113*** (0.0127) 1 = trust a youth/civil society organization to a great extent, Republic of Yemen 0.135*** (0.0159) 2 = trust a youth/civil society organization to a medium extent, Algeria 0.0724*** (0.00941) 2 = trust a youth/civil society organization to a medium extent, Arab Republic of Egypt 0.0298*** (0.00553) 2 = trust a youth/civil society organization to a medium extent, Iraq 0.0810*** (0.00948) 2 = trust a youth/civil society organization to a medium extent, Jordan 0.0283*** (0.00491) (continued next page) 246 TRUST, VOICE, AND INCENTIVES Table 9A.2 Predicted probability of trusting a youth/civil society organization on becoming a member of one: Selected MENA countries, 2010–11 (continued) Independent variable Predicted probability 2 = trust a youth/civil society organization to a medium extent, Lebanon 0.0664*** (0.00805) 2 = trust a youth/civil society organization to a medium extent, Saudi Arabia 0.0937*** (0.0100) 2 = trust a youth/civil society organization to a medium extent, Republic of Yemen 0.112*** (0.0117) 3 = trust a youth/civil society organization to a limited extent, Algeria 0.0611*** (0.00838) 3 = trust a youth/civil society organization to a limited extent, Arab Republic of Egypt 0.0250*** (0.00493) 3 = trust a youth/civil society organization to a limited extent, Iraq 0.0685*** (0.00899) 3 = trust a youth/civil society organization to a limited extent, Jordan 0.0237*** (0.00439) 3 = trust a youth/civil society organization to a limited extent, Lebanon 0.0560*** (0.00750) 3 = trust a youth/civil society organization to a limited extent, Saudi Arabia 0.0793*** (0.00875) 3 = trust a youth/civil society organization to a limited extent, Republic of Yemen 0.0953*** (0.0112) 4 = absolutely do not trust youth/civil society organizations, Algeria 0.0610*** (0.00856) 4 = absolutely do not trust youth/civil society organizations, Arab Republic of Egypt 0.0250*** (0.00491) 4 = absolutely do not trust youth/civil society organizations, Iraq 0.0684*** (0.00979) 4 = absolutely do not trust youth/civil society organizations, Jordan 0.0236*** (0.00453) 4 = absolutely do not trust youth/civil society organizations, Lebanon 0.0559*** (0.00747) 4 = absolutely do not trust youth/civil society organizations, Saudi Arabia 0.0792*** (0.0101) 4 = absolutely do not trust youth/civil society organizations, Republic of Yemen 0.0951*** (0.0122) No. of observations 7,885 Source: Arab Barometer, 2010–11 (Wave II). Note: All controls are at their mean value. Standard errors are in parentheses. MENA = Middle East and North Africa. *p < .10 **p < .05 ***p < .01 Notes 9 percent; “go to courts,” 2 percent; “take issue to a minister,” 3 percent; “take issue to 1. These data are taken from a Transitional local government,” 3 percent; “ask the local Governance Project survey conducted in MP [member of parliament],” 1 percent. 2007 on constituent approaches when deal- No constituents responded “take issue to a ing with government. In Algeria, the responses political party” or “other.” In Morocco, of constituents (n = 232) were as follows: responses from constituents to the same sur- “take issue directly to agency involved,” vey (n = 768) were as follows: “take issue 56 percent; “take issue to family and friends,” directly to agency involved,” 49 percent; 26 percent; “take issue to a religious person,” “take issue to family and friends,” 31 percent; EFFECTS OF LOWER CITIZEN TRUST ON CITIZEN ENGAGEMENT 247 “take issue to local government,” 7 percent; 21 percent (n = 1,283, 2006); Jordan, 6 percent “take issue to a religious person,” 2 percent; (n = 1,130, 2006); Lebanon, 18 percent “go to courts,” 3 percent; “ask the local MP,” (n = 1,179, 2007); Morocco, 13 percent 2 percent; “take issue to a minister,” (n = 1,272, 2006); Republic of Yemen, 2 percent; “take issue to a political party,” 28 percent (n = 663, 2007); and West Bank and 1 percent; “other,” 4 percent. Gaza, 22 percent (n = 1,265, 2006). 2. Using the Arab Barometer data (2010–11), 7. The binary logit model included membership we ran a multinomial regression model that in a political party (a binary variable coded 1 if consisted of a categorical dependent variable yes and 0 if no) as a dependent variable and labeled as “how easy is it to access services in the following explanatory variables: (1) trust case your rights have been violated” and in political parties (a categorical variable coded coded as 1 = very easy, 2 = easy, 3 = difficult, as follows: 1 = trust to a great extent; 2 = trust 4 = very difficult, and 5 = never tried. The to a medium extent; 3 = trust to a limited independent variables were the following: extent; 4 = absolutely do not trust parties); satisfaction with government’s efforts to (2) age; (3) gender; (4) income; (5) educational improve service delivery, income, age, gender, attainment; and (6) urban/rural status. urban/rural, and income quintile. Holding 8. The binary logit model included membership everything constant, the relative log odds of in a youth organization (a binary variable not trying to file a complaint versus believing coded as 1 if yes and 0 if no) as a dependent it is very easy decreases by 3.145 when people variable and the following explanatory believe that a government’s efforts are very variables: (1) trust in CSOs—associations, good compared with believing it is very bad clubs, volunteer youth groups, etc. (a cate- (the result is statistically significant at the .05 gorical variable coded as follows: 1 = trust to level). In other words, they are more likely to a great extent; 2 = trust to medium extent; file a complaint when they are more satisfied 3 = trust to a limited extent; 4 = absolutely do with the government’s efforts. not trust parties); (2) age; (3) gender; 3. In the GAC report on the Republic of Yemen (4) income; (5) educational attainment; and (World Bank 2010), regression results suggest (6) urban/rural status. that, in general, males, the wealthy, and the 9. Interviews with local councils and mayors in more educated are more likely to have had northern areas of Jordan, October 22–24, contact with public services providers than 2014. females, poor respondents, and less educated 10. For example, see International Crisis Group respondents. Similarly, urban respondents are (2012) and Middle East Monitor (2013). more likely to have had contact with public 11. Bold and Svensson (2010) noted that there services providers than rural respondents. was considerable variation, “ranging from all 4. As demonstrated by the very low levels of schools in Ismalia to none in Fayoum.” There citizen awareness of the existence of such was also wide variation across schools. institutions in areas close to their residence According to the PETS survey results, the (Gallup World Poll, 2013). schools that reported receiving funds from 5. According to data from the 2010–11 Arab boards of trustees and communities recorded Barometer and Transitional Governance a total of LE 148,124 during the fiscal Project, the following percentages of respon- 2007–08. The mean amount received was dents declared themselves members of a polit- LE 2,743 per school, but the median was ical party: Algeria, 2 percent (n = 1,207, much lower, LE 714 per school, reflecting the 2011); Egypt, 3 percent (n = 2,510, 2012); uneven distribution across schools. Iraq, 3 percent (n = 1,196, 2011); Jordan, 12. For discussions of wasta, see Kilani and 1 percent (n = 1,178, 2010); Lebanon, Sakija (2002). 12 percent (n = 1,383, 2011); Libya, 5 percent 13. MPs in Morocco and Algeria also appear (n = 1,001, 2013); Republic of Yemen, more likely to be more responsive to men 31 percent (n = 1,147, 2011); Tunisia, than to women. On average, 20–29 percent 2 percent (n = 1,185, 2011); and West Bank of requests for constituent services made to and Gaza, 14 percent (n = 1,190, 2010). deputies in Morocco and Algeria are from 6. According to the 2006–08 Arab Barometer females (see Benstead 2014). data, the following percentages of respondents 14. There is also interesting regional variation in declared themselves members of a CSO: Algeria, the Republic of Yemen. For example, over 248 TRUST, VOICE, AND INCENTIVES 50 percent of respondents from Taiz, Hajjah, Corruption, the Judicial System and the and Sana’a reported making additional pay- Quality of Governmental Services in Egypt.” ments for their eldest child’s education in the Research, Monitoring and Governance Unit, last school year, whereas 10 percent of Social Contract Center, Washington, DC. respondents from Al Mahwit, Abyan, and h t t p : / / s i t e r e s o u r c e s .w o r l d b a n k . o r g Al-Baidha reported doing so. / I N T L AW J U S T I N S T/ R e s ou rc e s / E g y pt 15. For example, a survey conducted in Egypt in _Corruption _ JudicialSystem_Government 2009 found that 41 percent of respondents _Services_Report.pdf. believed a bribe is generally a “previously Al M., S. M. bin Rashid. 2014.“How to Fight known thing that happens spontaneously ISIS: A ‘Pre-Packaged Franchise of Hate.” between the civil servant and the citizen,” Huffington Post, October 5. http://www 16 percent said that people offer the bribe .huffingtonpost.com /sheikh-mohammed-bin unprompted, and 30 percent said a civil ser- -rashid-al-maktoum / how-to-fight vant openly requests the payment (Al-Gharini, -isis_b_5936620.html. Al-Rashidi, and Al-Gamal 2009). Arab Social Media Report. 2014. “Citizen 16. “There is also a strong correlation, 0.66, Engagement and Public Services in the Arab between the percentage of respondents who World: The Potential of Social Media.” have been asked to pay a bribe for a service G over na nce a nd I n novat ion Prog ra m , and the percentage of respondents who Mohammed Bin Rashid School of believe they have a valid reason to complain Government, Dubai. about the service” (World Bank 2010, 19). Aslam, G. 2014. “Notes from Fieldwork in Yemen 17. In a survey conducted by the Yemen Polling Conducted for SA Flagship.” Unpublished raw Center (2013), almost two-thirds of respon- data, World Bank, Washington, DC. dents identified poor government services as a Bayat, A. 2010. Life as Politics: How Ordinary major reason for their grievances (on a four- People Change the Middle East. Amsterdam: scale answer, 42 percent identified poor gov- University of Amsterdam. ernment services as a very large reason, and Bennet, R. 2011. “Building Civic Participation 21 percent identified it as a somewhat large and Enforcing the Law: Qena, Egypt, 1999– reason). Other factors were the insensitivity 20 0 6.” I n In n o vatio n s fo r S u c c e ssf ul of the central government (47 percent, very Societies. Princeton, NJ: Princeton University large; 18 percent, somewhat large) and lack of Press. local authority and autonomy (60 percent, Benstead, L. J. 2014. “Do Quotas Improve very large; 15 percent, somewhat large). Constituent Service Provision to Women? 18. Based on a regional online survey adminis- Evidence from North Africa.” Unpublished tered in 22 Arab economies (Algeria, Bahrain, manuscript, Por tland State Universit y, Comoros, Djibouti, Egypt, Iraq, Jordan, Portland, OR. Kuwait, Lebanon, Libya, Mauritania, Bold, T., and J. Svensson. 2010. “Baseline Data Morocco, Oman, Qatar, the Republic of Report from the Community Score Card Pilot Yemen, Saudi Arabia, Sudan, the Syrian Arab in Ismailia Governorate.” World Bank, Republic, Tunisia, the United Arab Emirates, Washington, DC. and West Bank and Gaza). The survey sam- Faour, M. 2012 The Arab World’s Education pled 3,654 respondents from February to Report Card: School Climate and Citizenship May 2014. Skills. Washington, DC: Carnegie Middle East 19. Based on a Pew Research Center survey con- C enter. ht t p: //c a r ne g ie endow ment ducted in 2012 that included Egypt, Jordan, .org /2012/02/01/arab-world -s-education Lebanon, and Tunisia from the MENA region -report-card-school-climate-and-citizenship and 21 nations internationally. The results -skills. are based on those who already use social Fox, J. 2014. “Social Accountability: What media sites. Does the Evidence Really Say ?” Global Pa r t nersh ip for Globa l Accou ntabilit y (GPSA) Working Paper No. 1, World Bank, References Wa s h i n g t o n , D C . h t t p s : / /d e m o c r a c y A l - G h a r i n i , K . , I . A . A l - R a s h id i , a n d spotdotnet.files.wordpress .com /2014/09 M. Al-Gamal. 2009. “The Report of the /social-accountability-what -does-evidence National Survey of the Citizens’ Opinions on -really-say-gpsa-working -paper-1.pdf. EFFECTS OF LOWER CITIZEN TRUST ON CITIZEN ENGAGEMENT 249 Gengler, J, M. Tessler, D. Al-Emadi, and A. Diop. cadtm.org/IMG/pdf/ World _Protests_2006 2013. “Civic Life and Democratic Citizenship -2013-Final-2.pdf. in Qatar: Findings from the First Qatar World Ringold, D., A. Holla, and M. Koziol. 2012. Values Survey.” Middle E ast L aw and Citizens and Service Delivery: Assessing the Governance 5 (3). Use of Social Accountability Approaches in Gerges, F. 2014. “ISIS and the Third Wave of the Human Development Sectors. Jihadism.” Current History (December): Washington, DC: World Bank. http://elibrary 339–43. .wo rl d b a n k . o r g /d o i / p d f / 10 .159 6 Hardin, R. 1996. “Trustworthiness.” Ethics 107 /978-0-8213-8980-5. (1): 26–42. Shediac, R., R. T. Shehadi, J. Bhargava, and International Crisis Group. 2012. “Popular H . Sam man. 2013. “G enerations Protest in North Africa and the Middle East A: Differences and Similarities across the Arab (IX): Dallying with Reform in a Divided G enerat ion s.” S t ra t eg y&. ht t p: // w w w Jordan.” Middle East/North Africa Report .strategyand.pwc.com/global /home /what-we No. 118, March 12. http://www.crisisgroup -think/reports-white-papers/article -display .org /~/media / Files / M iddle%20E ast%20 /arab-generations. North%20Africa/Iran%20Gulf/Jordan/118 World Bank. 2007. Decentralization and Local -popular-protest-in-north-africa-and-the Governance in MENA: A Survey of Policies, -middle-east-ix-dallying-with-reform-in-a Institutions, and Practices. A Review of -divided-jordan.pdf. Decentralization Experiences in Eight Middle Khatib, L. 2013. “Corruption in Qatar? The East and North Africa Countries. Link between the Governance Regime and Wash i ng ton , DC: World B a n k. ht t p: // Anti-Corruption Indicators.” Working Paper d o c u m e n t s .wo rl d b a n k . o r g /c u r at e d /e n No. 40, European Research Centre for Anti- / 2 0 0 7/ 0 3 / 1 2 5 6 7 8 9 8 /d e c e n t r a l i z a t i o n Corruption and State-Building (ERCAS). -local-governance -mena -survey-policies h t t p : / / w w w. a g a i n s t c o r r u p t i o n . e u / w p -institutions-practices -review -decentraliza -content /uploads/2013/12/ W P- 40 - Qatar tion-experience-eight -middle-east-north -paper.pdf. -africa-countries. Khatib, L, and E. Lust, eds. 2014. Taking to the ———. 2010. Governance and Anti Corruption Streets. The Transformation of Arab Activism. Diagnostic Survey. Washington, DC: World Baltimore, MD: Johns Hopkins University B a n k . ht t p: // w w w-wd s .worldba n k .or g Press. /external/default/WDSContentServer/WDSP Khemani, S. 2013. “Buying Votes vs. Supplying /IB/2012/08/20/000425962_201208201215 Public Services: Political Incentives to Under 20/Rendered/PDF/705830ESW0P1170usehold Invest in Pro-Poor Policies.” Policy Research 0Final.Report.pdf. Wo r k i n g P a p e r 6 3 3 9, Wo r l d B a n k , Washington, DC. Kilani, S., and B. Sakija. 2002. Wasta: The Data sources Declared Secret . A mman: Jordan Press Arab Barometer, 2006–08 (Wave I), http://www Foundation. .arabbarometer.org/content/arab-barometer-i Middle East Monitor. 2013. “Students’ Protests Arab Barometer, 2010–11 (Wave II), http://www Sweep Egypt.” December 15. https://www .arabbarometer.org/instruments-and-data . m i d d l e e a s t m o n i t o r. c o m / n e w s / a f r i c a -files. http://www.transparency.org/research /8824-students-protests-sweep-egypt. /cpi/overview Morris, S. D., and J. L. Klesner. 2010. “Corruption GAC (Governance and Anti-Corruption) Country and Trust: Theoretical Considerations and Diagnostic Survey, World Bank, http://web Evidence from Mexico.” Comparative Political .worldb a n k .or g / W B S I T E / E X T E R N A L Studies 43 (10): 1258–85. /TOPICS/EXTPUBLICSECTORANDGOVE Olson, M. 1965. The Logic of Collective Action: RNANCE/0,,contentMDK:23399116~page Public Goods and the Theory of Groups. PK : 210 058 ~pi PK : 210 0 62~t he Site Cambridge, MA: Harvard University Press. PK:286305,00.html Ortiz, I., S. Burke, M. Berrada, and H. Cortes. Gallup World Poll, http://www.gallup.com 2013. “World Protests 2006–2013.” Initiative /services/170945/world-poll.aspx for Policy Dialogue and Friedrich-Ebert- Governance and Local Development survey, Stiftung New York Working Paper. http:// Prog ra m on G over na nce a nd L o c a l 250 TRUST, VOICE, AND INCENTIVES Development, Yale University, http://gld Pew Research Center, http://www.pewresearch .commons.yale.edu/research/ .org PETS (Public Expenditure Tracking Survey), Tra nsit iona l G over na nce Projec t , ht t p: // World Bank, ht tp: //web.worldbank.org transitionalgovernanceproject.org/ /WBSITE/EXTERNAL/TOPICS World Values Survey, 2010–14 (Wave 6), http:// /EXTSOCIALDEVELOPMENT/EXTPCE w w w.w o r l d v a l u e s s u r v e y. o r g / W V S NG/0,,contentMDK:20507700~pagePK:148 DocumentationWV6.isp 956~piPK:216618~theSitePK:410306,0 0 Yemen Polling Center, http://www.yemenpolling .html .org/consultancy/ New Openings PART V I n this report, we have analyzed the expectations and realities in the delivery of education and health services in the Middle East and North Africa (MENA), and we have explained these in the context of the cycle of performance. Chapter 1 revealed that citizens of the MENA region are not satisfied with services and expect their governments to improve them. Chapter 2 cautioned that, although citizens of the MENA countries largely enjoy access to education and health services, they face significant challenges in the quality of services. Chapter 3 then described some local communities that have overcome these quality challenges. Building on chapter 4, which illustrated the historical and cultural context within which the cycle of performance operates in the MENA region, chapter 5 explored the influence of state and nonstate institutions on state performance and underlined the weaknesses in the existing accountability mechanisms facing policy makers, public servants, and service providers in most MENA countries. Chapters 6 and 7 illus- trated how these institutional weaknesses affect the efforts of teachers and health workers to deliver services, and how the effects may differ in different local contexts. Finally, chapters 8 and 9 discussed how service delivery and state performance affect citizens’ trust in public institutions, and how such trust translates into the nature of citizen engagement. Because of the complex circumstances facing MENA countries, it is necessary to build on evidence of local successes and positive trends that buck the cycle of gener- ally poor performance. In this part, chapters 10, 11, and 12 identify the bases for improvement and encouragement—for citizens, civil servants, policy makers, and donors alike—to act on those successes and trends. As we discuss in chapter 10, con- fl icts, crises, and political transitions may give national and local leaders a unique opportunity to tackle service delivery challenges as well as boost trust and construc- tive engagement. In chapter 11, we acknowledge that donors, including the World Bank Group, need to learn from their own (often failed) efforts to support education and health services. Finally, chapter 12 suggests possible incremental approaches to systemic reforms, options for empowering communities and local leaders to find local solutions, and possible quick wins. 251 Transitions, Conflicts, and Refugees: Both an Opportunity 10 for and a Strain on the Cycle of Performance • Dramatic changes such as the regime transitions, conflicts, and refugee crises facing the Middle East and North Africa region today strain the cycle of performance, but they also can spark a needed change. • Transitions in power and the uncertainty they bring initially undermine political and administrative institutions as well as performance, but they do offer opportunities for positive reforms. They also affect trust and engagement in a positive way—at least at first. • A conflict destroys institutions while increasing the demand for services. At first, performance declines. And yet the conflict provides a rallying point for citizens, and ultimately trust, engagement, and institution building can be restored. • An influx of refugees also places high demands on service delivery and opens the possibility of institutional reform. Incumbents may find it necessary to establish new institutional arrangements in response to the stress, and these institutions may prompt better performance. T his chapter explores how the dramatic with a framework through which we can challenges facing the Middle East consider how these changes impinge on ser- today affect the possibilities for vice delivery. Political transitions, confl icts, change in the cycle of performance (figure and refugee crises have affected, and may be 10.1). We take as a starting point that crises expected to affect, the cycle of performance can open possibilities for reform. They can at various points, altering engagement, insti- disrupt existing institutions, alter elite coali- tutions, performance, and trust. Moreover, tions, and place actors—citizens, policy mak- political crises sometimes create countervail- ers, and elites alike—in a domain of losses,1 ing pressures—for example, heightening making them more likely to take risks engagement while diminishing trust. (Skocpol 1979; Keeler 1993; Weyland 2004; In this chapter, we draw on diverse empiri- Hunter 2006). The performance cycle intro- cal and theoretical findings from within and duced in the previous chapters provides us beyond the Middle East and North Africa 253 254 TRUST, VOICE, AND INCENTIVES FIGURE 10.1 The cycle of performance uprisings led to conflicts that failed to over- throw regimes—a short-lived uprising in Bahrain and a much longer one in the Syrian Citizens’ engagement: Arab Republic that prompted regional inter- formal and informal vention led to refugee crises and created state- less spaces that were fertile ground for militant organizations. A small set of countries, includ- ing Morocco, the Gulf states, and Jordan, largely avoided (at least until now) major Institutions: changes, in part by undertaking reforms Cycle of political aimed at increasing public support for the performance at administrative, regime in power. In short, the MENA region Citizens’ trust in and social the national and is currently in the midst of dramatic change. public institutions local levels Transitions In Egypt, Libya, and Tunisia, the Arab Uprisings saw long-standing dictators fall from power, beginning transition processes Performance: effort and ability driven from below. Activists took to the to meet citizens’ needs streets, old ruling parties were discredited, and party buildings were set ablaze. Since then, these countries have seen new political parties emerge and elections called. Their citi- (MENA) region. The evidence is too weak zens are engaged—in the streets, through and the processes are too stochastic to argue social media, and at the polls, and debates that the impacts we point out are inevitable or abound over a variety of issues, including elec- exhaustive. Nor are we arguing that these are tion rules, constitutional changes, transitional the only pressures that affect the cycle. Rather, justice, and decentralization. Transitions, the goal of this discussion is to stimulate then, impinge on all parts of the cycle: institu- thinking about opportunities for the future. tions, performance, trust, and engagement. Such thinking is particularly important in the MENA region today as it witnesses the Institutions most intense regionwide transformations in at least a century. The Arab Uprisings in The very essence of a transition is a struggle 2010–11 began a political rupture, the rum- over the rules of the game, and thus transi- blings of which had been heard for a decade. tions entail changes in both the strength and The turmoil that started in Tunisia soon the form of institutions at all levels of the sys- spread across the region, taking different tem. The extent and nature of institutional forms in different places. The Arab Republic change during transitions are driven by vari- of Egypt, Libya, and Tunisia overthrew long- ous factors: the preexisting institutional standing leaders, and the Republic of Yemen strength and form, state resources and capac- replaced President Ali Abdullah Salah, albeit ity, leadership, and heightened engagement. with a handpicked successor in the hope of a In the short run, the uncertainty inherent carefully managed transition. The ruptures in transition periods weakens institutions; put these countries on transition courses that rules and norms are uprooted and new ones varied, from a reversion to authoritarianism in are not yet established. And yet the extent to Egypt to relatively successful democratization which a transition undermines the provision in Tunisia to the breakdown into conflict in of public services may depend on the nature Libya and the Republic of Yemen. Elsewhere, of the state’s political institutions at the TRANSITIONS, CONFLIC TS, AND REFUGEES: BOTH AN OPPORTUNIT Y FOR AND A STRAIN ON THE CYCLE OF PERFORMANCE 255 outset of the transition. When a strong state social actors and institutions, which often was in place prior to the political rupture, a come into conflict with the state and non- transition may be less likely to undermine the state institutions in place before the provision of services. For example, Tunisia transition. For example, in Egypt’s Sinai the enjoys a more developed bureaucracy and transition resulted in the spread of Islamic- stronger state than Egypt and Libya. Even based unofficial courts, which have come though strikes in Tunisia have crippled the into confl ict with both state and previously economy and the government has not been state-tolerated tribal courts (Revkin forth- able to pursue reforms as effectively as citi- coming). Observers argue that new groups zens have demanded, it does not appear that (gangs or mafias) are emerging even in areas entrenched “deep state” forces have been able around Cairo because of the state’s inability to sabotage services to achieve their own to meet citizens’ needs.3 Similarly, in Iraq the political gains. 2 By contrast, in Egypt the informal economy grew considerably after state is weaker, and the political forces deter- 2003, especially in the areas of small-scale mined to stymie the transition were able to urban services, and by 2006 an estimated sabotage service provision (especially electric- 80 percent of the labor force was engaged in ity, water, and solid waste management) an informal economy (Looney 2006). to achieve personal gain (Hubbard and Finally, transitions provide an opportunity Kirkpatrick 2013). At the extreme, in Libya to debate and form new institutional arrange- the state was too weak to retain control over ments, placing new items on the agenda. For strategic resources and institutions, ulti- example, in transitioning countries universal mately resulting in a civil war that, as we dis- health care has been the subject of dialogues cuss shortly, is undermining education, over new constitutions and policies, and health, and other services. it has appeared in party programs and in Transitions allow space for innovation as increasingly vocal advocacy campaigns (Saleh well as the possibility of decentralization and et al. 2014). Similarly, although decentraliza- local solutions. When national and regional tion has long been on the reform agenda, it is controls in what were previously highly cen- now debated with new interest. Moreover, tralized systems weaken, local leaders are able countries in transition receive more attention to implement innovative solutions. School and the resources and expertise needed to put principals, chief medical officers, municipal in place such reforms (Institute for Integrated councils, and directorates previously con- Transitions 2013). Not everyone welcomes strained by inefficient arrangements can cre- such an intervention, but it may make it dif- ate new ones. For example, a study of Egypt’s ficult for naysayers to avoid reform and education system based on open-ended sur- strengthen the hands of reformers. At least in veys of school administrators found that the the short run, there are pressures for institu- early period of crisis in Egypt prompted tional reforms that promise greater service administrators to make decisions, reformulat- provision. ing policies, procedures, and responsibilities within their schools. This freed savvy admin- Performance istrators from sometimes unnecessary, ineffi- cient constraints, although it also left less Unfortunately, transitions are likely to reduce adaptable ones without support. The crisis a state’s ability to provide services in the near was thus “both a threat and an opportunity” term. Demands for education, health, and for reform (Rissmann-Joyce 2014). other services increase, but the MENA coun- In the same vein, transitions open the way tries in transition have thus far seen a decline for nonstate actors to fill the vacuum left by in performance. the weakening state. Local governance, In part, performance suffers due to the including dispute resolution, security, and uncertainty surrounding transitions, the out- service provision, can be captured by new flow of experienced personnel, and the 256 TRUST, VOICE, AND INCENTIVES fewer resources. The extent to which these state’s performance, and leadership. In the changes occur depends in part on choices early period of the transition, trust—or at made during the transition. The removal of least optimism—skyrockets. Not only have personnel associated with the ancien regime the leaders so closely associated with repres- (often in response to popular demands) and sion and exploitation been removed, but the appearance of new entrants as political regime change that was unthinkable only a appointments can leave ministries and state short time before has been achieved. People offices with inexperienced staffs. The unsta- have high hopes for the future. It is a period ble environment of transitions can also lead that Leszek Balcerowicz, the Polish economist to a loss of resources. Tourism declines, and former fi nance minister, has called the which had a particularly high impact in period of “extraordinary politics,” in which Tunisia and Egypt. Seven percent of the gross people “are pretty euphoric” because of their domestic product (GDP) in Tunisia (UNWTO “freshly regained freedom” (Balcerowicz and 2011), accounting for 450,000 jobs, is based Gelb 1995; Balcerowicz 1997). on tourism; 5.6 percent in Egypt.4 Moreover, According to the polls, at least initially an unprecedented freedom to strike and dem- trust is restored. In 2013, citizens in the tran- onstrate cripples production. In Tunisia, for sitioning countries anticipated better pros- example, work stoppages in the state-owned pects for good governance, less corruption, Gafsa Phosphate Company caused output to an improved economy, and better security plummet from 7.5 million tons before 2011 in the future. This stood in sharp contrast to 3.3 million tons in 2014 (Saleh 2014). to those in economies that did not undergo Volunteerism alleviated some of these pres- political rupture, where majorities of citi- sures in the early days of the transition. A zens expected these conditions to worsen sense of pride and nationalism can mobilize (figure 10.2). citizens to organize food drives for the poor, This period of heightened trust can be clean their neighborhoods, or otherwise short-lived, however. In Egypt, distrust contribute material and human resources. 5 between elites who had, hand in hand, taken But this mobilization diminished rapidly, to the streets began to emerge within months leaving transitional governments strapped for of President Hosni Mubarak’s downfall. resources. By April, non-Islamist activists and party At the same time, transitions create addi- leaders were grumbling that the Muslim tional demand for education and health Brotherhood was colluding with the military, services. In part, heightened expectations while Islamists were warning that non- drive the demand. The abrupt change has Islamists would corrupt Egypt’s society. physical and mental repercussions as well. Distrust between the Copts and Muslims also Indeed, as Mohamed Elmahdy, professor of emerged. In Tunisia, similar distrust among psychiatry at Al-Azhar University, has noted, citizens took shape. In response to repeal of the uncertainty, insecurity, and feelings of the ban on headscarves and the appearance loss that follow revolutionary change lead to of displays of religiosity in the public sphere, a spike in depression, post-traumatic stress Islamists’ opponents often decried the social disorder (PTSD), and other psychological change. Overall, citizens’ trust in institutions problems that require attention (Elmahdy and actors has plummeted in transitional 2012). 6 These issues manifest themselves countries in the MENA region.7 both emotionally and physically. Ultimately, people decide whether they trust state institutions and actors based on the ability of the latter to provide services, Trust maintain stability, and stimulate eco- Transitions also affect citizens’ trust in public nomic growth. Indeed, in North Africa’s institutions. In general, trust derives from citi- transitional countries a majority of citizens zens’ historical experience with the state, the believe they are worse off today than before TRANSITIONS, CONFLIC TS, AND REFUGEES: BOTH AN OPPORTUNIT Y FOR AND A STRAIN ON THE CYCLE OF PERFORMANCE 257 FIGURE 10.2 Survey of citizens of transitioning economies: Prospects for good governance, the economy, government corruption, security and safety, the overall economy, and quality of government leadership: MENA region, 2013 a. As a result of the recent protests and revolts in the Arab world, will the prospects for good governance get better or worse in these economies? 100 90 80 70 60 Percent 50 40 30 20 10 0 ya . p. sia co q za ria n an ep no Ira Re Ga Lib oc rd ge ni ,R ba Tu Jo or b Al nd en ra Le M m ka t, A Ye an yp tB Eg es W b. As a result of the recent protests and revolts in the Arab world, will the economic prospects get better or worse in these economies? 100 90 80 70 60 50 Percent 40 30 20 10 0 ya . p. sia q co ia za an n ep Ira r no Re Ga Lib oc rd ge ni ,R ba Tu Jo or b Al d en ra an M Le m ,A k Ye t an yp tB Eg es W Stay the same Get worse Get better (continued next page) 258 TRUST, VOICE, AND INCENTIVES FIGURE 10.2 Survey of citizens of transitioning economies: Prospects for good governance, the economy, government corruption, security and safety, the overall economy, and quality of government leadership: MENA region, 2013 (continued) c. As a result of the recent protests and revolts in the Arab world, will corruption in government get better or worse in these economies? 100 90 80 70 60 Percent 50 40 30 20 10 0 ya sia Re ab ria co Lib r ge ni oc p. t, A Tu Al or yp M Eg d. As a result of the recent protests and revolts in the Arab world, will security and safety get better or worse in these economies? 100 90 80 70 60 Percent 50 40 30 20 10 0 . ya a p. q ria za an on co ep si Ira Re Ga Lib rd ge an ni oc ,R Tu Jo b Al b or nd en ra Le M m ka t, A Ye an yp tB Eg es W Stay the same Get worse Get better (continued next page) TRANSITIONS, CONFLIC TS, AND REFUGEES: BOTH AN OPPORTUNIT Y FOR AND A STRAIN ON THE CYCLE OF PERFORMANCE 259 FIGURE 10.2 Survey of citizens of transitioning economies: Prospects for good governance, the economy, government corruption, security and safety, the overall economy, and quality of government leadership: MENA region, 2013 (continued) e. As a result of the recent protests and revolts in the Arab world, will the overall economy get better or worse in these economies? 100 90 80 70 60 Percent 50 40 30 20 10 0 ya p. ria co sia Re Lib ge oc ni b Al or Tu ra M t, A yp Eg f. As a result of the recent protests and revolts in the Arab world, will the quality of government leadership get better or worse in these economies? 100 90 80 70 60 Percent 50 40 30 20 10 0 ya sia p. ria co Re Lib ge ni oc Tu b Al or ra M ,A pt y Eg Stay the same Get worse Get better Source: Gallup World Poll, 2013. 260 TRUST, VOICE, AND INCENTIVES 2011 (Transitional Governance Project). 8 Indeed, in the MENA region as in transi- Moreover, since the transitions began, tions elsewhere12 the fall of the regimes Tunisians and Egyptians have tended to rate opened the floodgates, giving citizens a economic concerns as the most important chance to mobilize and voice demands. issues facing their country—64 percent9 in Tunisian teachers went on strike within Tunisia (Transitional Governance Project) weeks of President Zine El Abidine Ben Ali’s and more than 50 percent in Egypt.10 In downfall, demanding improvements in the Libya, the vast majority see security as the country’s education system and the status of main issue.11 All three countries believe their teachers (Education International 2011, governments and political parties are not 2012). Egyptian doctors staged sit-ins at hos- dealing with these problems effectively. pitals, transit and dockworkers went on strike, students took to the streets, and mill workers called for better working conditions Engagement (Faiola 2011; Rollins 2014). After the fall of Finally, transitions change citizen engage- President Ali Abdullah Saleh’s government, ment with the state and offer an opportunity the Republic of Yemen saw strikes from port to break out of the low-efficiency equilib- workers, employees at the Ministry of Youth rium. Engagement was not entirely absent and Sports, postal workers, bus drivers, before the 2010–11 uprisings, but the forms health professionals, and, perhaps most visi- that it took were often indirect and under- bly, garbage collectors, who organized a ground (Khatib and Lust 2014). Particularly nationwide strike that lasted in some parts of in Egypt, Libya, and Tunisia, where incum- the country up to three months, “filling the bent regimes were pushed from power, new cities with the unbearable stench of filth and political parties were allowed to form, previ- decay” (Alwazir 2012). These calls for better ously banned actors were allowed to engage, conditions and an end to corruption had and new civil society organizations were severe repercussions for the education and established. For example, in Egypt the health systems and often frightened away Muslim Brotherhood, which previously was international investors. officially prohibited from engaging in poli- Newfound freedoms of association led to tics, came into power; political parties were the formation of new organizations and established for the first time in Libya; and in greater mobilization. For example, after the all countries the previously strong grip of the fall of Mubarak Egypt saw the formation ruling parties was eliminated. These events of two new labor unions: the Egyptian offered a chance for political participation to Federation of Independent Trade Unions move from the informal to the formal sphere (EFITU), with 261 new trade unions and (Khatib 2013). some 2 .45 million members, and the However, opp or t u n it ie s were not Egyptian Democratic Labor Congress expanded for everyone. Today, those once (EDLC), with 246 unions. The EFIT U close to the former regime often fi nd them- included the Independent Teachers Union of selves unable to participate because of either Egypt (ISTT), which sought to move teachers popular pressure or lustration laws. For many on temporary contracts to permanent status others, however, the previous constraints on after they had been in a post for three years, voice and participation have faded, and, at to advocate for wage increases, and to pro- least in the initial periods, they are eager to vide teachers with greater leverage in the engage the state, often for the first time. They teaching process. As Ayman Albaili of the make demands in strikes and demonstra- ISTT explained, “The previous regime con- tions, speak their mind through social and sidered education as a commodity which par- traditional media, run in elections, and flock ents could afford or not. The role of teachers to the polls. Information flows more easily, was marginalised in the decision making and there is high public engagement. process on education. Our union wants to TRANSITIONS, CONFLIC TS, AND REFUGEES: BOTH AN OPPORTUNIT Y FOR AND A STRAIN ON THE CYCLE OF PERFORMANCE 261 reassert teachers in the education process and transition saw an upsurge and then a decline improve their status and rights” (Education in the enthusiasm for participation. At the International 2011). Labor strikes almost national level, citizens are refraining from doubled in the year of the revolution (see voting. In Egypt, only 47.14 percent of the table 10.1). voting-age population went to the polls in the And yet over time citizens’ enthusiasm for 2014 presidential elections (International voicing demands begins to flag, placing con- IDEA, 2015), even though the voting period straints on these movements. People become was extended and citizens were pressured to tired of clogged streets, nonfunctioning turn out to the polls. By contrast, in 2012, industries, and disrupted services, and many 49.14 percent turned out to vote (International even become willing to support the introduc- IDEA, 2015). In Tunisia, the parliamentary tion of measures that would clamp down on elections also saw diminished enthusiasm. In mobilization. Thus Egypt’s Presidential 2012, 82 percent of Tunisians surveyed said Decree 107 enacted in 2013 (the Protest Law) they planned to go to the polls for the next was subjected to less backlash than one might election, but as the election approached in have predicted at the downfall of Mubarak. 2014, this number dropped to 54 percent The law requires organizers of public gather- ( Tr a n s i t i o n a l G ove r n a n c e P r o j e c t). ings to notify the police at least three days in Eventually, only 45.39 percent of the voting- advance of a campaign with specific informa- a ge p opu l at ion t u r ne d out to vot e tion on the place of gathering or route of the (International IDEA, 2015).13 In Libya, too, procession, the start and end times of the engagement in the political process has event, the subject of the event, the demands declined, with 48.72 percent voting in the and slogans featured at the event, and the fi rst General National Congress elections of names of the individuals or group organizing July 2012 and only 15.64 percent in the par- the event, with a place of residence and con- liamentary elections of 2014 (International tact information (Article 8). Arguably, the IDEA, 2015). enthusiasm for demonstrating also dies down Perhaps most important, however, transi- because the results are minimal. Far from tions allow the development of new efforts to creating better economic conditions, mobili- monitor and hold providers accountable. zations are blamed for contributing to the Community monitoring and initiatives economic slowdown. Most citizens do not see by citizens and civil society organizations their conditions improve, and so they lose emerge (such as client surveys, feedback faith that the government can solve their mechanisms, score cards, polls, independent problems. Engagement slowly returns to pre- research) to highlight failures and hold transition levels. administrative and political actors account- Engagement in the formal political process able. For example, in Tunisia Al-Bawsala, an also spikes but then diminishes over time. In independent nonprofit nongovernmental both Egypt and Tunisia, for example, the organization, provides citizens with daily updates on the workings of the Constituent Assembly, brings representatives and their Table 10.1 Labor strikes: Arab Republic of Egypt, 2006–12 constituents together in at times combat- ive public meetings, and to a lesser extent Year Number of labor strikes engages in mobilizing communities to solve 2006 222 local issues.14 2007 614 Such efforts are not entirely new, nor are 2008 609 they restricted to transitioning countries. For 2009 700 example, in Jordan and the Republic of 2010 584 Yemen providers instituted 24-hour service May 2011 – April 2012 1,137 hotlines as early as 2007. And in Jordan, Source: Abdalla 2012, 2. organizations such as Leaders of Tomorrow 262 TRUST, VOICE, AND INCENTIVES used creative techniques and convened com- The Libyan transition has deteriorated into munity meetings to foster information shar- civil war, and the Republic of Yemen threat- ing and communication between citizens and ens to do the same. Calls for change were policy makers. Yet these efforts often face repressed brutally in Bahrain and the Syrian obstacles and resistance from entrenched Arab Republic, with the Syrian civil war con- forces within established regimes. In transi- tinuing in full force. Violence has escalated as tioning countries, governments taking their well in Iraq. Ungoverned spaces in Syria and seats for the first time are more likely to Iraq have given rise to the Islamic State of embrace such initiatives. Thus, for example, Iraq and Syria (ISIS), which is expanding in Tunisia in 2011 the office of the prime with bloody vengeance. Moreover, the pres- minister launched its first national scorecard sures in neighboring states, combined with for 10 public services; 8,500 citizens partici- unresolved domestic conflicts, have led to the pated, and the results were published online. outbreak of violence in Lebanon. And fight- A second national scorecard on services and ing between Israel and the West Bank and benefits from the National Health Insurance Gaza rose to devastating levels throughout Fund was launched in 2012. the summer of 2014. In short, various forms Very different transition processes can of civil and interstate war are spreading unfold, however, affecting the opportunities across the region. Such wars entail a number for engagement. In Egypt, the rollback of of challenges: the diminished capacity of the institutional protections for freedom of state to control its population and administer speech, protest, and party participation— services; an increase in the need for educa- highlighted by the banning of the Muslim tion, health, and other services in the face of Brotherhood and April 6 movements, prohi- declining resources; and the emergence of bition of the Freedom and Justice Party, mass new actors and forms of engagement. Wars arrests, and death sentences—sent a clear may also affect citizens’ trust in public insti- message to citizens that they have limited tutions and their engagement, although how latitude to voice their demands. It is not sur- they do so is less transparent. prising, then, that citizens and elites alike are Before turning to each of these issues, it is choosing to step back from engaging the important to note that the impact of civil state. As one revolutionary activist and for- wars on the cycle of performance is likely to mer member of parliament (MP) explained, vary, depending on several factors. In gen- now is the time to seek to address people’s eral, as Stewart, Huang, and Want (2000) basic needs, not to put forth platforms or run note, the costs of confl ict are highest when in elections.15 By contrast, in Tunisia citizens conflict is geographically pervasive and are frustrated with political parties and a government is undermined to the point democracy, but they do not face the same that it is unable to effectively collect taxes or institutional constraints as Egyptians in provide services. The impacts of conflict voicing demands. Elites of all political also vary with a set of interconnected fac- persuasions continue to engage in politics, tors: the duration of the confl ict, the level mobilizing supporters in part by promises of of international intervention and support, public services and economic growth. Finally, the nature of the economy (such as the extent in Libya it is the weakness of the state that of subsistence agriculture, reliance on inter- ultimately prompted the disintegration of national trade, dependence on few indus- order, with many choosing bullets over tries), institutions (for example, the strength ballots. of civil society organizations and quasi- governmental and governmental institu- tions), and the character of the society (such Conflict as the initial level of poverty and vulnerabil- The MENA region has also witnessed a ity, social support networks)—see Stewart horrific escalation of conflict since 2011. and Humphreys (1997). TRANSITIONS, CONFLIC TS, AND REFUGEES: BOTH AN OPPORTUNIT Y FOR AND A STRAIN ON THE CYCLE OF PERFORMANCE 263 Institutions nations engaged in long civil wars actually experience an increase in economic activity Civil wars affect institutions primarily by after the cessation of conflict, whereas those diminishing the state’s ability to control geo- engaged in short civil wars do not. Conflict, graphic areas and populations. The level of particularly long civil wars, may undermine control may change over time, as the back old institutions but, as just noted, lead to the and forth of government and rebel control emergence of new ones. When new institu- over swaths of land and populations in tions are developed in a context in which Syria demonstrate. When and where the gov- mobilizing citizen support and resources is ernment is not in control, new and diverse crucial, they may ultimately promote greater institutions and governance arrangements levels of voice and accountability. In the short emerge under nonstate actors. Indeed, as run, however, the war is likely to undermine Watkins (2014) argues, “confl ict zones are the provision of services. characterized by states that have lost this monopoly in reality or in the eyes of their people. Thus, what is ‘legitimate’ becomes Performance unmoored from its Weberian foundation. In this context, alternative forms of power, con- The death and destruction of war increase trol, and coercion develop to fill the void. the demand for services and simultaneously Nowhere is this void more visible than at the make it more difficult for the state to provide margins of the state where warlords and non- them. The need for health services escalates state armed groups principally operate” (see to treat injuries as well as the increased prev- also Schneckener 2006). alence of distress. Indeed, the World Health Even where the state remains in control, Organization estimates that between one- central authority over outlying areas is third and one-half of the individuals who diminished, with attention diverted to the witness such political violence experience war effort. This opens the way for local elites PTSD, depression, anxiety, and other mental and providers—both from the state and from disorders (Sausa 2013). outside of it—to establish new rules and The demands for services are compounded procedures. by the need to repair the infrastructure for Several studies of the long-term impacts of education, water, electricity, and transporta- civil war suggest that some such wars pro- tion destroyed in the war. Education and mote stronger institutions that may even health inputs are limited by the destruction of encourage accountability. Bellows and Miguel facilities, the loss of teachers and health care (2006), examining the community-level workers through death or migration, and the effects of the 11-year civil war in Sierra Leone, inability of families to contribute to educa- found no evidence of uniform, persistent tion costs.17 At the same time, the war forces adverse effects of the civil war violence on the government to shift resources to military local institutions. Violence destroyed some of support. Meanwhile, the economy is stifled the existing institutions, but also led to the by the decline in tourist receipts, foreign creation of new ones. In general, they found direct investment, remittances, and industrial that areas that had experienced violence were output, and, in some cases, by international associated with higher levels of mobilization sanctions (Stewart, Huang, and Want and collective action, which could possibly 2000).18 Gupta et al. (2004) have argued that lead to better postwar political accountability. governments resolve economic problems by This fi nding fits well with those of Levitsky sacrificing macroeconomic stability rather and Way (2012) and Bermeo (2014), who than by cutting spending on social services have argued that revolutionary conflict makes (perhaps because of the incumbents’ need to regimes more resilient.16 It may also help maintain public support), but it is neverthe- explain the finding by Collier (1999) that less clear that confl ict hinders performance 264 TRUST, VOICE, AND INCENTIVES (see also Stewart, Huang, and Want 2000; For example, in Syria and Bahrain, doctors Blomberg, Hess, and Orphanides 2004). caring for opposition fighters reportedly have Conflict affects achievements in education been targeted by the regimes in power or and health in both the short and long run. imprisoned, and in Syria, some hospitals Cross-national studies of health and conflict have been taken over by the national army, have found confl ict associated with under- leaving opposition forces unable to seek nourishment, lower height, reduced life medical attention without facing arrest and expectancy, and higher infant mortality rates torture (Batniji et al. 2014). There is good (Stewart, Huang, and Want 2000; Ghobarah, reason to believe that such lopsided access to Huth, and Russett 2004; Gates, Hegre, and health care is a problem in rebel-held areas as Nygård 2012). Often, women suffer espe- well. Indeed, such politically charged conflict cially from lack of health services during environments place well-intentioned health pregnancy and childbirth (Stewart, Huang, care workers in precarious positions, making and Want 2000). The extent to which the it even more difficult for them to provide health system deteriorates appears to depend services. in part on the extent to which a government Where the infrastructure is weak at the was committed to providing services before outset, problems are likely to be particularly the war. In the past, some governments such acute. For example, in the Republic of Yemen as El Salvador, Mozambique, and Nicaragua the health system comprised only 842 public remained committed to social objectives even health centers, which reached only 68 percent during war, while others such as Liberia, of the population, at the outset of the conflict Somalia, and Uganda invested little and real- in 2011. Not surprisingly, as the conflict con- ized particularly poor infant mortality rates tinued, citizens’ conditions deteriorated, par- relative to their GDP (Stewart, Huang, and ticularly in the center and west of the country. Want 2000). The Joint Economic and Social Assessment The negative impacts of conflict on health for the Republic of Yemen explained: are readily apparent in the MENA region. For example, the sanctions placed on Syria All sectors of life in Yemen were affected since May 2011 have led to an increase in the during 2011, and as might be expected costs of health care and pharmaceuticals and the nutrition situation deteriorated to a shortage of medical supplies not pro- during that time. Given the weak duced locally, including those for cancer, dia- starting point . . . , most vulnerable betes, and heart disease. Sanctions have also Yemeni families began 2011 ill-prepared resulted in power shortages, which have to absorb the shocks induced by the exposed the vulnerable to extreme tempera- crisis. As a result, conflict, general tures and interrupted vaccination programs insecurity, and rising prices for essential (Sen 2013). In Iraq, sanctions led to a rise in commodities aggravated hunger and infant mortality from 47 to 108 deaths per malnutrition. Vulnerable populations 1,000 births from 1994 to 1999, with the experienced a lack of access to food, mortality rates of children under 5 increasing reduced diversity in diets, changes in from 56 to 131 per 1,000 births (Batniji et al. breastfeeding practices as a result of the 2014). In the West Bank and Gaza, the sanc- crisis, and reduced access to clean water tions, occupation, and warfare related to because of fuel prices, which led to the long-standing conflict with Israel have catastrophic results. Compounding the increased food insecurity: 33 percent of the above, the overall breakdown of social residents of the West Bank and Gaza did not services as a result of the unrest created have enough food in 2013, compared with conditions whereby addressing the 27 percent in 2011 (UNRWA 2014). intensifying problem of children’s Political tensions also undermine the abil- health and nutrition became even more ity of health care workers to provide services. challenging. (World Bank 2012, 81) TRANSITIONS, CONFLIC TS, AND REFUGEES: BOTH AN OPPORTUNIT Y FOR AND A STRAIN ON THE CYCLE OF PERFORMANCE 265 Meanwhile, long-standing conflicts and The costs of conflict affect segments of occupation can undermine good governance society unevenly. In some countries such as mechanisms, making the provision of public Syria and the Republic of Yemen, it is the health care difficult, even when the confl ict most vulnerable who are unable to afford pri- is not raging. Giacaman et al. (2009) aptly vate medical services and whose situation describe the situation in the West Bank deteriorates most precariously (Batniji et al. and Gaza: 2014). Often, impacts are also felt geographi- cally. For example, in the West Bank and The absence of any control by the Gaza malnutrition is geographically clustered, Palestinian National Authority over determined more by access to food and road water, land, the environment, and closures than by financial means. So, too, m o ve m e n t w it hi n t he oc c u pi e d during the conflict in Iraq, childhood mortal- Palestinian territory has made a public- ity was regionally determined because those health approach to health-system living in areas under UN sanctions experi- development difficult, if not impossible. enced greater increases in mortality rates than These issues have been exacerbated by those in the autonomous northern regions. the dysf u n c tio n al politic al a n d Geographical concentrations of conflict institutional systems of the authority; are particularly evident when conflicts lead to the damaging effects on ministries widespread internal displacement. The prob- of using the authorit y resources lems mirror those of refugees. Put briefly, for patron age to secure loyalt y; “the sudden and rapid movement of refugees marginalization of the Palestinian and IDPs [internally displaced persons] into Legislative Council; and corruption and low- and middle-income cities can generate a cronyism, all of which led to a rapid host of negative shocks. When arriving in increase in the number of health-service large numbers, new arrivals can generate employees of the Palestinian National stresses on already dilapidated water and Authority without evident improvement sanitation services, exacerbate conflicts over in the quality of health services. tenure and access to land, and generate com- Conflict affects education as well. Schools petition for resources with host populations” are destroyed, and school calendars are dis- (Muggah 2012). rupted. Parents often keep their children at home, fearing for their safety, and children Trust and engagement who do make it to classrooms find it difficult to concentrate on their studies. As noted in The existing evidence on the linkages World Development Report 2011, “when between conflict and trust and engagement is children are late coming home, a parent has less clear. The extent to which citizens con- good reason to fear for their lives and physi- tinue to trust public institutions is likely to cal safety. Everyday experiences, such as depend on their prior relationship with the going to school, to work, or to market, regime, as well as the extent to which they become occasions for fear” (World Bank believe the regime will prevail. On the one 2011). The conflict in the summer of 2014 in hand, political violence undermines trust the West Bank and Gaza destroyed 26 between citizens and the state, between pro- schools and partially destroyed 232 schools, viders and ministry officials, and between forcing delay of the start of school (The citizens themselves (Sausa 2013). Conflict Guardian 2014).19 Longer wars, which lead situations may thus strengthen citizens’ reli- to displacement of families, are likely to have ance on their social kin, with the result that even greater impacts. Often, as studies of they feel and act less like citizens and more Cambodia, Guatemala, and Rwanda have like family or tribal members. shown, the impacts of conflict on the educa- At the same time, confl ict may push citi- tion system are long felt.20 zens seeking protection to trust the state 266 TRUST, VOICE, AND INCENTIVES more deeply. Indeed, García-Ponce and provide catalysts for social confl ict that can Pasquale (2014) have found that people diminish trust in public institutions, and they exhibit more trust in their national govern- face unique constraints and incentives that ment and head of state when they have suf- shape engagement. fered from political violence. They argue that citizens see the government as capable of Performance imposing order and providing security. Yet they also note that people may fear govern- We begin this discussion of refugees with ment repression and thus express greater performance because it is the step in the cycle levels of trust even when they actually feel of performance most clearly affected by refu- differently. gee crises. Refugees place additional demands Civil war can also alter the ability of citi- on a state, vying with nationals for educa- zens to engage by changing constraints and tion, health care, housing, and other goods their access to information. The change in and services. And the additional burden can constraints is twofold. First, the deteriora- be exceptionally high; refugees tend to be tion of order opens up new options and disproportionately women, children, and incentives for engagement, including fi nan- the elderly; the healthy younger men leave to cial incentives to join armed groups. Second, fight (Lischer 2009). at the same time in both opposition- and Today, the MENA region is particularly government-controlled areas, the space for affected by the confl icts in Iraq, Libya, and dissent is often limited because purported Syria. The Syrian conflict has led to a refugee “traitors” are punished. crisis that heavily affects its neighbors. As of Finally, in addition to these constraints September 2014, 1,185,275 Syrian refugees there is reason to question the extent to were in Lebanon (24.58 percent of its popu- which citizens engage in contributing to or lation), 615,792 in Jordan (over 7.75 percent demanding improvements in health, educa- of its population), 843,625 in Turkey tion, or other services in the midst of ongoing (1.1 percent of its population), and 215,303 conflicts. For example, rather than seek edu- in Iraq (0.60 percent of its population). These cation boys in conflict zones often leave refugees were primarily located in northern school to join the fighting or support their Kurdish areas (2.58 percent of the popula- families. This may help explain why the ratio tion of Iraqi Kurdistan). Elsewhere, data of girls’ to boys’ primary enrollment, as well from the United Nations High Commissioner as illiteracy ratios, improve during war years for Refugees (UNHCR) for January 2014 (Stewart, Huang, and Want 2000; Sausa indicated that about 3,300 refugees and 2013). However, it also suggests that engage- 53,600 internally displaced persons were in ment in education diminishes during war. Libya (UNHCR 2014). After the escalation of violence in Libya in the summer of 2014, reportedly up to one-third of the population Refugees (about 1.8 million Libyans) fled to Tunisia Confl icts also push refugees to neighboring (Gall 2014). A similar spike in IDPs has been states, thereby affecting the education, health, seen in Iraq and Syria as ISIS has expanded and other systems there as well. According to across the region.21 World Development Report 2011, neighbor- The extent to which the service delivery ing countries host nearly three-quarters of the system is able to absorb refugees and perform world’s refugees (World Bank 2011). And well depends in part on the manner in which that is certainly the case in the MENA region, refugees are incorporated into society, the where countries bordering on confl icts bear availability of additional resources from the the brunt of refugee crises (multiple times for international community, and the state’s Jordan). Refugees place particular strains on capacity to adjust to new demands. The a state’s ability to deliver services; they can Lebanese government allowed refugees but TRANSITIONS, CONFLIC TS, AND REFUGEES: BOTH AN OPPORTUNIT Y FOR AND A STRAIN ON THE CYCLE OF PERFORMANCE 267 did not place them in camps, a decision that children. One man described his trials in was politically driven. However, refugees enrolling his two children in school in were then scattered across the country, pri- Amman. He searched and searched, deter- marily in hard-to-reach areas, making it dif- mined to enroll them, and finally went to one ficult to provide health and other services principal day after day. “She threw me out because of the challenges of access and trans- the door, and I came back in through the portation (Naufal 2012). Even in Jordan, the window,” he explained, until at last she high flow of refugees made it difficult for the admitted his children.22 state and UNHCR to keep up with demand. The additional burden on the service deliv- Thus, for example, refugees report that wait- ery system affects not only refugees but ing for UNHCR registration, or having had it nationals as well. As the Migration Policy expire, is the main cause of being unable to Centre notes about Lebanon, “The ever access services in Jordanian health clinics increasing number of refugees weighs on the (CARE Jordan 2013). capabilities of host communities, which are Service delivery systems in host countries poor and lack the resources and adequate may become overwhelmed. In Jordan, a 2013 economic, educational and sanitary infra- CARE rapid community assessment of living structures. The host communities’ economic conditions for refugees found that hospitals situation is difficult and the refugees consti- and health care clinics refused to treat refu- tute a heavy burden” (Naufal 2012). Similar gees, often citing a lack of beds, medications, problems are found in Turkey. There, locals and other supplies. Over 50 percent of respon- complain that they are unable to get immedi- dents in the CARE survey reported that they ate treatment in hospitals because the hospi- had paid for private health care services since tals are overcrowded with Syrians. Although entering Jordan as refugees, in some cases reportedly no central order has been issued to more than JD 1,000 (CARE Jordan 2013). give Syrians priority, the wounded arriving (The majority of these were Syrians not regis- on a daily basis may crowd out Turkish citi- tered with the UNHCR.) Even refugees who zens (Özden 2013). are registered and can access public health care At times, the uneven cost of confl ict can find themselves in need of medications and also have a reverse effect. For example, as specialty care that they are unable to afford noted in chapter 2, Palestinian student refu- (IFRCJRC 2012). Moreover, in focus group gees served by the United Nations Relief and discussions Syrians living in Madaba noted Works Agency (UNRWA) schools score bet- that they were unable to access public clinics ter in international benchmarking tests in and hospitals in the city, and were instead math and science than their public school referred to Amman, which was costly in terms peers. UNRWA operates one of the largest of transportation (CARE Jordan 2013). nongovernmental school systems in the Similar issues are faced in education. The Middle East. It manages nearly 700 schools, 2013 CARE survey conducted in four large hires 17,000 staff, educates more than Jordanian cities (Madaba, Mafraq, Irbid, and 500,000 refugee students a year, and oper- Zarqa) found that in three of the four cities ates in four areas (Jordan, Lebanon, Syria, Syrians had more school-age children out of and the West Bank and Gaza). school than enrolled in it. This fi nding may UNRWA students in Jordan and the West stem to some extent from families’ need for Bank and Gaza on average achieve scores child labor, particularly in female-headed that are 23–80 points higher (equivalent to households. However, even in Irbid, where about a year’s worth of learning) than public 73 percent of children were not enrolled, only school students, even after controlling for stu- 10 percent of those out of school reported dent characteristics and for urban or rural working (CARE Jordan 2013). Another contexts (Patrinos et al. 2014). In part, this explanation may be found in the difficulty may be because Palestinian refugee students that some Syrians report in enrolling their are very aware of the adversities they face and 268 TRUST, VOICE, AND INCENTIVES view education as a path toward maintaining responses by citizens. For example, a 2013 hope for the future. UNRWA schools offer assessment of the impacts of the Syrian refu- relevant and equitable learning opportunities gees in Lebanon cited particular problems of by ensuring that they recruit and maintain a solid waste management, water, and electric- strong teaching force that is able to manage ity, which has led to a deterioration in living the challenges their students face, by provid- conditions for both the refugee and host ing students with academic guidance and communities (World Bank 2013). When refu- socioeconomic support, and by promoting a gees are located near the border of a conflict, high level of school and community engage- they can serve as the base of rebel social net- ment (Patrinos et al. 2014). Together, with a works, facilitating the spread of weapons and great sense of community and commitment violence, exacerbating bilateral tensions that to a common goal, schools are able to provide can in turn increase tensions with the host a relevant education despite the challenges. communities (Puerto Gomez and Christensen 2010). Finally, although refugee crises do not Trust, engagement, and institutions entail the same degree of institutional change The impact of refugee crises on trust, engage- experienced during transitions, they never- ment, and institutions is also less than clear. theless require the creation or adaptation of The ability of the state to provide high levels rules and procedures governing access to of services to citizens affects trust in public education, health, and other services. In the institutions, and where refugees disrupt tenu- short run, such institutional changes may be ous social balances, as they do in Lebanon, minimal, particularly when refugees are the state’s strategies toward what are seen as granted access to services under the auspices more or less favorable social groups affect of UNHCR. However, in the long run, states trust. As for engagement, refugees bring new are often forced to implement greater change. actors into the system who, as just described, In some cases, this can lead to the creation of may have unique constraints and conditions parallel systems, while in others it may lead that shape their demands on the state. to more systemic changes and investments. In Moreover, the presence of refugees can Turkey, for example, the government passed spark citizen mobilization along ethnic, sec- a comprehensive migration law that estab- tarian, or class lines, strengthening nonstate lished a legal system to protect and aid actors and institutions. Increased engagement refugees. Such measures are particularly may be particularly likely when animosities important because 65 percent of the Syrian have traditionally existed between refugee refugees in Turkey live outside of the camps. ethnic groups and those of the host country And yet these steps are also difficult, espe- (Puerto Gomez and Christensen 2010). cially when the host communities resent the Another catalyst for engagement is when ref- stresses of in-migration. For example, as ugees alter the balance between rival ethnic Ibrahim Saif, Jordan’s minister of planning or sectarian groups in the host country, such and international cooperation, explained, at as in Lebanon with the arrival of Palestinian the same time that the Jordanian people are refugees in the 1960s and Syrian refugees becoming tired of the strains of hosting Syrian today. Economic inequalities may also spark refugees, it is becoming ever more apparent engagement because the refugees may that the refugees will not be returning home increase the stresses on the host economies, any time soon. The government thus recog- particularly when resources flow to refugee nizes the need for institutional reforms that communities alone. address long-term problems, and yet it is Other factors also may influence the difficult to implement them in the current extent to which refugees affect engagement. climate. 23 This form of compassion fatigue They may increase environmental stresses, means that the impact of refugees on the cycle particularly on land and water, prompting of performance is likely to change over time. TRANSITIONS, CONFLIC TS, AND REFUGEES: BOTH AN OPPORTUNIT Y FOR AND A STRAIN ON THE CYCLE OF PERFORMANCE 269 Conclusion grave threats of further deterioration tend to induce people to take particularly bold coun- Today, the MENA region is inundated termeasures. This is in contrast to situations by major crises—regime change, wars, and in which people can choose between gains; refugees—spreading from North Africa to such a choice makes them exercise great, and the Arabian Peninsula. The results of these often excessive, caution. According to the ruptures remain to be seen, and no easy fore- research, the intersection of two conditions— casts can be drawn. However, they do offer the assumption of power by a new leader and the eruption of severe problems that put this opportunities for changing the cycle of per- leader in the domain of losses—is crucial for formance, and even open up the possibility the initiation of drastic adjustments. Because that costly events today may prompt changes such problems often trigger elite renovation, in institutions, trust, and engagement that the crisis appears to be the main factor in the result in better performance tomorrow. adoption of drastic reforms. During transitions, elites often experience a 2. On the difficulties faced in Tunisia’s transi- brief honeymoon period with much of the tion, see Bertelsmann Foundation (2014), population, during which they can gain citi- Freedom House (2012, 2014), and World zens’ acquiescence to reform. Such opportu- Bank (2014). nities may be short-lived, and the extent to 3. Interviews with Egyptian activists and observ- which they can be exploited is likely to ers, Cairo, August 2014. More generally, see Muggah (2012). depend on factors such as the strength of 4. For Tunisia, a report from the World Travel state institutions, the degree of polarization and Tourism Council (2014a) claims the within society, and the levels of regional or direct contributions were 7.3 percent, to international intervention. However, if citi- total 15.2 percent of GDP. For Egypt, the zens see improvements they may remain direct contributions to GDP from tourism engaged and continue to support reform. were 5.6 percent, and the total impact on It is therefore critical to seize the opportu- GDP (direct and indirect contributions) was nities offered in crises to buoy service provi- 12.6 percent (World Travel and Tourism sion, press for institutional reforms, and Council 2014b). maintain citizens’ trust and positive engage- 5. On Egypt, see Rissmann-Joyce (2014). ment. Even in the midst of the enormous dif- On Egypt, Libya, and Tunisia, see John D. Gerhart Center for Philanthropy and Civic ficulties faced by citizens and their states, Engagement (2013). there is an opportunity to escape the cycle of 6. In addition, on the impact of China’s Cultural poor performance. Preparing to face these Revolution, see Islam, Raschky, and Smyth challenges and seeking ways to open new (2011); on the negative effects of conflict on opportunities for breaking the cycle of poor pregnant women’s health and the long-term performance require a clearer understanding health prospects of their children, see Kaitz of how the international community, local et al. (2009) and Zapata et al. (1992). policy makers, civil society, and citizens can 7. By December 2013, 68 percent of Libyans work together to build citizen trust and felt the General National Congress was engagement, and to motivate public servants performing poorly (Transitional Governance and service providers toward providing better Project). On Egypt and Tunisia, see Benstead et al. (2013). services. 8. In September 2013, 41 percent of Libyans believed they were worse off than before 2011, and 27 percent felt they were the same. Notes In 2014, 54 percent of Tunisians believed 1. Weyland (2004) explains how situational they were worse off than before 2011, and conditions affect individual choices, and how 24 percent said they were about the same. grave, acute problems—crises—trigger bold 9. From a 2014 Transitional Governance Project reforms. People who face the danger of losses survey. This figure includes 34 percent who prefer risky choices. Severe problems that pose cited fighting unemployment; 13 percent, 270 TRUST, VOICE, AND INCENTIVES restoring economic growth; 11 percent, fight- Works Agency for Palestine Refugees in the ing poverty and economic inequality; and Near East (Maigua 2014). 6 percent, fighting rising prices. 20. On Cambodia, see Akresh and de Walque 10. Four polls were conducted under the (2008) and de Walque (2006); on Guatemala, Transitional Governance Project in August, see Chamarbagwala and Morá (2011). September, October, and November 2011. The 21. For more details, see the UNHCR Syrian polls found that the percentage of Egyptians Refugee Regional Response Information- rating an economic concern as the top issue Sharing Portal, http://data.unhcr.org facing Egypt ranged from 53 percent (in /syrianrefugees/country.php?id=107. August) to 49 percent (in November.) Security 22. Interview with Syrian baker from Damascus was the second most frequently cited concern. area living in Amman, Jordan, by Ellen Lust, 11. Surveys from May, September, and December March 28, 2014. 2013 conducted under the Transitional 23. Skype interview with Ibrahim Saif, min- Governance Project. ister of planning and international coop- 12. The extent to which transitions foster mobili- eration, Amman, Jordan, and New Haven, zation, the duration and nature of this mobili- Connecticut, February 4, 2014. On the ten- zation, and its political repercussions depend sions between Syrian refugees and Jordanians, on the type of authoritarian regime, the level see, for example, REACH (2014). of prior organization by labor unions and other organizations, and the political role of unions and professional organizations— References see Valenzuela (1988). For examples of the Abdalla, N. 2012. “Egypt’s Workers—From upsurge in labor strikes during the transi- Protest Movement to Organized Labor.” SWP tions in Brazil, Peru, and the Philippines, see Comments 32. http://www.swp-berlin.org Haggard and Kaufman (1995, 62), and on / filead m in /contents /products /com ments Brazil see Nervo Codato (2006, 15). /2012C32_abn.pdf. 13. There were two rounds of presidential Akresh, R., and D. de Walque. 2008. “Armed elections. The turnout reported is the average Conflict and Schooling: Evidence from the turnout over the two rounds of voting. 1994 R. Genocide.” Working Paper No. 3516, 14. See http://www.albawsala.com for more Institute for the Study of Labor. http://ftp.iza information. .org/dp3516.pdf. 15. Interview with Egyptian activist in the revolu- Alwazir, A. Z. 2012. “Garbage Collectors and the tionary youth movement, Cairo, August 2014. Struggle for Workers’ Rights in Yemen.” 16. For a discussion of the impact of war Jadaliyya, June 1. http://www.jadaliyya.com and, more generally, of democratization, see /pages/index/5788/garbage-collectors-and-the Bermeo (2010). -struggle-for-workers-ri. 17. On these issues in Rwanda, see Akresh Balcerowicz, L. 1997. “The Interplay between and de Walque (2008), and in Guatemala, Economic and Political Transition.” In Lessons see Chamarbagwala and Morá (2011). from the Economic Transition: Central and 18. The World Bank’s World Development Eastern Europe in the 1990s , edited by Report 2011 reported that it is difficult to Salvatore Zecchini, 153 – 67. New York: return to prewar investment levels, even after Springer Science + Business Media. conflict ends (World Bank 2011). For more Balcerowicz, L., and A. Gelb. 1995. “Macropolitics on the impact on investment in Uganda, see in Transition to a Market Economy: A Three- Deininger (2003). Also see Collier (1999) and Year Perspective.” Proceedings of the World Stewart and Humphreys (1997). Bank Annual Conference on Development 19. Estimates vary slightly. Another source states Economics. Washington, DC: World Bank. that at least 280 schools and kindergartens Batniji, R., L. Khatib, M. Cammett, J. Sweet, were damaged or destroyed (Jalbout, Dryden- S. Basu, A. Jamal, P. Wise, and R. Giacaman. Peterson, and Watkins 2014), whereas a third 2014. “Governance and Health in the Arab source found that 25 government schools World.” The Lancet 383 (9914): 343–55. were completely destroyed and at least 207 Bellows, J., and E. Miguel. 2006. “War and others were damaged, including 75 run Institutions.” American Economic Review 96 by UNRWA, the United Nations Relief and (2): 394–99. TRANSITIONS, CONFLIC TS, AND REFUGEES: BOTH AN OPPORTUNIT Y FOR AND A STRAIN ON THE CYCLE OF PERFORMANCE 271 Benstead, L., E. Lust, D. Malouche, G. Soltan, January 24. http://www.ei-ie.org/en/news and J. Wichmann. 2013. “Islamists Aren’t /news_details/1639. the Obstacle: How to Build Democracy in ———. 2012. “Tunisia: Strikes to Improve Egypt and Tunisia.” ForeignAffairs.com, Education System and Teachers’ Status.” February 25. http://www.foreignaffairs.com Education International News, May 23. /articles/138940/lindsay-benstead-ellen-m http://www.ei-ie.org/en/news/news_details -lust-dhafer-malouche-gamal-soltan-and-j /2170. / islamists-arent-the-obstacle. Elmahdy, M. 2012. “Revolution and Mental Bermeo, N. 2010. “A. Conflict and the Durability Health.” Egyptian Journal of Psychiatry 33 of Electoral Democracy.” In In War’s Wake: (3): 1110–15. International Conflict and the Fate of Liberal Faiola, A. 2011. “Egypt’s Labor Movement Democracy, edited by Elizabeth Kier and Blooms in Arab Spring.” Washington Post, Ro n a ld R . K r e b s . C a m b r id g e , U . K . : September 25. http://www.washingtonpost Cambridge University Press. . c o m / w o r l d / m i d d l e - e a s t /e g y p t s - l a b o r ———. 2014. “Conflict and Consolidation.” -movement-blooms-in-arab-spring/2011/09/25 Paper presented at the annual meeting of the /gIQAj6AfwK_story.html. American Political Science Association, Fre edom House. 2012 . “C ou nt r ie s at Washington, DC, August 28–31. t he C rossroads: Tu n isia.” ht t p: //w w w Bertelsmann Foundation. 2014. “Bertelsmann .f reedom house.org /repor t /cou nt ries Transformation Index 2014. Tunisia Country -crossroads/2012/tunisia#.VD0b-74qqZY. Report.” http://www.bti-project.org/reports ———. 2014. “Freedom in the World Report: /country-reports/mena/tun#democracy. Tunisia.” http://w w w.freedomhouse.org B l o m b e r g , S . B r o c k , G . D. H e s s , a n d / repor t /f reedom-world / 2014/t u n isia- 0 # A. Orphanides. 2004. “The Macroeconomic .VD0fB74qqZY. Consequences of Terrorism.” Journal of Gall, C. 2014. “Libyan Refugees Stream to Tunisia Monetary Economics 51 (5): 1007–32. for Care, and Tell of a Home That Is Torn CARE Jordan. 2013. “Syrian Refugees in Urban Apart.” New York Times, September 9. http:// Jordan: Baseline Assessment of Community- www.nytimes.com/2014/09/10/world/africa Identified Vulnerabilities among Syrian /libya-refugees-tunisia-tripoli.html?_r=1. Refugees Living in Irbid, Madaba, Mufraq, García-Ponce, O., and B. Pasquale. 2014. “How and Zarqa.” Amman: CARE Jordan. http:// Political Repression Shapes Attitudes toward www.care.org/sites/default/files/documents the State.” Unpublished manuscript. http:// / EMER-JOR-2013-Syrian-Refugees-in-Urban omargarciaponce.com /wp-content -Jordan.pdf. / upload s / 2013/ 07/G a rc ia Ponc e _ a nd Chamarbagwala, R., and H. E. Morá. 2011. “The _ Pasquale_2014.pdf. Human Capital Consequences of Civil War: Gates, S., H. Hegre, and H. M. Nygård. 2012. Evidence from Guatemala.” Journ al of “Development Consequences of A rmed Development Economics 94 (1): 41–61. Conf lict.” World De velopment 40 (9): Collier, P. 1999. “On the Economic Consequences 1713–22. of Civil War.” Oxford Economic Papers 51 Ghobarah, H. Adam, P. Huth, and B. Russett. (1): 168–83. 2004. “The Post-war Public Health Effects of Deininger, K. 2003. “Causes and Consequences Civil Confl ict.” Social Science and Medicine of Civil Strife: Micro-level Evidence from 59 (4): 860–994. Uganda.” Unpublished manuscript, World G i ac a m a n , R . , R . K h at ib, L . S h aba neh , Bank, Washing ton, DC . ht tp: //elibrar y A . R a m law i, B . Sabri, G . Sabati nelli, .wo rl d b a n k . o r g /d o i / p d f / 10 .159 6 M. Khawaja, and T. Laurance. 2009. “Health /1813-9450-3045. Status and Health Services in the Occupied de Walque, D. 2006. “The Socio-Demographic Palestinian Territory.” The Lancet 373 (9666): L egacy of the K h mer Rouge Period in 837–49. Cambodia.” Population Studies 60 (2): Gupta, S., B. Clements, R. Bhattacharya, and 223–31. S. Chakravarti. 2004. “F. Consequences of Education International. 2011. “Tunisia Faces A. Conflict and Terrorism in Low- and Middle- Te acher St r i ke , Prote st s aga i n st N ew Income Countries.” European Journal of Government.” Education International News, Political Economy 20: 403–21. 272 TRUST, VOICE, AND INCENTIVES Haggard, S., and R. R. Kaufman. 1995. The Policy-Making.” Comparative Political Studies Political Economy of Democratic Transitions. 25 (4): 433–86. Princeton, NJ: Princeton University Press. Khatib, L. 2013. “Corruption in Qatar? The Link Hubbard, B., and D. D. Kirkpatrick 2013. between the Governance Regime and Anti- “Sudden Improvements in Egypt Suggest a Corruption Indicators.” Working Paper No. Campaign to Undermine Morsi.” New York 40, European Centre for Anti-Corruption and Times , July 10. http://www.nytimes.com State-Building. http://www.againstcorruption /2013/07/11/world/middleeast/improvements .eu /wp-content /uploads/2013/12 / W P- 40 -in-egypt-suggest-a-campaign-that-undermined -Qatar-paper.pdf. -morsi.html?pagewanted=all. Khatib, L., and E. Lust, eds. 2014. Taking to the Hunter, J. 2006. “Institutional Change in Meiji Streets: The Transformation of Arab Activism. Japan: Image and Reality.” In Institutional Baltimore, MD: Johns Hopkins University Change in Japan, edited by Magnus Blomström Press. and Summer La Croix, 45–70. New York: Levitsky, S. R., and L. A. Way. 2012. “Beyond Routledge. Patronage: Violent Struggle, Ruling Party IFRCJRC (International Federation of Red Cross Cohesion, and Authoritarian Durability.” and Red Crescent Societies). 2012. “Syrian Perspectives on Politics 10 (4): 869–89. Refugees Living in the Community in Jordan: Lischer, S. K. 2009. “Causes and Consequences of Assessment Report.” Geneva. Conflict-Induced Displacement.” Civil Wars 9 Institute for Integrated Transitions. 2013. Inside (2): 142–55. the Transition Bubble: International Expert Looney, R. 2006. “Economic Consequences of A ssi st anc e in Tuni si a . ht t p: //w w w.ifit Con f lic t: T he R ise of I raq’s I n formal -t ransitions.org /publications /inside-the Economy.” Journal of Economic Issues 40 (4): -t ransition-bubble-international- ex per t 991–1007. -assistance-in-tunisia/inside-the-transition Maigua, P. 2014. “It’s Back to School for -bubble-en-full. Palestinian Children in Gaza amid Death and Islam, A., P. A. Raschky, and R. Smyth. 2011. Destruction.” http://www.unmultimedia.org “The Long-Term Health Effects of Mass /radio/english/2014/09/its-back-to-school-for Political Violence: Evidence from China’s -palestinian-children-in-gaza-amid-death-and Cultural Revolution.” Discussion Paper 32/11, -destruction/?utm_source=UN+Radio&utm Department of Economics, Monash University, _ medium=facebook#.VBc9kktmpfM. Melbourne, Australia. http://www.buseco Muggah, R. 2012. Researching the Urban .monash.edu.au/eco/research/papers/2011/321 D il e m m a : Urba n iz atio n , Po ve r t y a n d 1longtermhealthislamraschkysmyth.pdf. Violence. Ottawa: International Development Jalbout, M., S. Dryden-Peterson, and K. Watkins. Research Centre. http://www.idrc.ca/ EN 2014. “The Destruction of Gaza’s Schools and / PublishingImages/Researching-the-Urban the Future of Palestinian Children.” http:// -Dilemma-Baseline-study.pdf. www.brookings.edu/blogs /education-plus Naufal, H. 2012. “Syrian Refugees in Lebanon: -development/posts/2014 /08/04-gaza-school The Humanitarian Approach under Political - p a l e s t i n i a n - c h i l d r e n Div i sion .” Re se a rch Repor t 2012 /13, -jalbout-dryden-peterson-watkins. Migration Policy Centre, Florence. http:// John D. Gerhart Center for Philanthropy and www.migrationpolicycentre.eu/docs/MPC Civic Engagement. 2013. Giving in Transitions %202012%2013.pdf. and Transitions in Giving: Philanthropy in Nervo C. A. 2006. “Political Transition and Egypt, Libya and Tunisia 2011–2013. http:// Democratic Consolidation in Brazil: A www.aucegypt.edu/research/gerhart/rprogram Historical Perspective.” In Political Transition /Documents/PSE.pdf. and Democratic Consolidation: Studies on Kaitz, M., M. Levy, R. Ebstein, S. V. Faraone, and Contemporary Brazil, edited by Adriano D. Mankuta. 2009. “The Intergenerational Nervo Codato, 1–26. New York: Nova Effects of Trauma from Terror: A Real Science. Possibility.” Infant Mental Health Journal 30 Özden, S. 2013. “Syrian Refugees in Turkey.” (2): 158–79. Research Report 2013/05, Migration Policy Keeler, J. T. S. 1993. “Opening the Window for Centre, F. http://www.migrationpolicycentre Reform: Mandates, Crises, and Extraordinary .eu/docs/MPC-RR-2013-05.pdf. TRANSITIONS, CONFLIC TS, AND REFUGEES: BOTH AN OPPORTUNIT Y FOR AND A STRAIN ON THE CYCLE OF PERFORMANCE 273 Patrinos, H., H. Abdul-Hamid, J. Reyes, and Skocpol, T. 1979. States and Social Revolutions. J. Kelcey. 2014. “Learning in the Face of Cambridge, U.K.: Cambridge University Press. Adversity: The UNRWA Education Program Stewart, F., C. Huang, and M. Want. 2000. for Palestinian Refugees.” World Bank, “Internal Wars in Developing Countries: An Washington, DC. Empirical Overview of Economic and Social Puerto G., M., and A. Christensen. 2010. “The C o n s e qu e n c e s .” I n Wa r a n d I mpac t s of Ref uge e s on N eig hb or i ng Underdevelopment, Vol. 1, The Economic and Countries: A Development Challenge.” World Social Consequences of Conflict, edited by Development Report 2011 Background Note, Frances Stewart and Valpy Fitzgerald, 67–103. World Ban k, Wash ing ton, DC . ht t p: // Oxford, U.K.: Oxford University Press. siteresources.worldbank.org / EXTWDR2011 Stewart, F., and F. P. Humphreys. 1997. “Civil / Resources/6406082-1283882418764/WDR Conflict in Developing Countries over the Last _ Background_Paper_ Refugees.pdf. Quarter of a Century: An Empirical Overview REACH. 2014. “Livelihoods, Employment and of Economic and Social Consequences.” Tensions in Jordanian Communities Hosting Oxford Development Studies 25 (1): 11–42. Syrian Refugees.” British Embassy, Amman, The Guardian. 2014. “Gaza Children Return Jord a n. Fi le: /// Users /el 242 / Dow n loads to School after End of War with Israel.” /1407917949.pdf. September 14. http://www.theguardian.com Revkin, M. Forthcoming. “Triadic Legal Pluralism / world /2014/sep/14/gaza-children-school in North Sinai: A Case Study.” UCLA Journal -war-palestine-israel. of Islamic and Near Eastern Law 13. UNHCR (United Nations High Commissioner for Rissmann-J., Stacie. 2014. “Egyptian Education in Refugees). 2014. “2014 UNHCR Regional Times of Social and Political Unrest: Crisis or Operations Profile—North Africa.” http:// S erendipit y.” Inte r n ation al Jour n al of www.unhcr.org/pages/49e485f36.html#. Humanities and Social Science 3 (6): 302–07. UNRWA (United Nations Relief and Works Rollins, T. 2014. “Egypt’s New Prime Minister Agency). 2014. “Food Insecurity in Palestine Faces Labor Crisis, Strikes.” Al Monitor, Remains H igh.” ht tp: //w w w.unr wa.org February 26. http://www.al-monitor.com / newsroom /press-releases/food-insecurity /pulse/originals/2014/02/egypt-strikes-mahalla -palestine-remains-high. -workers-mehleb.html#. U N W TO (United Nations World Tourism Saleh, H. 2014. “Strikes and Political Divisions Organization). 2011. “UNWTO Supporting Plague Tunisia Phosphate Industry.” Financial Tourism in Tunisia.” http://www2.unwto.org Times, February 19. http://www.ft.com/intl/cms /press-release/2011-06-06/unwto-supporting /s/0/a890651a-931a-11e3-8ea7-00144feab7de -tourism-tunisia. .html#axzz3DKxZnp8d. Valenzuela, J. S. 1988. “Labor Movements in Saleh, S. S., M. Alameddine, N. M. Natafgi, Transition to Democracy: A Framework for A. Mataria, J. Nasher, M. Zeiton, S. Ahmad, Analysis.” Working Paper No. 104, Kellogg and S. Siddiqi. 2014. “The Path Towards Institute for International Studies, Notre Universal Health Coverage in the Arab Dame, I N, June. https:// kellogg.nd.edu Uprising Countries Tunisia, Egypt, Libya and /publications/workingpapers/WPS/104.pdf. Yemen.” The Lancet 383 (9914): 368–81. Watkins, A. 2014. “The Perils and Possibilities of Sausa, C. A. 2013. “Political Violence, Collective Engaging Non-State Armed Actors in Post- Functioning and Health: A Review of the Confl ict Peacebuilding.” Small Wars Journal, Literature.” Medicine, Conflict, and Survival March 7. http://smallwarsjournal.com/jrnl/art 29 (3): 169–97. /the-perils-and-possibilities-of-engaging-non Schneckener, U. 2006. “Fragile Statehood, Armed -state-armed-actors-in-post Non-State Actors and Security Governance.” -conflict-peacebuild. In Private Actors and Security Governance, Weyland, K. 2004. The Politics of Market Reform edited by Alan Bryden and Marina Caparini, in Fragile Democracies: Argentina, Brazil, 23–40. Geneva: Democratic Control of Armed Peru, and Venezuela. Princeton, NJ: Princeton Forces (DCAF). University Press. Sen, K. 2013. “Syria: Effects of Conflict and World Bank. 2011. World Development Report Sanctions on Public Health.” Journal of Public 2011: Confl ict, Security, and Development. Health 35 (2): 195–99. Wash i ng ton, DC: World B a n k. ht t p: // 274 TRUST, VOICE, AND INCENTIVES siteresources.worldbank.org / I N T W DRS World Travel and Tourism Council. 2014a. Travel / Resources/WDR2011_Full_Text.pdf. and Tourism: Economic Impact 2014: Tunisia. ———. 2012 . Joint Economic and Social London: World Travel and Tourism Council. Assessment for the Republic of Yemen. ———. 2014b. Travel and Tourism: Economic Washington, DC: World Bank. Impact 2014: Egypt. London: World Travel ———. 2013. Lebanon: Economic and Social and Tourism Council. Impact Assessment of the Syrian Confl ict. Z ap a t a , C . , A . R e b o l l e d o , E . A t a l a h , E xecutive S umm ar y . Washing ton, DC: B. Newman, and M.-C. King. 1992. “The World Bank. http://w w w.worldbank.org Influence of Social and Political Violence on /content/dam/ Worldbank/document/MNA the R isk of Pregnancy Complications.” / L B N - E S I A% 2 0 o f % 2 0 S y r i a n% 2 0 American Journal of Public Health 82 (5): C o n f l i c t -%2 0 E X%2 0 S U M M A RY%2 0 685–90. ENGLISH.pdf. ———. 2014. The Unfinished Revolution: Bringing Opportunity, Good Jobs and Greater Data sources Wealth to All Tunisians. Washington, DC: Gallup World Poll, http://www.gallup.com World Bank. http://www-wds.worldbank.org /services/170945/world-poll.aspx /external/default/WDSContentServer/WDSP International IDEA, http://www.idea.int/vt / IB/2014/09/16/000456286_20140916144712 /index.cfm /Rendered/PDF/861790DPR0P12800Box3853 Tra nsit iona l G over na nce Projec t , ht t p: // 14B00PUBLIC0.pdf. transitionalgovernanceproject.org/ Donor Support 11 • Donors have been only partly effective in supporting education and health services delivery in countries in the Middle East and North Africa (MENA). • World Bank nonlending activities have increasingly sought to address the weaknesses in institutions and citizen engagement that undermine service delivery performance; lending operations have done so to a lesser degree. • The usual practice of identifying policy reforms must be matched by a corresponding focus on how any given policy reform will actually be implemented and by whom. • Social accountability is a promising approach for the MENA countries, even if its implementation requires time, money, expertise, and fit in local contexts. • Evaluation of donor activities is essential. Evidence is lacking, however, on the effectiveness of the responses to such evaluations of Bank activities. T he quality of education and health in provinces to individual villages and specific the Middle East and North Africa schools and clinics—perform so much better (MENA) region remains low despite than others, despite facing similar challenges. higher spending and policy reforms. As we In this report, we argue that such variation have seen, school enrollment rates and access can be a major source of ideas and inspira- to basic health care facilities have steadily tion for policy makers seeking ways to improved, but students perform below improve the quality of service delivery. In the average in the classroom, and health care same way, it behooves donors, including the coverage and its use are highly inequitable. World Bank, to seek ways to enhance their However, these low averages are accompa- own effectiveness by examining and learning nied by considerable variation between and from variation in the quality of their engage- (especially) within countries, and much can ment with clients. Here, we examine the be learned from examining how it is that Bank’s own record of operational and non- some settings—from governorates and lending work (analytical, advisory, and 275 276 TRUST, VOICE, AND INCENTIVES technical assistance activities) in the MENA considered at the project design stage and region and how this assistance has changed which governance drivers were promoted over time. This examination will serve as a throughout implementation. Project perfor- basis for identifying those areas in which mance is analyzed by looking at the main strengthening may be required, where mutu- components of assessments of project ally beneficial partnerships have been the success and failure. The chapter concludes most fruitful, and where new opportunities by considering the outlook for the future of may be found. project design and implementation. Quality of service delivery has always been a crucial area of attention for the World Bank and other donors in the MENA An overview of the World Bank’s region. In looking at the Bank’s role in sup- education and health activities porting education and health services deliv- in the MENA region ery in the region over the last 25 years, we The Bank’s involvement in the MENA consider the following questions: How have region, as elsewhere in the world, takes vari- projects evolved? On what bases have defi- ous forms, but its efforts roughly fall into ciencies in service delivery been addressed? two categories: operational and nonlending Which governance mechanisms were taken (which mainly includes analytical, advisory, into consideration, and which were ignored? and technical assistance activities). The We start by looking at how World Bank volume of operational activity across the projects in the MENA region have been MENA region from 1994 to 2014 is summa- designed through the years: what were the rized in figure 11.1. challenges in the MENA countries, and how The primary purpose of the World Bank’s did the Bank respond? The Bank’s role is operational engagement in education and explored from a governance perspective, health issues in the MENA countries has asking which governance mechanisms were changed considerably over time, beginning FIGURE 11.1 Education and health lending, World Bank: MENA region, 1994–2014 450 400 350 300 US$, millions 250 200 150 100 50 0 sia . p. q an co on p. ti za ep Ira ou Re Re Ga oc rd n ni ,R ba ib Tu Jo or b. ic nd en Dj Le am ra M m ka t, A Isl Ye an yp n, tB Ira Eg es W Source: World Bank. Note: MENA = Middle East and North Africa. DONOR SUPPORT 277 as early as 1970. In health, for example, one comprehensive institutional strengthening can discern six phases: population lending, within the education and social protection 1970–79; primary health care, 1980–86; sectors. health reform, 1987–96; health outcomes Because of the existing weaknesses in and health systems, 1997–2000; global tar- institutions and accountability mechanisms gets and partnerships, 2001–06; and system (chapter 5) and in formal citizen engagement strengthening for results, 2007–present (see (chapter 9), an even greater emphasis on IEG 2009b).1 The Bank’s fi rst involvement governance and accountability in addressing in the health sector focused on improving service delivery is needed. The regional access to family planning services as a MENA Health, Nutrition, and Population response to the concern about the adverse Strategy, entitled Fairness and Accountability: effects of rapid population growth on eco- Engaging in Health Systems in the Middle nomic growth and poverty reduction. Direct East and North Africa (World Bank 2013b), lending in the health sector formally began calls for fair and accountable health systems, with the 1980 “Health Sector Policy Paper” thereby addressing the transformative socio- (World Bank 1980). This effort enabled the political changes that have shaken up the Bank to better address the health needs of MENA region since December 2010. It the poor by enhancing access to low-cost addresses systemic disparities as well as a fair primary health care. The first loan to distribution of costs, dignified provider- expand basic health services was made in patient interactions, and the obligation of 1981 to a MENA country, Tunisia. Two client countries to ensure that health care ser- new objectives were added following the vices are timely, effective, safe, appropriate, release of Financing Health Services in cost-conscious, and patient-centered. Developing Countries in 1987: making Meanwhile, increasingly Bank education health fi nancing more equitable and efficient and health operations have included compo- and reforming health systems to overcome nents that address accountability, transpar- systemic constraints (World Bank 1987). ency, and citizen engagement. The proportion Service delivery was linked to public sector of education projects seeking to improve efficiency and thus was subjected to the external and internal accountability and structural reforms of the 1980s and 1990s. transparency rose to over one-third during A new strategy paper in 1997 shifted the the last decade (figure 11.2). The increase has focus from access to health outcomes, much been less pronounced in health projects, like the shift in focus in the education sector although projects designed since the adoption (World Bank 1997). Nevertheless, improved of the new regional health strategy in 2013 health system performance in terms of seek greater emphasis of these governance equity, affordability, efficiency, quality, elements. and responsiveness to clients continued The challenges facing the World Bank in to be a priority, with the addition of education have differed throughout the increased financing of single-disease or region—from increasing overall access to single-intervention programs that are often education to supporting higher education launched within weak health systems. and, more recently, preprimary education and The emphasis on performance of the edu- the education-to-job transition. In line with cation and health systems continues today. the Bank’s 1995 “Education Sector Strategy The latest Bank health strategy continues to Paper,” basic education was supported as the adhere closely to the objectives of the 1997 highest priority in lending to countries with- strategy, but introduces a stronger focus on out universal literacy and with low enroll- governance and the importance of demon- ment rates (World Bank 1995). Primary strating results. Similarly, Bank strategies on education was also nurtured because the education and social protection promote sys- social rates of return were found to be highest tem development, including approaches to in most developing countries. In the MENA 278 TRUST, VOICE, AND INCENTIVES FIGURE 11.2 The changing role of voice, accountability, and countries such as Morocco in the 1990s, transparency in World Bank education and health projects: where primary enrollment was about MENA region, 1994–2013 70 percent at that time. Countries with improved outcomes such as Tunisia and 50 Jordan worked with the World Bank Percent of projects in the given sector 45 to enhance higher education and the 40 education-to-job transition (“knowledge 35 economy”). Gaps in urban and rural educa- 30 tion as well as gender equality were recurring 25 themes and continually considered in 20 project design. 15 Analytical and advisory activities have 10 undergone a remarkable expansion since 5 2001. Their monetary value peaked in fiscal 0 2013 (figure 11.3). 1994–2004 2005–13 1994–2004 2005–13 1994–2004 2005–13 Although the World Bank’s education Voice/external Internal Transparency lending portfolio is almost three times the accountability accountability size of its health lending portfolio, the health Education projects Health projects sector has dominated analytical work in the M ENA region over the last 12 years. Source: World Bank. Note: MENA = Middle East and North Africa. Especially in fiscal 2008 and 2009, analytical work in health was approved for loans at least t wice as much as in education. FIGURE 11.3 Analytical and advisory activity commitments, When differentiating between analytical education and health, World Bank: MENA region, FY01–13 work and technical assistance, it becomes apparent that health activities in the MENA 5 region have been especially strong in terms of technical assistance, particularly between fis- 4 cal 2008 and 2014. Analytical work has been less constant in the health sector; it has US$, millions 3 mostly consisted of Health Sector Reviews, disease-specific analyses (such as HIV/AIDS 2 and schistosomiasis), Health Policy Notes, and Regional Health Studies. Specific gover- 1 nance issues addressed include governance of the pharmaceutical sector (Jordan), dual 0 practice reform in medical practice (West Bank and Gaza), and assessment of hospital 01 02 03 04 05 06 07 08 09 10 11 12 13 FY FY FY FY FY FY FY FY FY FY FY FY FY performance (Tunisia). Generally, analysis in Source: World Bank. the education and health sectors has included Note: Actual total cumulative cost = bank budget + trust funds. The fiscal 2000–04 data on sector-specific analytical work, technical coordinated analytical activities offered only one option, “Prepared jointly with other donor(s).” Thus considerable caution should be exercised in any comparison with post–fiscal 2005 data, assistance, and the capacity-building pro- except if restricted solely to donor participation. All estimates are based on the authors’ grams of the World Bank Institute. Outside calculations. MENA = Middle East and North Africa. the education and health sectors per se, ana- lytical work can refer to education and health region, this was true of the Republic of services delivery in core diagnostics (public Yemen, where the primary enrollment rate expenditure reviews, poverty assessments) increased by 21.5 percent between 1999/2000 and multisector analytical work. and 2010/11 (U N E SCO I nstitute for When working toward improving service Statistics, education, 2012), but also of delivery in a client country, it is useful to look DONOR SUPPORT 279 at the client’s budget flows and interactions with numerous Public Expenditure Reviews. through the delivery chain in order to detect Assessments like these underline the need to inefficiencies or leakages—breaks in the treat service delivery as a cross-cutting issue chain. Analytical tools for conducting that must be analyzed and supported com- such core diagnostic work include the prehensively, with clear links to public sector World Bank’s Public Expenditure Tracking governance and public financial manage- Survey (PETS) and Quantitative Service ment. Finally, even though the World Bank Delivery Survey (QSDS), which aggregate has produced many comprehensive sector data on inputs, outputs, user charges, quality, analyses, the gaps in the data in this field are and other characteristics directly from the still large, especially because most of the service provider level so that more can studies conducted have focused on three be learned about “the linkages, leakage, and countries: Egypt, Morocco, and the Republic the way spending is transformed into ser- of Yemen. vices” (Dehn, Reinikka, and Svensson 2002). Several PETS and QSDS activities have been conducted in the MENA region for edu- Factors shaping variation in cation and health, and they form the basis for operational effectiveness much of our analysis of performance at the In placing the World Bank’s activities in the point of service delivery in chapters 6 and 7. MENA countries in context, it helps to have For example, in 2011 the PETS and QSDS a broader sense of the “aid landscape” in the surveys conducted in Morocco shed light on region, something that aidData enables us to the flow and use of resources in the public do.2 Of the more than 14,500 projects in the health sector. Some of the PETS and QSDS aidData database, 7 percent are in MENA tools have been funded by other donors such countries, and of the 12 donors (including the as the European Commission and used with World Bank) associated with these projects, other research methods to gain a comprehen- 8 work in the MENA region. How well does sive understanding of the health or education the World Bank perform in the MENA expenditure chain in a given MENA country. region compared with its performance in In the Republic of Yemen, for example, track- other regions? And in turn how does that ing education expenditures means analyzing performance stack up against the perfor- public resource management and teacher mance of all other donors in the MENA absenteeism rather than undertaking a region compared with their own performance “pure” PETS in the education sector. The in other regions? The preliminary answer is reason is that Yemeni schools receive few or that, on average, donors perform worse in the no cash resources, rendering a PETS MENA region than they do elsewhere. The approach, which emphasizes estimating fiscal World Bank, however, does slightly better leakages from cash resources allocated at the than average (though likely it is not statisti- school level, unsuitable. Thus the 2006 World cally significant). At the very least, the World Bank study of the education sector in the Bank does not seem to perform worse than Republic of Yemen used an absenteeism sur- average, unlike other donors.3 vey to examine the leakages in wage and sal- Even so, the World Bank’s involvement in ary expenditures. the MENA region has faced several chal- Other World Bank products of this kind lenges in recent years. The regional political include its Public Expenditure and Financial economy has played a distinct role, affecting Accountability Country Assessment, which the Bank’s engagement, in particular through determines how to improve public fi nancial the nature and type of Bank-client relation- management (PFM). Country Procurement ships, which have recently changed because Assessments have been carried out in the of the Arab Uprisings and transformations Arab Republic of Egypt, Iraq, Jordan, in the region. Before 2011, the Bank’s rela- Morocco, and the Republic of Yemen, along tionship with clients was often not all it 280 TRUST, VOICE, AND INCENTIVES appeared to be. During President Zine El Over time, it became apparent that quality Abidine Ben Ali’s regime in Tunisia, for was not necessarily the result of just delivering example, the country had the Bank’s biggest inputs, and yet the Bank struggled to move lending portfolio in education and health, beyond these tasks. In Jordan, for example, adding up to almost $395 million between where seven Bank-assisted education projects fiscal 1994 and 2014 (see figure 11.1). were initiated or fully delivered in the 1990s, Internationally, Tunisia was perceived as a the Operations Evaluation Department (OED) strong, diligent client, and it was praised for Country Assistance Evaluation reasoned that its comparably good economic and social “despite the generally satisfactory outcome for development. T he latest I ndependent each of the seven projects in terms of their Evaluation Group (IEG) Country Program objectives (dominated by construction) . . . the Evaluation for Tunisia, which analyzed the assistance strategy focusing on primary World Bank’s engagement from fiscal 2005 and secondary education, vocational training, to 2013, reveals a different picture, however, and lately higher education during the stating that the “relationship was broadly nineties, has not yet made a significant contri- characterized by tight government control bution to improvements in the quality of edu- and relative passivity on the part of the Bank, cation at any level” (Operations Evaluation in particular after 2007” (IEG forthcoming). Department 2004). The Tunisian government sometimes inter- In the years that followed, teacher train- fered in the Bank’s work; it would inhibit the ing combined with stronger parent-student dissemination of some analytical work, pre- involvement in schooling became a useful vent the pursuit of some key work (for exam- strategy for targeting enhanced education ple, Public Expenditure Reviews, Investment qualit y, frequently incorporating the Climate Assessments), and intervene in the concept of participation. For example, Bank’s interaction with stakeholders. Even measures that were recommended in the though the analysis that was actually pro- early 2000s to improve the quality of duced was of high quality, the Bank’s reputa- education in Egypt included involvement tion as a provider of independent analysis of the community through school councils, was affected by the Tunisian case. Examples enhancement of teacher capacity (as well like this demonstrate that the Bank’s engage- as an emphasis on student learning evalua- ment in the MENA region has often been tions), installation of new technology, and welcomed, but only to a certain extent, even competition in private sector engagement though many countries in the region have (World Bank 2002). performed quite well compared with others In addition to teacher training, the contri- in the developing world. bution that parents and students could play Beyond broad political economy consider- in raising school quality emerged in concert ations, internal assessments of Bank projects with accountability frameworks and closer have lamented the challenge of moving from attention to concepts of incentive structures. the provision of inputs (buildings, supplies, The 2008 MENA education flagship report staff training) to ensuring the quality of out- powerfully concluded that where “reforms in comes (learning, public health functions, cura- the past have tended to focus on the engineer- tive care). Concerns about the quality of ing of the education system (building schools, education have been a focus of World Bank printing textbooks, hiring teachers) . . . engagement since at least the early 1990s, but today’s reforms should rely more on the use at that time the main priority in the MENA of incentives and accountability measures” region was increasing the access to education (World Bank 2008). Moreover, to further for the burgeoning young populations. Thus strengthen accountability from a demand- country assistance in education was designed side perspective, social accountability tools comprehensively, and the objectives of projects incorporating the citizen’s voice began to be were dominated by construction components. introduced. DONOR SUPPORT 281 One of the most influential (if contested) with an emphasis on provider ability and the concepts of the last 20 years in development availability of key inputs. In fact, this was the has been “good governance.” Under this World Bank’s overall education policy at broad umbrella has emerged a related set of that time and thus was not applied just to the ideas—for example, voice, participation, MENA region. As the 2006 IEG evaluation accountability, transparency—which are of World Bank support to primary education believed to be essential to shaping incentive noted, “Only about one in five projects had structures that promote high-quality service an explicit objective to improve student learn- delivery (Ringold et al. 2012). Supported by a ing outcomes. This does not mean that proj- combination of hard-won experience, newly ects were unconcerned about quality: almost gained insights, and academic research, all aimed for improvements in educational these concepts now guide World Bank activi- quality, but until recently this was mostly ties, from analytical work to outputs on seen in terms of delivery of inputs and the ground. An example is the stronger services” (IEG 2006). A look at education emphasis on participation and the role of projects throughout the 2000s reveals that “communities” as both a means and a target little has changed until recently for the of development assistance (see World Bank MENA region. 2001). More diverse areas of participation More or less the same assessments can be have been incorporated since the mid-2000s, made of the health sector. In the late 1990s when social accountability and other related and early 2000s, improvements in the quality concepts and corresponding tools were intro- of care were sought by providing training duced. For the MENA region, the essential and medical equipment as well as rehabilitat- significance of social accountability became ing facilities. Providers’ efforts were targeted evident in 2010–11 because the revolutions in only indirectly through the implementation four countries were not only about the peo- of quality and staffi ng norms, and monitor- ple’s desire for change but, more important, ing mechanisms or incentives were seldom their desire to be heard and be out of the introduced into project design. The Jordan shadow of their former omnipresent leaders. Operations Evaluation Department Country The shift just described toward including Assistance Evaluation of 2004 asserted that “participation” in project design has been “while primary health care and health sector clearly visible in education, where commu- staffing have improved as a result of training nity participation was strongly encouraged as and construction fi nanced by these projects, of the early 2000s. Mothers’ and fathers’ the overall results in terms of long term pol- councils were introduced or strengthened in icy reform in the health sector have not been schools throughout the MENA region, and i mpre s sive” (O p erat ion s Eva lu at ion local councils or community committees Department 2004). became involved through the development of In education (and to a lesser extent in community-based action plans or similar ini- health), upgrading management information tiatives. Here, however, a distinction must be systems was frequently invoked as a basis made between passively addressing the com- for improving education quality, as were munity and actively involving the community other relevant governance drivers such as at the decision-making level, because many transparency and access to information. earlier projects had referred to involvement However, measures that were recommended when elaborating on the concept of commu- or implemented in education, such as provid- nity participation. ing information on teaching practices and Also in the early 2000s, student learning publicly available student learning outcomes, was becoming the main focal point in educa- were aimed at enhancing education manage- tion, although it now depended mainly on ment mostly at the administrative level and teacher training and its quality. Education not at the school level. This thesis is sup- projects thus continued to be input-oriented, ported by an upcoming study on World Bank 282 TRUST, VOICE, AND INCENTIVES support for school autonomy and account- sought throughout the mid-2000s by donors, ability for fiscal 2003–13 (World Bank including the World Bank. As the 2006 IEG forthcoming). It finds that the World Bank’s evaluation of World Bank support to pri- program in the MENA region has been rela- mary education argued, “past education tively less active in supporting school auton- reforms failed to focus on incentives and omy and accountability when compared with public accountability“ and “few Bank- other regions. Although 14 projects are sup- suppor ted countr y prog rams directly porting school autonomy and accountability addressed teacher recruitment and perfor- in the MENA region (slightly above average mance incentives; particularly lacking are when compared with other regions), the vast performance incentives related to student majority of them include such support at the learning outcomes.” Rightly, it concluded subcomponent level rather than the compo- that the “new road requires a new balance of nent level, and there is no stand-alone school engineering, incentives, and public account- autonomy and accountability project. ability measures” (IEG 2006). In the health sector, the organizational In recent years, things have changed. In and management tools introduced by the education, analytical work in particular has World Bank from the 1990s on to enhance begun to follow “the new road” in introduc- hospital performance were mostly directed at ing new governance tools for education. The the level of the Ministry of Health rather 2011 SABER4 evaluation for Tunisia, for than the hospital. The same could be said of example, listed “Teacher Motivation” as one the procedures and information systems of eight goals and focused on formal mecha- deemed necessary for improving decision nisms to hold teachers accountable as well as making and resource allocation, although the performance-related incentives (World Bank low level of support for autonomy/autono- 2011b). Assessment tools such as SABER aim mous management is no doubt rooted in the for a comprehensive analysis of factors that centralized systems in the MENA countries, affect education services delivery, including where virtually all the decision-making early childhood development, education power resides in the line ministries. As for resilience, school autonomy and accountabil- accountability, the governance concept was ity, school finance, school health and school mostly supported as an “internal” process in feeding programs, student assessment, and service delivery—that is, monitoring mecha- teacher and workforce development. During nisms and information transparency within the fi rst phase of an assessment, policy data the service delivery chain. Even though par- are analyzed based on an analytical frame- ticipatory elements were increasingly intro- work that uses global evidence to identify the duced throughout the early and mid-2000s, policies and institutions that matter most for (external) instruments to ensure that clients promoting learning for all. The second and would be able to hold the provider account- third phases explore quality of service deliv- able were rarely introduced or promoted by ery, focusing on policy implementation and the Bank. the quality of education services provided. Thus in the area of governance in social When looking at how World Bank interven- services delivery, the World Bank’s involve- tions are working toward the school auton- ment in the MENA region in the early 2000s omy and accountabilit y policy goals, was primarily aimed at ensuring the avail- however, SABER analysis shows that little ability of key inputs and improving provider has been done so far in the MENA region ability, seemingly ignoring the role of pro- compared with other regions (see figure 11.4). vider effort and (external) accountability The highest proportion of World Bank activ- mechanisms in social services delivery. Since ities lies in the area of the role of school then, the role of incentives and accountabil- councils in school governance or participa- ity has increasingly been acknowledged, and tion (policy goal 3); it accounts for a little the mechanisms for addressing them were more than 15 percent of interventions. DONOR SUPPORT 283 FIGURE 11.4 SABER analysis of World Bank school autonomy and accountability interventions, by region, FY03-13 80 70 60 50 Percent 40 30 20 10 0 AFR EAP ECA LAC MENA SAR Budget autonomy Autonomy in personnel management Role of school council (participation) Assessment Accountability Source: World Bank forthcoming. Note: SABER = Systems Approach for Better Education Results; AFR = Africa; EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin America and the Caribbean; MENA = Middle East and North Africa; SAR = South Asia. Relatively less emphasis has been on account- of direct empowerment as a goal of project ability in finance, school operations, and recipients was novel. learning (policy goal 5). Also in analytical work, the World Bank Another notable example of an analytical has pursued a stronger “citizen perspective”— tool is the new Benchmarking Governance as most of all in the MENA sector strategy for a Tool for Promoting Change initiative, 2013–18 (World Bank 2013b). The recently which today includes 100 universities in the introduced governance mechanisms, which MENA region (World Bank 2013a). It have to date mainly appeared in analytical focuses on governance in higher education, work but also to an extent in lending, are with an emphasis on management orienta- the expression of a shift in perspective for tion, autonomy, accountability, participation, World Bank support of social services in and public/private sector differentiation. the MENA region. Social accountability In health, the World Bank’s perspective is tools, in particular, are being used more also changing to include an increasingly frequently in the human development bottom-up focus on engaging with client sectors. Methodologies such as community countries. In 2011 a Japan Social Development scorecards have been piloted in the region, Fund (JSDF) project was approved, and its focusing on citizens as the ultimate stake- key objective was to “improve the quality and holders and thus adding to the Bank’s work responsiveness of health service delivery for on the supply side of governance (Beddies reproductive health through community et al. 2011). involvement and empowerment in planning and delivering health services in underserved areas of the Recipient’s peri-urban and rural The challenge of project governorates” (World Bank 2011a). Even implementation in the though the frequency of community partici- MENA region pation as a stated objective of development These promising advances still face con- projects had been steadily increasing since the siderable challenges when it comes to early 2000s, the inclusion in the JSDF project implementation via the World Bank’s 284 TRUST, VOICE, AND INCENTIVES lending instruments. In recent years, most system efficiency. The efficiency of investment Bank lending in the MENA region in edu- of project resources was modest, as project cation and health has taken the form of activities were either not completed or were investment lending, and, of the array of funded by other donors, some activities were instruments available for this purpose, two implemented late, and there was instability in in particular—the Adaptable Program Loan the implementing agency” (IEG 2013). APLs (APL) and the Learning and Innovation are designed specifically for strengthening Loan (LIL)—have been established for the express purpose of providing more flexi- bility and space for innovation during FIGURE 11.5 World Bank education and health project implementation (see box 11.1). In lending instruments: MENA region, 1994–2014 principle, these instruments should be well placed to facilitate the discovery of country- APL Other specific strategies for enhancing the quality 12% 7% DPL of service delivery in MENA countries. 3% ERL Over the period 1994–2014, 7 percent of 10% the Bank’s lending instruments were APLs, which accounted for 6 projects out of 82 in IPF 1% the MENA education and health lending LIL portfolio (figure 11.5). Two of these were 2% education projects with two phases (and therefore four projects in total), and two were SIM 10% health projects, approved in 2001 and 2008, including one for additional financing for the Djibouti Health Sector Development pro- TAL gram. All six projects were rated satisfactory 1% SIL or moderately satisfactory. The Djibouti 54% Second School Access and Improvement project was rated moderately unsatisfactory Source: World Bank. by the Independent Evaluation Group Note: MENA = Middle East and North Africa; SIL = Specific Investment Implementation Completion Report (IEG Loan; SIM = Sector Investment and Maintenance Loan; TAL = Technical Assistance Loan; APL = Adaptable Program Loan; DPL = Development ICR) review, largely because “evidence was Policy Loan; ERL = Emergency Recovery Loan; IPF = Investment Project not provided on improvements in quality and Financing; LIL = Learning and Innovation Loan. BOX 11.1 World Bank’s adaptable program loans and learning and innovation loans Adaptable Program Loans (APL) provide phased “risky,” or time-sensitive projects in order to pilot support for long-term development programs. An promising initiatives and build a consensus around APL is actually a series of loans in which each loan them or to experiment with an approach in order builds on the lessons learned from the previous loan(s) to develop locally based models prior to a larger- in the series. APLs are used when sustained changes scale intervention. LILs are predominantly used in in institutions, organizations, or behavior are deemed sectors or situations in which behavioral change central to implementing a program successfully. and stakeholder attitudes are critical to progress Learning and Innovation Loans (LIL) fi nance and where “prescriptive” approaches may not (at a level of $5 million or less) small, experimental, work well. DONOR SUPPORT 285 BOX 11.2 The independent evaluation group’s process for conducting implementation completion reports: How it works Project outcomes and World Bank performance different rating for Bank performance, it was either are assessed and rated in an Implementation “moderately unsatisfactory” or “unsatisfactory.” Completion Report (ICR), a document written by Such IEG ratings generally reflect significant weak- the Bank project team and subsequently reviewed nesses in project design and project implementa- by the Independent Evaluation Group (IEG) team. tion, while recognizing the often complex situations As part of its ICR review, the IEG independently on the ground. The difference between the ratings rates project performance. If the IEG rating differs reported by the project team in the ICR and by IEG from the ICR rating, the IEG explains the differ- in its ICR review often stems from the somewhat ence. On average, the IEG rating tends to be worse more subjective and insightful approach of the for- than the original ICR rating. In the case of MENA mer and the highly objective and rigorous approach human development projects, if the IEG reported a of the latter. institutional capacity, but in the MENA disbursed” (IEG 2011). Based on its modest region at least their performance does not relevance, modest efficacy, and negligible effi- stand out—they have not performed espe- ciency, the project was rated unsatisfactory. cially well or especially poorly (see box 11.2). As for the Yemen Higher Education Assessments of two early LILs in the edu- project, the ICRR asserted that the technical cation sector (approved in 2001) reached less and analytical work ordinarily essential for ambivalent conclusions: each was rated designing a higher education sector strategy unsatisfactory in both the ICR and the ICRR was not carried out. This could be a short- (Implementation Completion and Results coming of the LIL mechanism because its Report). “Based on the modest ratings for rel- capacity is limited (it is small) and because evance, efficacy, and efficiency,” the Morocco doing genuinely innovative work requires a Adult Literacy (Alpha Maroc) program was level of organizational capability that (by rated unsatisfactory, whereas the relevance definition) ministries struggling to implement and efficacy of the Yemen Higher Education key services may lack. Even so, again no pre- Learning and Innovation project were found requisites for piloting were put into place, to be modest at best (IEG 2009a; IEG 2011). which would have been a central objective of Although its objectives generally addressed a LIL. Thus weak project management could the Country Assistance Strategy priorities of also be a reason for the unsatisfactory project transparent governance, efficient public performance. administration, and creating social benefits A more recent Bank programmatic and human capital through investment in innovation— the Program for Results education, the Yemen Higher Education proj- (PfoR)—seeks to shift the core incentives for ect aimed to develop a sector strategy “with- development projects from the delivery of out the benefit of essential technical and inputs to disbursements linked to tangible analytic work, and the design failed to put and verifiable outcomes. 5 Disbursement- into place the prerequisites for piloting, the linked indicators (DLIs) play a critical role in one highly relevant objective. . . . There is no this process; they provide the participating evidence of efficiency; in fact, despite an government with incentives to achieve key attempt to significantly scale up project activ- program milestones and improve perfor- ities, less than two-thirds of the credit was mance. DLIs can be outcomes, outputs, 286 TRUST, VOICE, AND INCENTIVES intermediate outcomes, or process indicators. closing in 2006. According to the ICR, the Of the PfoR portfolio and pipeline opera- project’s specific goals for education pro- tions, 34 percent are mapped to the World grams and processes were overly ambitious— Bank’s Health, Nutrition, and Population even with four closing date extensions and (HNP) Global Practice, followed by the 13 years of project implementation, only 2 of Governance Global Practice with 14 percent. 18 targets were confi rmed as achieved. The The HNP sector of the Bank has prepared problems with this project were related to and implemented the most PfoR projects thus incomplete social analysis and public aware- far. The Morocco Health Sector Support ness outreach, incomplete preparation of Program for Results project is currently school grant activity, lack of a baseline for under way in the MENA region, and three evaluation of development outcomes, and more PfoR projects are in the pipeline. One lack of satisfactory monitoring and evalua- project was recently approved, and one is tion designs. Surprisingly, more or less the under implementation (National Initiative for same assessments were made in the IEG Human Development Support, Phase II, report on higher education development proj- Morocco)—see World Bank (2012). Taken ects in Egypt, Jordan, and the Republic of together, MENA countries account for up to Yemen, as well as in the ICR for the 25 percent of total PfoR project fi nancing, Moroccan National Initiative for Human second only to Africa with 37 percent. There Development Support project. is no indication yet of how PfoR projects are In education, projects in the MENA region performing in the MENA region, but such a seem to have focused on inputs or immediate fundamental shift in how incentives and conditions (sometimes successfully) but with- imperatives are structured in loan agreements out reliable measurements of student learn- will surely yield interesting findings when the ing. There was thus no indication of whether assessments are eventually conducted. these inputs actually made a difference in More broadly, project performance terms of student outcomes. For several proj- remains critical in both the education and ects, the links among outputs, outcomes, and health sectors in MENA countries. The rea- project development objectives were not sons for project weaknesses vary—whether clear, and the indicators did not capture the in the concept review stage or implementa- real results of the operation. tion stage; however, the final assessment con- In health, assessments of programs in the tinues to be project failure. Regarding overall MENA region also found that a focus on the project design, Denizer, Kaufmann, and availability of key inputs undermined project Kraay (2013) assert that “project size mat- effectiveness because institutional aspects ters,” meaning that large projects usually were neglected. In the Moroccan Social comprise several components and are thus Priorities program (Basic Health Project), for more complex, which has a negative bearing example, an emphasis on infrastructure, on successful project completion. Project per- drugs, and equipment resulted in timely pro- formance is therefore affected by project size curement. However, little attention was paid and design. Denizer et al.’s quantitative anal- to the actual quality of the services delivered. ysis holds up in the MENA region, where a Despite efforts to avoid the outcome of an majority of ICRs relate project difficulties to earlier project in which newly built and overly ambitious project goals for too short a stocked facilities could not open because of period of time. lack of personnel, this outcome was repeated. Indeed, 56 of 72 closed projects in educa- But this weakness in project implementation tion and health had to revise their closing was hardly acknowledged in the ICR, where dates. One example is the Secondary the Bank’s performance was rated as satisfac- Education Enhancement project in Egypt. It tory because “identified unexpected prob- was implemented in 1999, but did not close lems were found to be primarily of an until 2012, six years after its scheduled institutional nature with structural roots, DONOR SUPPORT 287 residing in domains beyond the scope of the least as important as country fixed effects in current project” (World Bank 2004). The accounting for variation in project outcomes. IEG review of the ICR review then reversed Accordingly, the ICR for the Yemen Higher this rating, finding that although “most phys- Education Learning and Innovation project ical targets were met, several facilities were stated that the weak project delivery was also closed and those that remain open are not related to overloaded task team leaders fully functional, service quality is low, and (TTLs) and slowly responding sector man- they are underutilized” (IEG 2004). Notably, agement. The TTL and other team members the project ICR was prepared in 2004, changed four times during the project, and and the project had been approved in 1996. the “frequent turnover of the staff resulted in Project thinking and expectations thus vary discontinuity and inconsistency in the quality in the time between project design and its of technical supervision” (World Bank 2009). assessment. This was a key issue for most of Another essential point for project design the projects because their time frames were is highlighted in the IEG health financing revised extensively and frequently. evaluation (IEG forthcoming). Specifically, During project implementation, the task the evaluation asks for systematic country team composes ISRs (Implementation Status diagnostics that address health financing as a and Results Reports) to record the project cross-cutting issue in country engagements status. Often, however, the ISRs are not cre- and also relate it to questions of public ated with the due diligence needed to foresee finance and budget management, ensuring an or respond to project complications. For analysis of equity in health services use and example, the ICR for the Yemen Higher finance, financial protection, and financial Education Learning and Innovation project sustainability. Thus there is an increasingly (which closed in 2008) noted that the strong emphasis on approaching service “ISR for the mid-term review was recorded delivery from a range of disciplinary perspec- 11 months after the review took place” tives, which is essential when working toward (World Bank 2009). Furthermore, it stated, improving the governance of social services. “Ratings in ISRs were unrealistic, with PDO There are some strong early warning indi- [project development objective] progress cators of project outcome ratings. For exam- rated ‘satisfactory’ even though the disburse- ple, higher preparation costs are often ment and overall implementation was very associated with eventual low project outcome slow, with missed deadlines repeatedly for ratings. These costs can stem from several planned activities, and the (legally required) factors that seemingly do not improve despite six-monthly reports were not available to significant resources flowing into project substantiate the progress.” The activities of preparation—factors such as undue initial the project had changed without acknowl- project complexity or limited country owner- edging that the agreed-on indicators had ship. It is statistically proven that early warn- been modified. Until the last year of the proj- ing signals, including “problem project” flags ect, the ISR did not highlight this discon- and “monitoring and evaluation” flags often nect, which in the end led to a “moderately predict the project outcome rating. Based on unsatisfactory” rating in project outcome this finding, Denizer, Kaufmann, and Kraay and an “unsatisfactory” rating for Bank (2013) reasoned that the “overall rate of ‘sat- performance. isfactory’ World Bank projects could be The composition of the task team is improved if incentives to significantly restruc- important for project implementation. ture or simply cancel problem projects at the Denizer, Kaufmann, and Kraay (2013) found implementation stage were strengthened, and that “task manager characteristics are impor- by increased emphasis on monitoring and tant and have quantitatively large and signifi- evaluation over the life of the project.” This cant impacts on project performance.” They observation aligns with the earlier finding argued that task manager fixed effects are at that weak monitoring and evaluation often 288 TRUST, VOICE, AND INCENTIVES signal fatal problems in project implementa- politically sensitive decisions and fundamen- tion and evaluation. tal changes in the way technical ministries And what is the role of project restructur- operate (OED 2001). The apparent objectiv- ing in improved project outcomes? The lack ity of IEG reviews enables sensitive questions of incentives for restructuring in the Yemen and critical assessments of Bank procedures. Higher Education Learning and Innovation For example, the Jordan Country Assistance project possibly led the task team to resist Evaluation of 2004 asks why, in view of the formally restructuring the project despite importance of education to the government fundamental changes that were proposed by of Jordan, did the Bank fail to undertake “a the ministry and incorporated into the proj- review of the whole education sector at any ect during its implementation. The ICR, in time during the eighties or nineties in order rating the project outcome as moderately to determine priorities between the three lev- unsatisfactory, asserted that “if the team els of education and hence to ascertain the had focused fully on the implications of the most important areas for Bank assistance to changes proposed by the Ministry and con- the sector” (OED 2004). A few paragraphs sidered a formal restructuring, the tradeoffs later, it becomes apparent that by the time an and the capacity requirements could have education sector review was planned for been better taken into account and the indi- Jordan, the Bank had already financed cators revised to match agreed revised 10 education projects without an assessment objectives” (World Bank 2009). of the strategic objectives of the education Finally, Denizer, Kaufmann, and Kraay sector as a whole. Examples like this are (2013) found that the time between project numerous, and yet many of the IEG assess- completion and evaluation is important to ments echo each other, calling out the same project outcome ratings. True project out- challenges year after year. Thus the Bank’s comes only become apparent over time, high- responsiveness remains questionable: evi- lighting the need for a longer time period dence is lacking on the effectiveness of the between project completion and its evalua- Bank’s response to the evaluations. tion. Thus it is important to establish incen- When speaking frankly about the chal- tives that reward the long-term impact of lenges confronting all donors in the MENA projects more than their satisfactory comple- region, however, one should also look for, tion. This is especially true for the education and look to, those places and spaces where and health sectors, where projects may gener- more promising initiatives are under way. ate positive student or health outcomes only One such place might be the Republic of years after project completion. Yemen,6 where the Second Basic Education Development project explicitly concedes that “input-based approaches” to improving edu- Lessons and opportunities going cation have been ineffective and “there is no forward evidence that teacher training and supervi- Evaluations of sector and regional work offer sion . . . has resulted in better teaching,” all a comprehensive understanding of the com- while “quality indicators at the country level mon approaches and challenges in World remain poor” (World Bank 2013c). As such, Bank operations across sectors and regions. the project is adopting a competency-based Country Assistance Evaluations (CAEs), in approach in conjunction with a conditional particular, illustrate the bigger picture over cash transfer program to try to encourage time, reflecting on the Bank’s work in the both high attendance and a focus on learning country context. This is especially important outcomes. Adjustments have been made to for the MENA countries. Bank sector work accommodate the seemingly “overambitious such as in education and health has often design”7 that initially characterized the proj- been less effective than it should have been ect, and it is hoped that other necessary because recommendations have required adjustments will be made along the way. It is DONOR SUPPORT 289 far too early, of course, to make summary implementation dynamics in a country beset judgment on the success of this project. The with conflict can begin to make a real differ- point is that an honest reckoning is being ence in the lives of poor and marginalized made with past approaches, a shift has been groups. initiated to focus on outcomes (not just inputs), and midcourse adjustments are being undertaken as immediate feedback becomes Conclusion: Are there even more available. “lessons” to be learned? The Republic of Yemen is also home to the This report and this chapter have sought to Secondary Education Development and Girls provide both a realistic sense of the immense Access Project (SEDGAP)—see box 11.3. development challenges facing the MENA Here we see how a deep commitment to region in service delivery—for donors and BOX 11.3 Implementing education projects in the Republic of Yemen: The Secondary Education Development and Girls Access Project (SEDGAP) Mouna’s story Republic of Yemen has proven to be a demanding I got married when I was in seventh grade and gave work environment for the Bank. Project implemen- birth to nine children because my father-in-law tation and evaluation have been difficult and risky. didn’t allow me to use contraceptives. I insisted on However, it is in the personal stories like Mouna’s continuing to study. I took my children with me that one can learn about the true impact and suc- to the school and walked almost 8 kilometers to cess of a project against all odds. school. After fi nishing my education I worked as a volunteer teacher. When I joined the school, only The context 10 girls studied there, but after I started teaching, In the Republic of Yemen, the poverty rate rose to the number of girls went up to 72. 54.5 percent in 2012. The severe unrest since 2011 I have benefi ted from the female teacher con- and the high population growth rate are placing tracting scheme, which has a great impact on my growing pressures on the education system. life. My role now is not just teaching but also And yet the last decade has seen significant advocating for girls’ education. I make the best improvements. The basic education gross enrollment use of every social gathering and speak about ratio (GER) rose from 68 percent in school year the importance of education and sending girls to 1998–99 to 86 percent in 2011–12. Girls’ enroll- schools. ment has surged over the same time period, with the Mouna lives in the Republic of Yemen, one of GER rising from 42 percent to 76 percent for basic the poorest countries in the MENA region. Over education and from 16 percent to 23 percent for sec- the last 10 years, World Bank teams have worked ondary education. Nevertheless, many children drop to improve the living conditions and educational out of school early, particularly girls in rural areas. opportunities for women like Mouna and their An estimated 2 million 6- to 15-year-olds were out families. Over the period 1994–2014, the lending of school in 2011. School survival rates are also low: portfolio for the education and health sectors of the only half of those who enter grade 1 reach the end Republic of Yemen totaled almost $300 million, of basic education. And there is the persistent gen- making the country the third largest aid recipi- der inequality: the 2009/10 grade 6 completion rates ent in the MENA region for those respective sec- were 51 percent for girls and 71 percent for boys tors. Shaken by conflict and severe crises, the (61 percent for all). (continued next page) 290 TRUST, VOICE, AND INCENTIVES BOX 11.3 Implementing education projects in the Republic of Yemen: The Secondary Education Development and Girls Access Project (SEDGAP) (continued) The project Results and experiences The Secondary Education Development and Girls In spite of the difficult context, the Ministry of Access Project (SEDGAP) was approved in 2008, Education strenuously pursued implementation of and it closed in January 2015. The objective of the project as planned, especially the incentive pro- SEDGAP was to improve the gender equity, quality, grams. These programs appear to have had positive, and efficiency of secondary education in targeted dis- tangible impacts in the targeted districts—for exam- tricts, with a particular focus on girls in rural areas. ple, transportation incentives led to the improved The project fi nanced a set of supply- and demand- retention of boys and girls who had to travel long side interventions (civil works, provision of school distances to attend school and contributed to reduc- materials, capacity-building activities, and teacher ing the dropout rate and increasing the enrollment training), including incentive programs such as rate of students. Also, based on recent data from the conditional cash transfers and transportation incen- Ministry of Education, the conditional cash transfer tives. The project was initially cofinanced by five schemes have proved highly effective in increasing development partners for a total of $103.4 million. the enrollment of boys and girls from disadvantaged In 2010 the upheaval in the country resulted in a sig- areas. Stories like Mouna’s show that every small nificant reduction in project funds, to $47.2 million, step counts in improving the lives of the poor. as some development partners withdrew from the Source: World Bank staff. project or reduced their contribution. Note: Names have been changed to protect identities. domestic actors alike—as well as a vantage 21st century and ensuring health care for all. point from which opportunities for improve- If citizens are to trust the programs and sys- ment might be apparent. As this report tems supported by donors, donors may need explains, enhancing service delivery—and to reach out to citizens directly to build trust. especially addressing the quality challenges Simple measures such as providing citizens in education and health services—requires a with detailed information on donor support thorough understanding of political, admin- objectives, interventions, and cost, disaggre- istrative, and social institutions; performance gated to the village level, would be a step for- indicators; and citizen trust and engagement ward and a possible model for promoting at the national level and in local contexts. It transparency domestically. also requires devising solutions that would A n important factor in identif ying fit—and gradually enhance—the cycle of opportunities is being able to locate and performance and would be embraced by local learn from those who have already made leaders and their communities. the system work and building on the The key challenge is building high- insights gained from them. If governments, capability education and health systems as donors, and others are to engage in the type part of the whole cycle of performance. If they of analysis that yields such insights, it is are embedded in functioning accountability important that the World Bank itself be mechanisms that are trusted by citizens and self-critical and that it analyze the varia- that benefit from citizens’ feedback and tions in its own operational performance, action, education and health systems would seeking opportunities for improvement. be better able to build on the initial logistical This chapter reveals how the Bank has successes of infrastructure provision to tackle changed its approach to (and rationale for) the more complex but vexing tasks of ensuring service delivery over the years. However, it quality learning that prepares students for the has consistently struggled to actively DONOR SUPPORT 291 support more than the provision of basic will draw on the community participation inputs, even when deploying instruments approach pioneered in the mid-2000s after expressly designed to facilitate more “adap- publication of World Development Report tive” implementation. Perhaps the harder 2000/2001 (World Bank 2001). And one can question, because of recent experiences in reasonably ask to what extent social account- the MENA region, is whether the Bank, ability tools deployed by “communities” are with its current procedures and incentive likely to overcome powerful institutional structures, can support an approach to proj- pressures (or inertias). Even, or especially, on ect design and implementation more focused this point, however, there will be variation: on solving locally prioritized problems and the social compact binding citizens and state delivering the real results that citizens right- will be reimagined in different ways in dif- fully expect. Many of the concerns raised in ferent places in different sectors. Building the recent high-profi le IEG review, Learning the collective capability to be willing to try and Results in World Bank Operations: new things, to demand high standards from How the Bank Learns (IEG 2014), seem to each other, and to share with and learn from apply to the MENA region. The question, each other is the challenge ahead. then, is can the World Bank now be a leader in pioneering a different (and potentially Notes more effective) approach? A key lesson from this analysis is that how 1. The time horizons are broad and mostly mark the introduction of new concepts rather engagement is structured among donors, gov- than the completion or even abandonment of ernments, and citizens matters. In other the old ones. words, the usual focus on policy reforms in 2. See http://aiddata.org/. the abstract must be matched by a corre- 3. Thanks to Dan Honig for conducting this sponding focus on how any given policy will analysis. As part of his PhD research, Honig actually be implemented, and how the pre- extended the aidData database to enable vailing system of political imperatives and analysis of the larger coverage of projects incentives will shift in favor of learning and reported here. effective delivery (as opposed to just process 4. The World Bank’s Systems Approach for compliance). One might hope that the new Better Education Results (SABER) is an ini- “science of delivery” approach provides a tiative to provide comprehensive comparative data and knowledge on education systems space within which such issues can be around the world. explored by looking at the nature of a prob- 5. h t t p : / / w e b . w o r l d b a n k . o r g / W B S I T E lem and developing a hypothesis while being / EXTERNAL/PROJECTS/0,,contentMDK agnostic about the solution; by using evidence :23215867~pagePK:41367~piPK:51533~the to inform the implementation of solutions; by SitePK:40941,00.html. taking an adaptable, creative, and context- 6. See also Strengthening World Bank Portfolio driven approach; and by being able to cap- Performance in the Republic of Yemen ture cumulative knowledge when finding and (World Bank 2014) for a fuller accounting of fitting local solutions. how the Bank is actively seeking to improve For now, social accountability is a prom- the quality of its engagement in the Republic ising approach for the MENA region, even if of Yemen. 7. See page 2 of http://www-wds.worldbank its implementation requires time, money, .org/external/default/WDSContentServer and expertise so that it can be appropriately /WDSP/MNA/2014/08/23/090224b082667 adapted to local contexts. 8 At the 2014 34b/1_0/Rendered/PDF/Yemen00Republi0R spring meetings of the World Bank, main- eport000Sequence004.pdf. streaming citizen engagement officially 8. h t t p : / / s i t e r e s o u r c e s . w o r l d b a n k . o r g became an imperative across the MENA /EXTSOCIALDEVELOPMENT portfolio. It remains to be seen exactly what /Resources/244362-1193949504055 form this imperative will take and how it /Scalingup.pdf. 292 TRUST, VOICE, AND INCENTIVES References ———. 20 0 4. “Jordan: A n Evaluation of World Bank Assistance for Poverty Reduction, Beddies, S., M. Felicio, G. Dedu, F. Fal, and Education, and Health: A Country Assistance C. Vagneron. 2011. “Middle East and North Evaluation.” World Bank, Washington, DC. Africa Local Service Delivery Initiative: R i ngold , D., A . Hol la , M . Koz iol , a nd Promoting Social Accountability and Demand S. Srinivasan. 2012. Citizens and Service for Good Governance.” Arab World Brief No. Delive r y: A ssessing the Use of Soci al 3, April, World Bank, Washington, DC. Accountabilit y Approaches in the Dehn, J., R. Reinikka, and J. Svensson. 2002. Human Development Sectors. Washington, Survey Tools for Assessing Service Delivery. DC: World Bank. Development Research Group. Washington, World Bank. 1980. “Health Sector Policy Paper.” DC: World Bank. World Bank, Washington, DC. Denizer, C., D. Kaufmann, and A. Kraay. 2013. ———. 1987. Financing Health Services in “Good Countries or Good Projects? Macro Developing Countries. Washington, DC: and Micro Correlates of World Bank Project World Bank. Performance.” Journ al of Development ———. 1995. “Education Sector Strategy Paper.” Economics 105: 288–302. World Bank, Washington, DC. IEG (Independent Evaluation Group). 2004. ———. 1997. “Health, Nutrition, Population “Implementation Completion and Results (HNP) Sector Strategy 1997.” World Bank, Report (ICR) Review: Basic Health Project.” Washington, DC. World Bank, Washington, DC. ———. 2001. World Development Report ———. 2006. From Schooling Access to Learning 2000/2001: Attacking Poverty. Washington, Outcomes: An Unfinished Agenda: An DC: World Bank. Evaluation of World Bank Support to Primary ———. 2002. Arab Republic of Egypt: Education Education. Washington, DC: World Bank. Sector Review: Progress and Priorities for the ———. 2009a. “Implementation Completion Future, Vol. 1. Washington, DC: World Bank. Report (ICR) Review: Alpha Maroc Project.” ———. 2004. “Kingdom of Morocco: Social World Bank, Washington, DC. Priorities Program: Basic Health Project: ———. 2009b. Improving Effectiveness and Implementation Completion and Results Outcomes for the Poor in Health, Nutrition, Report.” World Bank, Washington, DC. and Population: An Evaluation of World Bank ———. 2008. The Road Not Traveled: Education S uppor t since 1997. Washing ton, DC: Reform in the Middle East and North Africa. World Bank. Washington, DC: World Bank. ———. 2011. “Implementation Completion ———. 2009. “Republic of Yemen: Higher Report (ICR) Review: Higher Education Education Learning and Innovation Project: L e a r n i n g a n d I n n o v a t i o n P r o j e c t .” Implementation Completion and Results World Bank, Washington, DC. Report.” World Bank, Washington, DC. ———. 2013. “Implementation Completion ———. 2011a. “Republic of Tunisia: JSDF Repor t ( IC R) Rev iew: S e cond S chool Grant for Community Health Collaborative Access and Improvement.” World Bank, Project. Grant No. TF099374.” World Bank, Washington, DC. Washington, DC. ——— . 2 014 . L e a r n i n g a n d R e s u l t s i n ———. 2011b. “SABER Country Report for World Bank Operations: How the Bank Tunisia.” World Bank, Washington, DC. Learns. Washington, DC: World Bank. ———. 2012. “Kingdom of Morocco: National ———. Forthcoming. “Tunisia Country Program Initiative for Human Development Support Eva lu at io n , F Y05 –13 .” Wo rld B a n k , Project (INDH): Implementation Completion Washington, DC. a n d R e s u l t s R e p o r t .” Wo r l d B a n k , ———. Forthcoming. World Bank Group Support Washington, DC. to Health Fin ancing for Improving ———. 2013a. Benchmarking Governance as a Health System Performance. Washington, DC: Tool for Promoting Change: 100 Universities World Bank. in MENA Paving the Way. Washington, DC: Operations Evaluation Department. 2001. World Bank. “Yemen Country Assistance Evaluation.” ———. 2013b. Fairness and Accountability: World Bank, Washington, DC. Engaging in Health Systems in the Middle DONOR SUPPORT 293 East and North Africa. Washington, DC: /WBSITE/EXTERNAL/TOPICS World Bank. / EXTSOCIALDEVELOPMENT/EXTPCENG ———. 2013c. “Republic of Yemen: Second Basic /0,,contentMDK:20507700~pagePK:148956 Education Development Project: Project ~piPK:216618~theSitePK:410306,00.html A p p r a i s a l D o c u m e n t .” Wo r l d B a n k , QSDS (Quantitative Service Delivery Survey), Washington, DC. World Bank, ht tp: //web.worldbank.org ———. 2014. S tre ng the ning World Bank /WBSITE/EXTERNAL/TOPICS Portfolio Performance in the Republic of / EXTPOVERTY/EXTPSIA/0,,contentMDK: Yemen. Washington, DC: World Bank. 20467190 ~isC U R L:Y~menu PK:1108016 ———. Forthcoming. World Bank Support ~p a g e P K :148956 ~pi P K : 216 618 ~t h e S it for School Autonomy and Accountability ePK:490130,00.html FY03–13. Washington, DC: World Bank. SABER (Systems Approach for Better Education Results), World Bank, http://saber.worldbank .org/index.cfm Data sources UNESCO Institute for Statistics, education, PETS (Public Expenditure Tracking Survey), http://www.uis.unesco.org/Education/Pages World Bank, ht tp: //web.worldbank.org /default.aspx Incremental Changes in the Social Contract and 12 Local Empowerment • Enhancing service quality requires building institutions and promoting citizens’ trust and engagement. • The social contract among policy makers, public servants, citizens, and providers can be improved through incremental changes to strengthen accountability mechanisms, internal control, and performance management as well as modify recruitment and promotion mechanisms. • Empowering communities and local leaders and building coalitions of reformers inside and outside of government can help in the design and implementation of best-fit solutions. This concluding chapter is devoted to possible of instability and fragility may generate a solutions to the social services delivery prob- desire for quick wins. With that in mind, in lems in the countries of the Middle East and this chapter we explore possible approaches North Africa (MENA). Clearly, many policy to incremental systemic reforms, options for makers across the MENA countries want to empowering communities and local leaders deliver visible results and, in doing so, bolster to find local solutions, and possible quick their authority and public support within an wins to enhance citizens’ experience with the atmosphere of political fragility. They may state, especially in the areas of education and not have the will, the power, or the support health care. to drastically shake up the entire system of Instead of sector-specific reforms, such as service delivery institutions and accountabil- teacher policies, school management, or ity mechanisms, but they may be willing and provider payment mechanisms and quality able to make incremental systemic changes management in hospitals, we focus on cross- and allow local initiative in motivating and cutting governance elements that critically supporting teachers and health workers. As affect service delivery performance. We argue we explained in chapter 10, conflicts, crises, that the roots of service delivery problems, and transitions in the MENA region also such as provider absenteeism, poor quality of may create opportunities for wide-ranging teaching or care, and shortages of medicines systemic reforms. Similarly, the atmosphere and textbooks, can often be traced to these 295 296 TRUST, VOICE, AND INCENTIVES cross-cutting governance elements. Examples servants accountable.1 However, only strong of these elements are transparency, public administrative institutions can enable the sector management including civil service, state and providers to actually respond to independent accountability institutions such citizens’ voices and their needs. Indeed, an as courts, and an enabling environment for ability to respond to citizens’ feedback on the citizen action. Efforts at the cross-cutting quality of service delivery is crucial to sus- governance level need to accompany sector- taining citizens’ trust and participation. specific solutions in education and health. Likewise, policy reforms, and the mecha- nisms for their implementation, need to draw on evidence, including fi ndings from moni- Lessons from experience toring, evaluation, and citizens’ feedback in Experience has shown that policy reforms— local contexts. and public spending—are typically necessary Experience also suggests that for institu- but not sufficient to improve the delivery of tional reforms to have an impact, they have social services. Drawing on the literature and to emerge from problem-led learning especially on evidence from the MENA processes, facilitate the finding and fitting of region, our analysis indicates that without context-specific solutions, and engage broad adequate political and administrative institu- groups to ensure that new institutions are tions, accountability mechanisms, and inter- shared and embedded. Historically, institu- nalized norms of personal responsibility and tional reforms that overlooked contextual public service, improvements in education realities often failed. The problem-driven and health services will not come simply iterative adaptation approach to institutional through policy reforms, through moderniza- reforms that is showing more promise calls tion of schools and health facilities, or for reform interventions to focus on solving through training of educators and health problems through purposive muddling that professionals. To foster better performance, includes active, ongoing experiential learn- policy reforms and investments need the ing, with engagement by broad sets of agents backing of institutions—especially incentives who together ensure that reforms are viable and norms embedded in both formal and and relevant (Andrews 2013; Andrews, informal accountability relationships—and Pritchett, and Woolcock 2013). citizens’ trust and engagement. The success of institutional and policy Similarly, narrowly focused institutional reforms depends on the actual incentives that reforms within the education and health sec- prevail for stakeholders associated with a tors will have only a limited impact if they specific problem in a specific setting. are not anchored in broader institutional Understanding the way in which political, changes. Experience suggests that perfor- economic, and social interests intersect can mance in service delivery improves when help in designing both institutional and pol- political institutions are the primary drivers icy reforms in a way that will strengthen pro- of outcomes, or—as our case studies in reform coalitions and encourage compromise. chapter 3 illustrate—when skillful leaders use In this respect, political economy analysis them to tap into and exploit social institu- can help connect data about decisions, the de tions for better outcomes. Decentralization, jure design of institutions, and the de facto incorporated in a broad package of reforms use of institutions, thereby identifying con- aimed at putting more power into the hands straints and opportunities for progressive of local officials, can help strengthen incen- change. Within the existing constraints, an tives for better performance if supported by incremental, problem-driven approach to adequate accountability mechanisms and institutional and policy reforms can combine resources. Meanwhile, enabling environ- considerations about feasibility with consid- ments for collective action are needed so that erations about finding solutions that are citizens can hold local and central public robust and meaningful. This approach can INCREMENTAL CHANGES IN THE SOCIAL CONTRAC T AND LOCAL EMPOWERMENT 297 adapt reform design to align with the existing outcomes. Instead, greater improvements in reform space and gradually expand it (Fritz, service delivery can be achieved by aligning Levy, and Ort 2014). the incentives of policy makers, public ser- Recognizing that there is no magic bullet vants, and service providers with the needs on these fronts, we do not try to provide a and rights of citizens. The way forward can comprehensive set of solutions. Instead, we involve strategic incrementalism to give pub- highlight examples of success in the region lic servants and providers incentives to and elsewhere to inspire policy makers, pub- improve their performance and promote citi- lic servants, citizens, and donors across the zens’ trust and engagement. Incremental MENA countries. The MENA region is striv- changes in incentives at the center of govern- ing to overcome its complicated political ment, in sectors, and at the local level can transitions and tragic conflicts, and the key is transform the social contract in the MENA to seek and learn from positive examples. countries, sending them down the path to As Ibish recently observed when reflecting on greater meritocracy in the public and private the challenges the region is facing, “Under sectors alike. Such changes will elevate the such circumstances, it is an intellectual and value of education and skills in meeting the political moral duty to look for (but not needs of the economy and society. Measures invent) real evidence that allows one to retain promoting transparency, accountability, and a sense of decency and openness to a better norms of personal responsibility and public future. And such evidence genuinely does service can play an important role in this exist in the Arab world today, despite a ‘big effort, both at the center of government as picture’ that is, or at least currently seems, so well as at the level of service providers and unremittingly appalling” (Ibish 2014). their respective sectoral departments. Develop effective accountability institutions Strengthening accountability to monitor the performance of service pro- and incentives through viders and provide tools for the resolution of incremental changes complaints related to service delivery. Capacity is important, but it is not a promis- ing area for improving the quality of service Independent oversight institutions such as delivery in the MENA countries. Where supreme audit institutions, ombudsmen, and capacity is very low, institutional incentives courts are critical components of national are not, of course, sufficient to improve per- accountability systems. As auditing moves formance. For example, incentives for nurses beyond fiduciary matters to questions of in clinics will have little impact on the quality results and effectiveness, it can make govern- of health care if the nurses lack the skills ment more transparent and more accountable needed to serve patients, and those incentives for what has been accomplished with public that bring teachers to classrooms may still fail money. Supreme audit institutions have to provide a good education if the classrooms increasingly recognized the importance of are overflowing with students. Overall, with demonstrating relevance to citizens by being the exception of low-income countries such as a credible source of independent and objec- Djibouti and the Republic of Yemen and situ- tive insight and guidance to support benefi- ations of conflict and refugee crises, low ser- cial change in government and service vice delivery capacity does not appear to be delivery. An ombudsman, who deals with the key binding constraint in the education complaints from the public about decisions, and health sectors in the MENA countries. actions, or omissions of public administra- Meanwhile, the evidence suggests that tion, serves to protect the people against vio- undertaking policy reforms, building schools lations of rights, abuses of power, errors, and hospitals, and providing equipment and negligence, unfair decisions, and maladmin- training are alone unlikely to yield better istration, and to improve public services 298 TRUST, VOICE, AND INCENTIVES while making the government’s actions administrative agencies, complaints, and more open and its administration more particularly behaviors such as unjustly refus- accountable to the public. Courts, particu- ing receipt and handling of petitions on the larly administrative courts, can help support grounds that they may be later pinpointed by citizens’ exercise of their legal rights to health audit and inspection.” The hotline also services and sanction service providers for receives “reports of corruption and fraud of failing to implement their mandates. Because public officials, including bribery, idleness, accountability institutions disseminate their embezzlement and the misappropriation of findings publicly, assist citizens, and interact public funds.” This mechanism has been with the state and service providers, they also widely disseminated in Korean society and help to strengthen citizens’ trust and engage- has a dedicated page on the SAO’s website. ment in the state. In addition, the SAO has established special- For their part, governments must ensure ized channels for receiving audit requests from that accountability institutions are accessible citizens about matters of public interest and to all citizens. Awareness and information corruption. A committee of auditor generals campaigns should be conducted so that citi- and civil society experts assesses these requests zens know their rights and how to exercise and responds to citizens within 30 days them. Accountability institutions should (Effective Institutions Platform 2014). receive adequate financial and personnel Accountability institutions have also resources and operate under clear regulatory proven effective at the subnational levels. In frameworks. Whistleblower legislation needs Pakistan’s Khyber Pakhtunkhwa province, to be in place to protect citizens who file for example, the provincial ombudsman complaints, and laws on libel and slander provided redress to about 1,800 citizens’ should not be used against citizens as a means grievances during 2012–13 and both carried of protecting corrupt or incompetent offi- out an awareness campaign and instituted a cials. Information on complaints to and citizen report card for 10 basic services investigations by accountability institutions (including education, health, and sanitation) should be made public and easily accessible. that targeted potential as well as actual Legal aid services should be made available service users.2 to ensure that poor and vulnerable citizens have access to accountability mechanisms. In Strengthen internal controls and perfor- facilitating access for all, governments can mance management, including mechanisms effectively partner with civil society organi- to share and act on information. zations. Governments should also experi- ment with the establishment of specialized A government can improve its perfor- accountability institutions geared toward mance by focusing on results in policy advice, the education and health sectors, such as central and departmental management medical-legal partnerships and health care processes, and public accountability. The rel- ombudsmen. ative priority of these areas is different in Accountability institutions can create for- each country. Countries worldwide have mal mechanisms to actively seek citizens’ adopted performance-oriented approaches to input and complaints and report back on the management, budgeting, personnel, and status of their requests, feedback, and institutional structures. input. The Superior Audit Office (SAO) of the However, such approaches are effective Republic of Korea established a complaint only when they are built on an effective sys- hotline and whistle-blower mechanism tem of control. Combating absenteeism and through which citizens can report areas sus- gaps in the distribution of key inputs such as pected of irregularities or corruption and instructional materials in schools and medi- request audits. The 188 hotline collects cines in health facilities depends on having “reports on unjust handling of petitions by internal management controls in place. INCREMENTAL CHANGES IN THE SOCIAL CONTRAC T AND LOCAL EMPOWERMENT 299 Service delivery in MENA countries would performance (such as pay for performance). benefit from strengthening the basic record- Finally, private service providers could oper- ing, monitoring, and management of inputs ate under a uniform regulatory framework, and transactions in the service delivery in parallel with public service providers, and chain and from greater transparency in the offer citizens choices grounded in the acces- public finance management and procure- sibility of information on performance such ment systems. Stronger ex post control and as standards-based school and hospital processes of internal control in the public assessments and accreditations. sector would facilitate policy implementa- Greater transparency of government tion and operation of the education and performance across departments and service health systems. delivery systems and providers can nurture Because of the complexity of the education formal and informal accountability for per- and health services delivery chains, social formance. Making information about the institutions also play an important role in performance of schools, health facilities, and motivating behavior, especially if the internal the overall education and health systems controls are weak and public management readily available to policy makers, public ser- moves from input controls to performance vants, providers, and the public can draw management. In accountability and control, attention to the deficiencies and variations in as in all other dimensions of management, results as well as in the service delivery pro- the informal systems—individuals’ motiva- cess. Such a measure is likely to create both tions, values, and attitudes—are as important informal and formal pressure for improve- as formal systems. Strategies to strengthen ment and can help in devising the appropriate control and accountability must take this into corrective and support measures. Introducing account or fail. Performance-oriented man- greater transparency in the relevant informa- agement can effectively complement (and tion on performance can increase citizens’ even partly replace) input and process con- trust in the state’s efforts to improve public trols when formal controls can be partly services, as well as empower action by citi- replaced by social controls as staff internalize zens at both the local and national levels. the appropriate values and norms as well as In this context, the MENA countries organizational goals. would need to strengthen the system of Supported by effective administrative and school and health facility inspections in order social accountability mechanisms, some to generate the relevant information on per- performance-oriented approaches would be formance and then to adjust human resource easy to implement in MENA countries. management and institutional capacity build- Information on results and the different ing in the education and health sectors in elements of performance, for example, could response to such information. Furthermore, be introduced into government reporting, information on the different performance ele- subjected to independent verification and ments of schools, health facilities, and the public debate, and used as part of public education and health systems can be bench- management. More advanced approaches marked across localities within countries to include introducing performance measures provide citizens with the relevant compara- into budgeting and management; delegating tive framework geographically and over time. responsibility to line ministries and agencies; Such information would cover the key aspects taking steps toward meritocracy in public of service provision such as continuity, com- employment; and privatizing and outsourcing prehensiveness, and appropriateness in health services that could be effectively provided by care services and student learning and the t he private sec tor a nd civ i l so ciet y. education-to-work transition in education Performance-oriented fi nancing of services services. could complement managerial accountability Other performance-oriented reforms in the by providing financial incentives for publ ic s e c tor wou ld t a ke t i me a nd 300 TRUST, VOICE, AND INCENTIVES much effort. Constructing a true performance in service delivery, and evaluating employee budget, for example, requires major changes performance, with consequences for promo- in information, costing, and measurement tions and a bonus structure (Iyer 2011). systems. Reformatting budget documents to Policy makers can encourage such agency- show workload or output data does not suf- led initiatives through public recognition fice; citizens also must receive accurate infor- and awards (the importance of which is mation on how spending options or choices further discussed in the next section) at both affect the services they receive from the gov- the subnational and national levels. For ernment (Schick 2011). Changing the atti- example, Dubai’s Executive Council intro- tudes of public servants and politicians duced the Hamdan Bin Mohammed Award toward performance requires creating specific for Smart Government of Dubai in 2013. incentives and controls as well as spreading During 2013–14, 32 awards recognized spe- an understanding of the education and health cific programs and services, including best systems and how the actions of key actors public service, best new government service, influence each other (OECD 2005). and best public-private partnership, as well as Although efforts to introduce performance best efforts such as best improvement team or management in a comprehensive, top-down best service center manager. Citizens voted to fashion have been met with only partial and select the winner from among the top three gradual success, many examples of success finalists; experts served as international can be found at the agency or sector level. judges in the preselection. The fi nal awards Such cases are particularly inspiring when were announced by the crown prince and, in succeeding against the odds in fragile, con- addition to public recognition, involved a fl ict-affected settings, as documented in the financial prize.3 Ministry of Basic and Secondary Education Governments can also establish programs and the Ministry of Finance and Economic that provide incentives for monitoring the Affairs in The Gambia; the Ministry of provision of quality services, apart from Public Works and Transport and Électricité private-public partnerships and reliance on du Laos in the Lao People’s Democratic civil society. An Egyptian conditional cash Republic; the Ministry of Finance and transfer program piloted in Ain es-Sira (a Economic Development and Local Councils suburb of Cairo) in 2009 helped poor women in Sierra Leone; and the Ministry of Health, ensure that their children went to, stayed in, the Central Bank, and the Ministry of Social and excelled at school, and also led to obvi- Solidarity in Timor-Leste (see Barma, ous improvements in health. Such programs, Huybens, and Viñuela 2014). when done well, can help repair relations Inspiring examples of promoting perfor- between the poor and the state. Hania mance management in the public sector are Sholkamy, the professor at the American also found in the MENA region. Morocco’s University of Cairo who helped to design the experience in redesigning a public agency program, explained: “Civil society can moni- (Caisse Nationale de Sécurité Social) involved tor programs, can organize social audits, can effective measures for enhancing account- provide auxiliary benefits and projects, can ability and efficiency as well as for streamlin- even take on the responsibility of providing ing and computerizing administrative work opportunities or better markets so that procedures, building a skilled staff, and con- families will find exits from poverty. But the ducting transparent inspections and audits state is the duty bearer in the case of social (Ferrali 2013). Jordan’s approach to creating protection, and for these transfers to work, a “citizen-friendly” Civil Status and Passports they must be entitlements, not handouts” Department consisted of overcoming a noto- (Sholkamy 2014). Civil society engagement at rious lack of motivation and corruption by the point of service delivery may indeed be overhauling the department’s highly central- supported by government as well as interna- ized structure, eliminating unnecessary steps tional development partners without posing INCREMENTAL CHANGES IN THE SOCIAL CONTRAC T AND LOCAL EMPOWERMENT 301 any risk to the underlying clientelist political national level are frequently not implemented systems. Civil society organizations, however, successfully. Often, such gaps are only exac- do not stand above the clientelistic pressures erbated by support of projects through that may be present in society, but some civil “parallel” institutions and vertical programs. society organizations, like some political Using such channels may seem more efficient actors, eschew such practices. in the short run, but they actually weaken Experimentation, even at the local service the state. delivery level, is important to test how reforms fit local and national institutions. Modify the mechanisms for selecting, Some well-meant control and performance- encouraging, and rewarding leaders, public oriented reforms can at times have limited or servants, and service providers in order to perverse impacts. For example, curriculum internalize norms of personal responsibility reforms implemented in the United Arab and public service. Emirates were geared toward improving edu- cation outcomes and creating an entrepre- Meeting citizens’ demands for education neurial sense of citizenship. But the students and health services delivery requires reorient- began to feel entitled (despite the fact that the ing the incentives and attitudes of policy curriculum was aimed at reducing their makers, public servants, and service provid- dependence on the state), perhaps because ers not only through formal accountability they were selected to attend a special school mechanisms but also through internalized (Jones 2013). Even mechanisms as simple as norms of personal responsibility and public time/date stamping machines were intro- service. Our case studies of local successes in duced to improve nurses’ job attendance in chapter 3 reveal that such norms are often a India. But after significantly reducing absen- powerful engine of good performance. teeism, the time-punch machines suffered Similarly, Al-Yahya (2009), drawing on a sur- from breakages (some of which were proven vey of administrators in public sector organi- deliberate), and staff reverted to their old zations in Saudi Arabia, suggested that norms habits of poor job attendance without any and organizational culture, including partici- disciplinary action (Banerjee, Glennerster, pative practices, are significant predicators of and Duflo 2008). In these cases, the reforms motivation and the effective utilization of in state institutions were undermined by competence. weak political accountability mechanisms, by Norms of personal responsibility and pub- social norms and institutions that remained lic service must be nurtured through human unchanged, and by the time lag between resource management in the overall public reforms and their observable results. sector and within the education and health Finally, effective implementation of poli- sectors. Selection, promotion, and recogni- cies and programs aimed at better provision tion mechanisms as well as training can help of services must also take into account the internalize values of collective purpose, pub- gap between the central government and its lic service, and individual responsibility for ministries and the local service providers. results, as well as specific professional values. Policy reforms by the national government This emphasis implies the need to nurture and ministries are unlikely to be implemented fundamental values such as fairness, equity, successfully in the absence of mechanisms justice, and social cohesion as part of public that bridge the gaps between them and local administration and service delivery. When service providers and motivate implementa- service providers, public servants, and policy tion at the local level. Currently, the weak or makers demonstrate such values in their distorted monitoring, control, accountability, actions, they are likely to strengthen trust in and incentive mechanisms permeating the the governmental and political system as a service delivery chain imply that well- whole. Where needed, the effort to internal- meaning otherwise impressive policies at the ize these values can be accompanied by 302 TRUST, VOICE, AND INCENTIVES deregulation of administrative controls that make teacher recruitment more selective by quash individual initiative and personal raising the standards for entry into teacher responsibility. education; raising the quality of teacher edu- The traditional centrally controlled cation schools; and raising the hiring stan- bureaucracy common in MENA countries is dards for new teachers. The second need is to a workable, robust system for internalizing make teachers more effective by supporting the norms needed to achieve quality service their development during their critical fi rst delivery for all. Internationally, such tradi- five years of teaching and assessing teachers’ tional bureaucracy models have been shown strengths and weaknesses; by offering train- to be effective, especially where the constitu- ing to remedy teachers’ identified weaknesses tional institutions of society have been dis- and leverage the skills of top performers; by rupted or discontinued; where the other matching teachers’ assignments to the needs institutions in society are not particularly of schools and students; and by building a well ordered; and where national culture professional community of teachers both attaches importance to the existence of a within schools and across the school system. strong, all-embracing concept of the state and Finally, to motivate teachers, experience sug- therefore a need for strong cultural consis- gests that no education system achieves high tency across the core public service (OECD teacher quality without aligning professional 2005). rewards, accountability pressures, and finan- Standard competitive examinations and cial rewards in a context-specific manner. For diplomas can promote fairness in the entry example, Finland, Singapore, and Ontario into public service and into service provider (Canada) have all established strong profes- positions such as teachers and health work- sional rewards for teaching, whereas ers. Pre-entry and on-the-job training for dif- Singapore relies on stronger accountability ferent categories of civil servants and service pressures. providers can foster the spirit of public service Similarly for health workers, the World and collective values. And transparency in Health Organization (2006) has called for recruitment and promotion systems can help management of the national health work- reduce patronage. Other measures can be force to move beyond salary and training in taken to limit political capture and improve the public sector to approaches requiring fairness and trust as well as employee motiva- upholding and strengthening the professional tion and performance. One measure is more ethos of health workers, building trust among advanced public employment approaches stakeholders, and linking people’s expecta- such as open and competitive processes for tions with health worker performance. fi lling each position. Another is strong indi- Norms of shared purpose and nonfi nancial vidual performance assessments, relying on incentives have been found to play an impor- job objectives and adherence to certain values tant role in motivating health professionals and norms as defined in a performance agree- (Biller-Andorno and Lee 2013). This entails ment and linked to promotion and advance- acknowledging their professionalism; ment. And yet another is transparent pay addressing professional goals such as recogni- differentials (which can also be a step toward tion, career development, and further qualifi- attracting qualified medical and teaching cation; and encouraging health workers to staff to posts in less desirable areas). meet their personal and organizational goals Teachers and health workers are the larg- as part of their working environment. est highly specialized groups of public ser- Finally, simple measures may go a long va nt s. T hei r ma nagement , i nclud i ng way in internalizing norms of personal attracting, grooming, and motivating them, responsibility and public service. World requires a holistic approach. Development Report 2015: Mind, Society, With respect to teachers, Bruns and Luque and Behavior reports that simply reminding (2015) have highlighted the need to, first, health workers, teachers, public servants, and INCREMENTAL CHANGES IN THE SOCIAL CONTRAC T AND LOCAL EMPOWERMENT 303 policy makers of the social expectations of and scale-up is standard practice (Manzi their performance can improve it (World 2012). One can also discern elements of such Bank 2015). In health facilities and schools, practices in today’s most effective public supportive supervision by peers, professional organizations in both developed and develop- associations, and civil society organizations ing countries (Grindle 1997; Levy 2014; has been shown to contribute to sustaining World Bank forthcoming). Such processes norms of professional behavior, as have social appeared as well in the early years of public cues in the form of recognition such as agencies in now-developed countries, includ- awards, token prizes (such as stars and ing those agencies that initially struggled to plaques to display in the workplace), and obtain political legitimacy and autonomy small gifts (a book or pen). The case studies such as the post office in the United States in chapter 3 suggest that such approaches are (Carpenter 2001). Over time, such agencies indeed accepted positively by teachers and and the services they delivered became seam- health workers, providing them with greater lessly embedded in the fabric of everyday life, satisfaction at work and the motivation to to the point that most citizens of developed meet the expectations of their students and countries took entirely for granted the daily patients as well as supervisors and the com- arrival of the mail, electricity, and clean munity at large. water, as well as education and health care. Such extraordinary service delivery systems Learn from intracountry variation to design were not born large, accountable, and effec- solutions that fit local contexts, to evaluate tive; they became so over time as they incre- and strengthen policy implementation, and mentally acquired the capability to implement to scale up local successes. more complex and contentious tasks, at scale, in a political context that eventually gave Whether at the level of the individual, them the support and legitimacy they needed group, or organization, iterative learning is (Lindert 2004). the pathway to mastering complex tasks. Today’s developing countries, however, When we fi rst learn to speak a language, to and especially those in the MENA region, play a musical instrument, or to ride a bicy- face additional challenges as they seek to cle, for example, we routinely make elemen- deliver quality services. Unlike their counter- tary errors; any objective measure of our parts in the private sector of today’s devel- initial “performance” would deem us fail- oped countries, they cannot presume the ures. But we try again, making more mis- presence of effective legal, financial, and reg- takes, soliciting feedback from and observing ulatory systems that make rapid institutional those who are better than us, and trying innovation and iteration possible. Indeed, again and again. Eventually, we can converse creating and sustaining such systems are in full sentences, play a recognizable tune, themselves a major part of the development and pedal along safely. Acquiring any profes- challenge. Moreover, many public adminis- sional skill, from songwriting to brain sur- trative systems in developing countries are gery, requires embarking on a long quest not being built from scratch; most have a from awkward novice to seasoned practitio- long pre- and postcolonial history, which ner; we learn such complex tasks primarily means that many decades of administrative by doing and mentoring (as opposed to, say, practice have seen the consolidation of all listening to long lectures). manner of internal procedures, incentives, The same basic logic and processes apply and expectations that are not always condu- to groups; they, too, must learn how to mas- cive to high performance. Indeed, there may ter complex tasks, but with the added diffi- be powerful forces committed to blocking the culty of doing so collectively. In businesses necessary reforms. and large organizations today, a process of In such challenging contexts, there is no experimentation, rapid feedback, iteration, universal strategy or toolkit for promoting 304 TRUST, VOICE, AND INCENTIVES change, but that does not mean nothing can methodological approach has its own set of be done. As this report has sought to demon- complementary strengths and weaknesses. strate, there is widespread variation in the Ideally, officials would draw on subnational quality of service delivery within MENA variation data to regularly commission their countries, much of which cannot be accounted own analytic case studies of unlikely successes for by standard economic or geographic fac- in service delivery as part of a broader strat- tors. Data on subnational variation in service egy for enhancing organizational learning. delivery can assist policy makers in identifying possible effective responses to performance challenges. Some of the insights emerging Empowering communities and from a detailed analysis of subnational varia- local leaders to find best-fit tion may be generalizable to crafting national solutions policy reforms—for example, they may reveal The government can create the authorization for human resource management lessons necessary to further empower communities learned about the qualifications, experiences, and local leaders to find best-fit solutions. and temperaments of those who are delivering Despite similarities in the overarching innovative bureaucratic solutions and quality dynamics of service provision, barriers and front-line services for citizens. However, other potential solutions to improving service deliv- insights will not. In these instances, it is the ery are linked to local conditions. Indeed, process by which solutions are discovered even when national-level policies are put in rather than the solutions themselves that place, they are implemented to very different should be the focus of attention. degrees and with diverse effects locally. It is Data on subnational variation can also help critical, then, that local communities be citizens make informed choices and demand authorized by the central government to find accountability. Documenting the actual the appropriate solutions and to implement change already being achieved by local actors experimental programs that may at times be somewhere can be a constructive basis for pro- scaled up to the national level. moting that change more broadly. Further Such an arrangement would require trust diagnostic work based on qualitative analysis and cooperation between the central and of local successes can facilitate their possible local governments and between the state and scale-up and identify bottlenecks to address in civil society. Central governments need to the wider institutional framework for service allow, or even demand, a degree of autonomy delivery. Box 12.1 provides an example of at the local level, and local elites need to trust local institutional reform geared toward trans- that they have the freedom to innovate within parency and the responsiveness of schools. limits. Civil society, citizens, and state actors Meanwhile, low-cost, timely access to also need to develop the mutual trust that information on the performance of local allows engagement of all forces. The effective schools and health facilities in nearby towns engagement of citizens, communities, and and elsewhere in the country can empower local leaders to actively seek solutions— citizens to approach local officials and com- including solutions to address the needs of munity leaders and seek action. Such informa- poor, marginalized, and disadvantaged tion can energize local leaders to pursue groups—also requires actionable information improvements and help them make hard and an outreach to those excluded as well as decisions about where and how finite strategic allies. It is not surprising that, resources—in the face of a wide array of vex- according to the evidence, political relations ing problems—should be optimally allocated. among citizens, civil society, and state leader- Ideally, both broad quantitative data and deep ship are the most important domain for qualitative data would be available to local improving performance incentives and leaders and policy makers seeking service accountability (Devarajan, Khemani, and del iver y i mprovements because each Wa lton 2011). From t he com mu n it y INCREMENTAL CHANGES IN THE SOCIAL CONTRAC T AND LOCAL EMPOWERMENT 305 BOX 12.1 How assessing schools and clearly informing the public promote learning in Dubai As Dubai has grown over the last two decades, the There is, however, one caveat: the stakeholders demand for private education has grown with it. who do act are those in a position to take advantage Today, 88 percent of all primary and secondary of the information offered to them. Although the school students attend private schools, most of better schools are improving, the weaker schools are which cater to the various communities in this city getting caught in a low-equilibrium trap from which and emirate in the United Arab Emirates. Fifteen they do not have the means, either material or curricula are available, including those for U.K., technical, to escape. The KHDA, aware of this U.S., Indian, Pakistani, Iranian, French, German, asymmetry, is taking steps to help stakeholders by Filipino, and Japanese students. Increasingly, producing guides for parents explaining what Emirati nationals are also opting for private characterizes good schools, supporting weak schools establishments. to help them improve, and holding events called The surge in demand led subnational authorities “What Works” to expose teachers and schools to to recognize the need to establish an entity—the best practices. Knowledge and Human Development Authority The KHDA’s initiatives offer lessons for other (K H DA) — to oversee the private sector. T he countries in the region and indeed internationally: immediate challenge for this new public entity was to • Adopt an approach that is entirely transparent, is identify an appropriate approach for regulating a openly accountable, and has strong stakeholder private education sector. participation—all hallmarks of good governance. About that time, the World Bank published The • Disseminate information about all aspects of the Road Not Traveled: Education Reform in the Middle system. School ratings, one of the key elements of East and North Africa (World Bank 2008). The the KHDA approach, not only stir public discus- report argued for better interaction between sion about the importance of school quality but government and constituents based on transparency, also, more significantly, create higher expectations accountability, and participation as a way to foster among education consumers and providers. How- improved national policies, programs, and services in ever, policy makers should not rely exclusively on education. these means alone because very real constraints, The KHDA adopted this approach in Dubai. The technical or material, may prevent certain schools KHDA inspects all schools on a yearly basis and from improving. makes the results of inspections available online, in • Experiment with incentive mechanisms but publications , a nd even via a mobile device remain flexible. The KHDA allows schools to application. It rates schools on eight criteria, resulting increase their fees if they receive better ratings, in an overall rating of unsatisfactory, acceptable, but at the moment this only serves as an incentive good, or outstanding. Ratings are given without to those schools near the cutoff for a higher rat- regard to price. Indeed, the Indian High School, one ing. Other kinds of incentives should be explored. of the least expensive secondary schools in Dubai, is • Leverage competition between schools but not rated “outstanding,” whereas schools far more exclusively. In Dubai, competition has promoted expensive are only rated “acceptable.” some school improvement, but it is not complete: Information on all aspects of the private education schools that have failed to progress over several system in Dubai is now available, and it has sparked years are discouraged. Indeed, it may be that col- publ ic debate i n t he med ia. Moreover, t he laboration rather than competition could support information is being used and so is useful. Parents change just as effectively. In the same way that are asking more questions when choosing schools teachers feel that collaboration with other teach- and interacting with teachers and school heads, who, ers is one of the most effective ways to learn how when given KHDA feedback, are striving to improve to become a more effective teacher, collaboration their teaching practices and school environment with other schools helps them improve as well. (World Bank 2014). 306 TRUST, VOICE, AND INCENTIVES perspective, evidence on citizen engagement Reform champions in government, civil and social accountability efforts points to society, and the private sector can make four important lessons (Mansuri and Rao more headway in improving service delivery 2013; Fox 2014): by building coalitions. Charismatic leaders who engender trust, mobilize engagement, 1. Information must be actionable. Citizens and fi nd allies in government and society can act on information only if an enabling often form such coalitions. As described in environment is in place to reduce the fear earlier chapters, the principal of the Kufor of reprisals. Incentives for information-led Quod Girls’ Secondary School in the West action increase with the likelihood that Bank, the leader of Jordan’s Sakhra com- the state will actually respond to citizens’ prehensive health care clinic, and the gover- voices. If citizens receive no response and nor of Qena, Egypt, established personal observe no improvements in performance, ties with like-minded officials, local civil trust and engagement are likely to decline, society actors, and citizens to mobilize and participation—whether spontaneous human and material resources for better or induced—will not be sustained. On the schools, health care services, and municipal other hand, positive results observable by services. citizens can generate trust and legitimacy Governments should provide political for the state as well as for agencies and space and incentives for such initiatives. concrete reforms. Many such efforts have emerged from civil 2. Only those local governments that are society organizations. Led by energetic pushed by citizens and civil society to be youth, they are aimed at bringing together more accountable are likely to become policy makers, government officials, service more responsive when bolstered by the providers, and citizens. For example, in increased funding and authority that Jordan the group Leaders of Tomorrow has comes with autonomy or decentraliza- established Diwanieh, a project that engages tion. Experience suggests, for example, local and national officials in direct discus- that the localities with more critical local sions with citizens before live audiences media tend to have more responsive local from across the country.4 They address a governments and service delivery. number of critical issues such as the use (or 3. Enabling environments are needed to misuse) of local revenue and inadequate actively encourage the voice and represen- local services. Governments should recog- tation of those who normally would be nize and encourage such efforts because excluded because of gender, ethnic, or they may ultimately increase public trust class bias. and engagement, particularly if coupled 4. Local voices that challenge unaccount- with other initiatives aimed at enhancing able authorities are by themselves likely performance. to be either ignored or squelched. Citizen More formal efforts could be encouraged action that has the backing of govern- as well. Local health committees and educa- ment allies who are both willing and able tion councils that include local officials, ser- to get involved, or that has forged links vice providers, citizens, and key members of with other citizen counterparts to build civil society may operate under a broad countervailing power, has a much greater umbrella of national initiatives (such as chance of addressing impunity. accreditation programs or national school improvement efforts) and could provide a forum for long-term efforts across multiple Build coalitions among champions of service projects. The local progress on national ini- delivery reform in government, civil society, tiatives can be tracked transparently through and the private sector, giving local actors publicly displayed “thermometers,” postings space to develop possible solutions. on Facebook and other social media, and INCREMENTAL CHANGES IN THE SOCIAL CONTRAC T AND LOCAL EMPOWERMENT 307 public announcements in mosques, schools, community leaders and networks can legiti- and other venues. mize programs and mobilize communities Moreover, coalitions can create new oppor- and resources to engage in efforts that pro- tunities for joint reform efforts. For example, mote service delivery. As one customary in Tunisia Al-Bawsala, a civil society organi- leader in Sana’a in the Republic of Yemen put zation initially formed as a watchdog of the it in a focus group discussion for this study, Constituent Assembly, and reform-minded customary authorities can serve as “a con- Assembly members worked together to orga- necting link between the community and the nize meetings with citizens in public forums traditional powers.” across the country. A meeting in Satour found If projects are not embedded in communi- the community with a severe water shortage ties, even the best-intentioned efforts can because of the mismanagement of water funds remain fruitless. In India, for example, the by the local committee. In response, effor t to establish Village E ducation Al-Bawsala worked with citizens and officials Committees (VECs) to monitor the perfor- at the local and governorate levels to identify mance of public schools was launched, but it the problem and find a solution—in this case, was not connected to local champions. As a an elected water management body that incor- result, although most villages had a VEC four porated representatives from all the local clans years after their establishment very few par- (Al-Bawsala 2014). ents knew of it or understood its purpose— Similarly, at the national level coalitions even some parents who were ostensibly engaging citizens can achieve tremendous members (Banerjee et al. 2010). Attempts to success in making otherwise unlikely circumvent local leaders can also generate reforms possible, including the establish- resistance and undermine efforts. Such local ment of high-level political institutions such leaders may see their roles as complementary as a new constitution. Here, Tunisia’s exam- with that of the state, but they also resist ple is highly inspiring. The Tunisian elec- reforms that would leave them sidelined. toral com m ission took adva nt age of Circumventing or ignoring local social relationships with political parties, govern- authorities, networks, and institutions can in ment, and the public to overcome inexperi- fact be counterproductive. This may be par- ence in volatile circumstances and organize ticularly true in postconflict situations, when elections for a National Constit uent such institutions fill important roles left Assembly that would rewrite the Tunisian empty by the withdrawal of state authority. constitution while helping to restore the Where traditional leaders and organizations public’s faith in elections (Tavana 2013). have stepped in to provide services, particu- Efforts to establish mechanisms for engag- larly in the context of the confl icts, refugee ing citizens in policy development, implemen- crises, and weak states found across the tation, and monitoring tend to be more MENA countries today, it is especially effective if building on the existing networks important to build coalitions of reformists of local authorities and social institutions. across state and society that take into account These networks and institutions can provide these local forces. valuable assets for mobilizing resources and legitimizing policies. For example, public Systematically collect feedback on public ser- health clinic directors have found that the vices from users, benchmark service delivery imams of local mosques can legitimize and and local governance performance, and dis- mobilize support for programs—they can seminate information on performance to assure their congregants that family planning provide a rigorous basis for citizen action. is consistent with God’s will, that immuniza- tions are not tainted, and that drugs and obe- To benchmark performance and measure sity defile God’s temple. Speaking from improvements, citizens, civil society organi- positions of social or religious authority, local zations, state officials, and service providers 308 TRUST, VOICE, AND INCENTIVES need information that has been systemati- providing public agencies with systematic cally collected on the quality and adequacy of feedback from the users of public services public services. The citizen report cards (Action Aid n.d.; Hakikazi Catalyst 2004). In (CRCs) implemented in Tunisia, Tanzania, Tunisia, citizens have welcomed the ability to and elsewhere are a simple but powerful tool provide feedback and monitor progress on for use in both promoting transparency and public services delivery (see box 12.2). BOX 12.2 Tunisia’s experience in building citizen feedback loops on service delivery Before the revolution of 2010–11, Tunisia had no and the needs and gaps that need to be addressed performance targets that citizens could use to hold over time. service providers accountable. Since 2011, the gov- ernment and civil society organizations (CSOs) Policy implications for raising accountability embarked on a program, facilitated by the World in Tunisia Bank, to develop qualitative and quantitative tools Since the revolution, Tunisia has for the fi rst time for citizen monitoring of service delivery appropriate adopted policy reforms of accountability in the for the Tunisian context. Participatory monitoring public sector, albeit gradually. For example, guidelines were prepared and used to train all the stakeholders involved. • A new law was adopted in 2011 on access to In the fi rst stage, communities were mobilized to information. participate, and their perceptions of health, • The government implemented in 2012 the first education, employment, and social assistance were participatory “barometer of public services.” identifi ed through qualitative tools, such as local • The government adopted policies on participatory community scorecards. This approach was applied outreach (2011) and accreditation in education through a pilot phase conducted in 6 of Tunisia’s 24 and health (2012). governorates that are home to both urban and rural • A decree in 2013 institutionalized participatory communities. Local CSO and community groups monitoring in one of Tunisia’s highest supreme were brought together, and training was provided audit institutions, the National Controller’s Body by the World Bank. The scorecards helped foster for Public Services. citizen empowerment and identified the key These measures created a framework for improving strengths and weaknesses of local services from the service delivery in areas in which provider behavior com mu n it y ’s p er sp e c t ive. Ac t ion pla n s for and incentives were lacking. And yet to fully sustain improving services were then developed with local and improve accountability and service delivery per- service providers. formance, improving demand-side governance will In the second stage, a quantitative assessment be critical to meeting citizens’ aspirations: using a household survey was conducted based on approaches such as citizen report cards and social • To date, the demand for public administrative audits, adapted to the Tunisian context. Using a information by citizens remains low because participatory approach, the questionnaire was citizens do not believe in their ability to act upon developed with the government, with support from such information toward achieving a change. CSOs, based on areas identified by communities in • A greater effort to proactively encourage citizen the fi rst stage through a qualitative technique. The access to information and participation is needed. quantitative assessment served as a pilot exercise for • Reforms to ensure freedom of information, freedom instituting a mechanism for routinely monitoring key of association, and participatory accountability performance indicators of the quality of service mechanisms for service delivery will help to align delivery. In the future, the tool will help track incentives for greater responsiveness. whether services regularly meet citizens’ expectations Source: World Bank (forthcoming). INCREMENTAL CHANGES IN THE SOCIAL CONTRAC T AND LOCAL EMPOWERMENT 309 More complex instruments, similar to the who share a common cause, build common Public Administration Performance Index ground beyond a locality or group, and (PAPI) developed in Vietnam, use citizen sur- launch networks to enable strategic collective veys to benchmark local governance across a action that can influence the power relations range of issues, allowing the relevant parties and incentive structures that determine to consider the relationship between service whether government actors really will delivery outcomes and other governance respond (Gigler and Bailur 2014). issues.5 Such efforts create incentives for gov- Information also provides a rigorous basis ernment agencies to undertake reforms and and a proactive agenda for communities, civil provide the targeted information needed for society organizations, or local governments those reforms. PAPI demonstrated how to use in pursuing a dialogue with service benchmarking can provide incentives. providers to improve the delivery of public A poorly performing province from the 2010 services. Extensive public campaigns using pilot study requested input from the team community meetings, websites, and social and independently developed an action plan media can generate a constant stream of to improve its future performance. More gen- information that invites the public and public erally, provinces included in the pilot study officials to recognize both obstacles and showed improvement when reassessed, sug- successes. gesting that they acted on the information. Similarly, the ongoing University Governance Closing the feedback loop, engage citizens Benchmarking exercise, which includes 100 and partners in civil society organizations universities in the MENA region, collected and the private sector to strengthen policy information useful for improving perfor- development, prioritization (with emphasis mance and measuring progress (World Bank on the most disadvantaged, poor, and vul- and Marseille Centre for Mediterranean nerable), public resource allocation, and Integration 2013). policy implementation. Providing citizens with information on performance and gathering citizen feedback Information must be linked with mecha- can be done cheaply and highly effectively nisms that allow citizens, service providers, with the help of today’s information commu- and officials to share and act on the infor- nication technology. Citizens can receive on mation they receive about performance their mobile phones regular updates about (Hanushek and Raymond 2005). Engaging the performance of the schools and hospitals local residents and civil society is key. Such in their locality and the ranking of schools engagement can bring new solutions to the and hospitals in their country. Government table, mobilize local resources, and help agencies or civil society organizations also ensure that programs and policies have can establish ways in which people can use buy-in from local actors. It can also provide their mobile devices to report provider soft accountability incentives, making pro- absences and the unavailability of services or viders more willing and able to invest in essential medicines. In Peru’s southern prov- offering good services and benchmarking ince of Puno, for example, designated citizen- performance. monitors oversee the quality of maternal and The public can be brought in as partners child health care and use information com- to assess needs, establish priorities, and con- munication technology to report violations of sider solutions. Sector-specific mechanisms users’ rights and engage with service provid- such as health boards and education clusters ers, the regional ombudsman’s office, CARE can seek the participation of clients, provid- Peru, and ForoSalud (a civil society health ers, community stakeholders, and officials. forum) on ways in which to improve mater- Town hall meetings and consultations can be nal and neonatal health services.6 Technology effective at establishing community priorities and information can help citizens find others and also strengthening social cohesion. 310 TRUST, VOICE, AND INCENTIVES Virtual consultations are also possible, such solutions, communities were able to resolve as that used in a “Government Asks” initia- long-standing problems.7 The combination of tive in the Brazilian state of Rio Grande do information and engagement also appeared Sul, where citizens are given an opportunity effective in designing and implementing solu- to give policy input, either choosing between tions in Uganda, where an intriguing study pair-wise policy choices or suggesting solu- found that public meetings in which informa- tions via mobile phone, Internet, or (in poor tion was distributed to citizens improved areas) face-to-face consultations. Such pro- health outcomes significantly (Björkman and cesses can have an enormous impact. The Svensson 2009). “Government Asks” initiative alone yielded Governments can also give third parties more than 1,300 citizen policy proposals, opportunities to monitor performance, with over 120,000 votes cast on prioritiza- engaging in quality assurance for both pub- tion. It led to higher budget allocations for lic and private providers. The MENA region primary health care, family health programs, has recently seen the emergence of civil and regional hospitals. society organizations aimed at monitoring In all cases, transparent and inclusive pro- public performance, from national parlia- cesses of consultation are essential. Inclusion ments to local officials and civil servants. In is particularly important in the current con- Tunisia, Al-Bawsala not only monitors the text in the MENA region, where the stresses Constituent Assembly, providing citizens of Syrian refugees, internally displaced per- with information about representatives’ sons, and confl ict are reshaping communi- attendance and voting, but also recently ties, undermining states, and placing extended this oversight function by launch- extraordinary pressures on local officials and ing two new programs, Marsad Baladia service providers. The influx of Syrian refu- and Marsad Budget, that will monitor gees is exacerbating the problems encoun- municipal activities and their budgets and tered in delivering water, transportation, make them publicly available in an easily solid waste collection, education, health care, digestible form. Globally, the social audi- and other services. In this context, the very tors who check government records against question of inclusionary processes raises on-the-ground project implementation can objections. It is difficult for citizens and local also raise citizens’ awareness of their rights, officials to accept the inclusion of noncitizen shine a spotlight on offi cials’ actions, pro- refugees in governance processes. And yet vide a space for collective interaction excluding these residents, who sometimes between officials and the public, and create outnumber local citizens, exacerbates incentives for better services. Sector-specific tensions. Moreover, it shuts out people who monitoring is possible as well. For example, are equally affected by local problems and in the Philippines the Department of may be able to contribute to solutions. It is Education collaborated with G-Watch and particularly important, although difficult, to civil society groups to establish Textbook ensure that marginalized voices (whether ref- Count, monitoring whether textbooks were ugees, women, the poor, or others) are repre- delivered on time and in the promised quan- sented and that the political capture of such tities and were of good quality (Guerzovich processes is avoided. and Rosenzweig n.d.). Inclusion is also important in efforts to systematically gather representative informa- tion on needs and priorities. A community Delivering quick wins development project in Jordan found Especially in countries emerging from a that when civil society organizations shock such as conflict and transition, policy conducted citizen surveys of local needs, makers, service providers, and civil society presented the results in a public meeting, should seek quick wins, investing in efforts and asked elected representatives to find that are clearly offering improvements in INCREMENTAL CHANGES IN THE SOCIAL CONTRAC T AND LOCAL EMPOWERMENT 311 service delivery, thereby inspiring citizens’ on feedback because failure to do so could trust and rendering the cycle of performance further undermine trust, engagement, and virtuous. ultimately performance. Popularize local successes; hold public awareness campaigns on citizens’ rights, Moving forward service delivery standards, and social norms; In this report, we have recognized that the expand opportunities for citizen engage- MENA region is at a critical juncture, with ment; and demand a response to citizens’ complicated transitions and tragic confl icts feedback. on one side and a tremendous potential based on rich human and natural resources on the States can identify and popularize local other. We have described how the majority of successes by using the media, awards, and the population in the region lacks economic other campaigns to draw attention to best opportunities, faces inequalities, demands practices and to the social norms of social justice, and expresses frustration responsive and clean service delivery and and mistrust. Our analysis suggests that government. Transparency in criteria and jus- with visionary leadership and inclusive insti- tification are critical in these efforts in order tutions, this vicious cycle of poor perfor- to avoid the real or perceived politicization or mance can be countered. A renewed social political capture of such rewards, which ulti- contract can allow citizens to receive better mately could undermine trust. Not only can services, and empower young women and school principals, chief medical officers, and men to realize their aspirations and potential other providers be honored at the national or and build a brighter future for the next district level, but the same practices can be generation. applied within facilities. Properly imple- We argue that because of the complex cir- mented, such efforts not only honor those cumstances found in MENA countries, it is who model best practices, but also provide an necessary to build on evidence of local suc- opportunity to involve them in teaching and cesses and positive trends at the level of insti- advising both peers and superiors. tutions, performance, and citizens’ trust and Policy makers and communities can also engagement. We hope that this report and its participate in public meetings aimed at needs recommendations will help citizens, public assessments, planning, and the implementa- servants, policy makers, and donors alike tion of projects directed at improving service jointly identify and build on the present foun- delivery on a small scale. Such meetings can dation to improve the delivery of social ser- be solution-focused, such as addressing vices, shifting the cycle of performance into a absenteeism and material shortages in service virtuous gear. An improved cycle of perfor- delivery. mance is what those living in the MENA The key is plugging into the existing countries deserve and what would enable awareness. For example, in Jordan a network them to fulfill their aspirations for the future. of civil society organizations established to conduct legal information and awareness Notes activities has expanded to include public awareness campaigns on drug abuse and 1. The term enabling environment is rarely defined with precision. Fox (2014), for exam- other issues. Such interventions can be ple, refers to actions by external allies that launched relatively quickly and provide citi- have two characteristics. First, they reduce zens with actionable information on their the actual and perceived risks and costs often rights and entitlements and on the existing inherent in collective action. Second, they support mechanisms toward their realization. bolster the actual and perceived efficacy of Interventions should, however, be combined collective action by increasing the likelihood with sufficient efforts and resources to deliver or degree of a positive institutional response. 312 TRUST, VOICE, AND INCENTIVES 2. Alsam, Ghazia, ombudsman, Khyber Incentives for Nurses in the India Public Pakhtunkhwa, Pakistan, unpublished com- Health Care System.” Journal of the European munication, based on “Aide Memoire (2014) Economic Association 6 (2–3): 487–500. Governance Support Project for Khyber Barma, N. H., E. Huybens, and L. Viñuela. 2014. Pakhtunkhwa, the Federally Administered Institutions Taking Root: Building State Tribal Areas (FATA) and Baluchistan— C apacit y in C halle nging C ontexts. P126425, Midterm review, Oct. 28–Nov. 15, Washington, DC: World Bank. 2013.” Biller-Andorno, N., and T. H. Lee. 2013. “Ethical 3. More information is available at http://www Physician Incentives—From Carrots and Sticks .dtmc.gov.ae and http://www.youtube.com to Shared Purpose.” New England Journal of /user/DTMCentre. Medicine 368 (11): 980–82. 4. See http://leadersot.org/initiatives/diwanieh. Björkman, M., and J. Svensson. 2009. “Power to 5. See http://www.papi.vn for more information. the People: Evidence from a Randomized Field 6. See http://governance.care2share.wikispaces.net Experiment on Community-Based Monitoring /Social+accountability+in+Latin+America and in Uganda.” Quarterly Journal of Economics http://www.ciudadanosaldia.org/que-hacemos 124 (2): 735–69. /realizamos-mediciones-y-analisis/rankincad Bruns, B., and J. Luque. 2015. “Great Teachers: .html. How to Raise Student Learning in Latin 7. Interview with U.S. Agency for Development America and the Caribbean.” World Bank, staff, Jordan. Washington, DC. Carpenter, D. 2001. The Forging of Bureaucratic Autonomy: Reputations, Networks, and References Policy Innovation in Executive Agencies, Ac t ion A id. n.d. “S o cia l Ac cou nt abi l it y 186 2 – 192 8 . P r i nc e ton , N J: P r i nc e ton Stocktaking Exercise for Africa: Poverty University Press. Reduction Strategy Monitoring in Tanzania.” Devarajan, S., S. Khemani, and M. Walton. 2011. http://iniciativatpa.org /2012/wp-content “ C i v i l S o c i e t y, P u b l i c A c t i o n a n d / u p l o a d s / 2 0 14 / 1 0 / Ta n z a n i a - P o v e r t y Accountability in Africa.” HKS Faculty -reduction-strategy-monitoring.pdf. Research Working Paper Series RWP11-036, Al-Bawsala. 2014. “Associations for Water John F. Kennedy School of Government, Management: The Case Study of Satour, the Harvard University. Governorate of Kasserine.” Unpublished man- Effective Institutions Platform. 2014. “Supreme uscript, Tunis, Tunisia. Audit Institutions and Stakeholder Engagement Al-Yahya, K. O. 2009. “Power-Influence in Practices: A Stocktaking Report.” http:// Decision Making, Competence Utilization, and iniciativatpa.org/2012/wp-content/uploads Organizational Culture in Public Organizations: /2014/09/Stock-take-report-on-SAIs-and-citizen The Arab World in Comparative Perspective.” -engagement2.pdf. Journal of Public Administration Research and Ferrali, R. 2013. “Redesigning a Public Agency: Theory 19 (2): 385–407. Social Security in Morocco, 2001–2010.” Andrews, M. 2013. The Limits of Institutional Innovations for Successful Societies, Princeton Reform in Development: Changing Rules for University. http://successfulsocieties.princeton R e ali stic S olution s . Ca mbridge, U. K.: .edu/. Cambridge University Press. Fox, J. 2014. “Social Accountability: What Does Andrews, M., L. Pritchett, and M. Woolcock. the Evidence Really Say?” Draft paper for 2013. “Escaping Capability Traps through Global Partnership for Social Accountability Problem Driven Iterative Adaptation (PDIA).” (GPSA), World Bank, Washington, DC. World Development 51: 234–44. Fritz, V., B. Levy, and R. Ort, eds. 2014. Problem- Banerjee, A., R. Banerji, E. Duflo, R. Glennerster, Driven Political Economy Analysis: The and S. Khemani. 2010. “Pitfalls of Participatory World Bank’s Experience. Washington, DC: Programs: Evidence from a Randomized World Bank. Evaluation in Education in India.” American Gigler, B.-S., and S. Bailur. 2014. “Closing the Economic Journal: Economic Policy 2: 1–30. Feedback Loop: Can Technology Bridge the Banerjee, A. V., R. Glennerster, and E. Duflo. A c c o u n t a b i l i t y G a p ? ” Wo r l d B a n k , 2008. “Putting a Band-Aid on a Corpse: Washington, DC. INCREMENTAL CHANGES IN THE SOCIAL CONTRAC T AND LOCAL EMPOWERMENT 313 Grindle, M. S., ed. 1997. “Getting Good Mansuri, G., and V. Rao. 2013. Localizing Government: Capacity Building in the Public Development: Does Participation Work? Sectors of Developing Countries.” Harvard Washington, DC: World Bank. St ud ies in I nternational Development , Manzi, J. 2012. Uncontrolled: The Surprising Cambridge, MA. Payoff of Trial and Error for Business, Politics, Guerzovich, F., and S. Rosenzweig. n.d. “Strategic and Society. New York: Basic Books. Dilemmas in Changing Contexts: G-Watch’s OECD (Organisation for Economic Co-operation Experience in the Philippine Education Sector.” and Development). 20 05. Mode r nising http://www.transparency - initiative.org/wp Government: The Way Forward. Paris: OECD. - content/uploads/2013/10/Strategic-Dilemmas Schick, A. 2011. “Repairing the Budget Contract -Context-FG-SR.pdf. between Citizens and the State.” OECD H a k i ka zi Cat a lyst. 20 0 4. “Pa r t icipator y Journal on Budgeting 11 (3). http://dx.doi Mon itoring of PR S and Pro - Poor .org/10.1787/budget-11-5kg3pdgctc8v. Expenditure in Selected Districts and Areas Sholkamy, H. 2014. “Feminist Social Protection: A of A rusha Region, Tanzania.” A rusha, Possible Lens for Poverty Alleviation and Social Tanzania. http://www.hakikazi.org/papers Justice.” Unpublished ma nuscript , T he / H ak ikazi - Pa r ticipator y-Mon itoring- of American University of Cairo. -PRS-2004.pdf. Tavana, D. 2013. “Preparing to Draft a New Hanushek, E. A., and M. E. Raymond. 2005. Social Contract: Tunisia’s National Constituent “Does School Accountability Lead to Improved Assembly Election, 2011.” Innovations for Student Performance?” Journal of Policy Successful Societies, Princeton University. Analysis and Management 24 (2): 297–327. http://successfulsocieties.princeton.edu/. Ibish, H. 2014. “Hisham, Hope and Despair.” World Bank. 2008. The Road Not Traveled: NOW, October 9. https://now.mmedia.me/lb Education Reform in the Middle East and /en /com mentar yanalysis /564209 North Africa. Washington, DC: World Bank. -hisham-hope-and-despair. — ——. 2014. The Road Traveled: Dubai’s Journey Iyer, D. 2011. “Creating a ‘Citizen Friendly’ towards Improving Private Educ ation. Department: Speeding Document Production Washington, DC: World Bank. i n Jorda n , 1991–96.” I n novat ions for ———. 2015. World Development Report 2015: Successful Societies, Princeton University. Mind, Society, and Behavior. Washington, http://successfulsocieties.princeton.edu/. DC: World Bank. Jones, C. 2013. “Bedouins into Bourgeois? Social ———. Forthcoming. “Opening the Black Box: Engineering for a Market Economy in the Contextual Drivers of Social Accountability UAE.” PhD diss., Yale University. Effectiveness.” Washington, DC. L ev y, B . 2014. Working with the G rain: World Bank and Marseille Centre for Mediterranean Integrating Governance and Growth in Integration. 2013. “Benchmarking Governance Development Strategies. New York: Oxford as a Tool for Promoting Change: 100 Universities University Press. in MENA Paving the Way.” http://cmimarseille Lindert, P. 2004. Growing Public: Social .org/highereducation/. Spending and Economic Growth since the World Health Organization. 2006. The World Eighteenth Century. New York: Cambridge Health Report, 2006: Working Together for University Press. Health. Geneva: WHO. ECO-AUDIT Environmental Benefits Statement The World Bank is committed to preserving Saved: endangered forests and natural resources. The • 3 trees Office of the Publisher has chosen to print • 2 million BTUs of Trust, Voice, and Incentives: Learning from total energy Local Success Stories in Service Delivery in • 336 pounds of net the Middle East and North Africa on recycled greenhouse gases paper with 10 percent postconsumer fiber in • 1,484 gallons of accordance with the recommended standards waste water for paper usage set by the Green Press Initia- • 118 pounds of tive, a nonprofit program supporting publishers solid waste in using fiber that is not sourced from endan- gered forests. For more information, visit www .greenpressinitiative.org.